by Katie Hafner & John Markoff    

(PDF Version)


posed a sinister, if also somewhat vague, threat. This is our attempt to
explain who they are and what drives them.
   This book tells three stories. Kevin Mitnick fit the public's perception
of an archetypal "dark-side" computer hacker. He was thought to be able
to manipulate credit ratings, tap telephones and take complete control
of distant computers. He saw himself as a brilliant computer renegade,
and he proved to be a formidable adversary for one of the world's leading
computer makers. In the end he was trapped by his own arrogance and
   The computer culture of the 1980s was as global as the youth culture
of the 1960s. A young West Berliner who called himself Pengo discov-
ered computers in his early teens. Because his parents had no under-
standing of the electronic world he had entered, no one knew that he
was doing anything wrong when he spent hours at a time in front of a
computer screen. To play out his outlaw fantasies, Pengo joined a group
that sold the fruits of its wanderings through international computer
networks to the Soviets. Eventually Pengo and his gang were torn apart
in a series of betrayals.
   Kevin and Pengo represent something close to the cyberpunk idea of
the computer "cowboy" who lives outside the law. Robert Tappan Morris
was different. The young Cornell University graduate student became
notorious when he wrote a program that brought down a nationwide
computer network. The son of a leading computer security researcher,
he grew up as the ultimate insider, a member of an insular and elite
community of computer scientists. Shy and thoughtful, Robert was
hardly a rebel. By releasing a program that crippled several thousand
computers in a matter of hours, he permanently altered the course of his
life and confirmed everyone's worst fears about what hackers could do.
The event marked a turning point: the private world of computer net-
works was suddenly of concern to the general public.
    More than stories about computers or technology, this book is about
the social consequences of computer networks and the communities that
have grown up around them. As the world's computer networks became
more closely linked in the 1980s, it was suddenly possible for anyone to
travel the electronic corridors that were once the preserve of a small
group of researchers. All three young men were seduced by the thrill of
exploring these international computer networks, but they all went too
far. Each drew national attention and contributed to a growing sense of
public unease about the risks that arise from society's increasing depen-
dence on computer networks.
I --

      In the 1960s and 1970s, to be a computer hacker was to wear a badge
   of honor. It singled one out as an intellectually restless soul compelled
   to stay awake for forty hours at a stretch in order to refine a program
   until it could be refined no more. It signified a dedication to computers
   that was construed as fanatical by outsiders but was a matter of course to
   the hackers themselves. The hackers from the Massachusetts Institute of
   Technology in particular adhered to what has been called the Hacker
   Ethic, which Steven Levy described in his 1984 book Hackers as a code
   of conduct that championed the free sharing of information and de-
   manded that hackers never harm the data they found. Hacking also
   meant anything either particularly clever or particularly wacky, with or
   without a computer, as long as the manipulation of a complex system
   was involved. Some hacks are legendary. There was the computer pro-
   gram that determined the minimum amount of time it took to cover the
   entire New York City subway system. After the program was written, a
   bunch of MIT hackers actually went to New York and tried it out. And
   there was the legendary hoax during the 1961 Rose Bowl game between
   Washington and Minnesota, when Cal Tech students made substitutions
   for the letters on the cards to be held up by the Washington Huskies
   fans at halftime. Instead of Washington, the cards spelled out Cal Tech.
   A palindromic music composition was considered a good hack (thus
   making Haydn, with his Palindrome Symphony, an honorary hacker).
   So was anything done to establish not merely a new record but a new
   category altogether in the Guinness Book of World Records.
      In the 1980s, a new generation appropriated the word "hacker" and,
   with help from the press, used it to define itself as password pirates and
   electronic burglars. With that, the public's perception of hackers
   changed. Hackers were no longer seen as benign explorers but malicious
      These hackers are significant because of what our fear of them says
   about our unease with new technologies. Arthur C. Clarke once said,
   "Any sufficiently advanced technology is indistinguishable from magic."
   For many in this country, hackers have become the new magicians: they
   have mastered the machines that control modern life. This is a time of
   transition, a time when young people are comfortable with a new tech-
   nology that intimidates their elders. It's not surprising that parents,
   federal investigators, prosecutors and judges often panic when con-
   fronted with something they believe is too complicated to understand.
      The fallout from this fear is already apparent. As we were finishing
   the book, something of a hacker hysteria was sweeping the nation. After

       a two-year combined federal and state investigation, in the spring and
       summer of 1990 more than thirty raids on young computer users took
       place across the country, followed by a second wave of searches and
       arrests a few months later. In an attempt to root out the high-tech tools
       being used by this new breed of "criminal," law-enforcement agents
       confiscated computers, modems, answering machines, telephones, fax
       machines and even stereo equipment. Federal agents have gone after
       computer hackers in 1990 as if they are the next scourge after Commu-
          Do young people who illegally enter computers really represent such a
       menace? We hope that from reading the following stories readers will
       learn that the answer isn't a simple one. All three of the young men we
       write about were caught up in what society views as criminal activities,
       yet none saw himself as a criminal. Each felt he was an explorer in a
       remarkable electronic world where the rules aren't clear. And each paid
       a price for his actions.
          It is possible that once computer networks become as commonplace
       as our national highway system, we will learn to treat them in much the
       same way. Rules of the road will emerge and people will learn to respect
       them for their own safety and for the common good. We hope that the
       stories of Kevin Mitnick, Pengo and Robert Morris illustrate not just the
       risks of computer networks but also their allure.

    PART   ONE
     I  t was partnership, if not exactly friendship, that kept the group
together. Each member possessed a special strength considered essential
for what needed to be done. Roscoe was the best computer programmer
and a natural leader. Susan Thunder prided herself on her knowledge of
military computers and a remarkable ability to manipulate people, espe-
cially men. Steven Rhoades was especially good with telephone equip-
ment. And aside from his sheer persistence, Kevin Mitnick had an
extraordinary talent for talking his way into anything. For a while,
during its early days in 1980, the group was untouchable.
   Susan was infatuated with Roscoe, but she never cared much for his
constant companion, Kevin Mitnick. For his part, Kevin barely gave
Susan the time of day. They learned to tolerate one another because of
Roscoe. But for all their mutual hostility, Susan and Kevin shared a
fascination with telephones and the telephone network; it was a fasci-
nation that came to dominate their lives. Susan, Kevin, Roscoe and
Steven were "phone phreaks." By their own definition, phreaks were
telephone hobbyists more expert at understanding the workings of the
Bell System than most Bell employees.
   The illegality of exploring the nooks and crannies of the phone system
added a sense of adventure to phreaking. But the mechanical cornpo-

16   •   Cyg£~PLlNK

nents of telephone networks were rapidly being replaced by computers
that switched calls electronically, opening a new and far more captivat-
ing world for the telephone underground. By 1980, the members of this
high-tech Los Angeles gang weren't just phone phreaks who talked to
each other on party lines and made free telephone calls. Kevin and
Roscoe, in particular, were taking phone phreaking into the growing
realm of computers. By the time they had learned how to manipulate
the very computers that controlled the phone system, they were calling
themselves computer hackers.
   Kevin was the only one of the original group to go even deeper, to
take an adolescent diversion to the point of obsession. Susan, Roscoe
and Steve liked the control and the thrill, and they enjoyed seeing their
pranks replayed for them in the newspapers. But almost a decade later it
would be Kevin, the one who hid from publicity, who would come to
personify the public's nightmare vision of the malevolent computer

                                .. "
Born in Altona, Illinois, in 1959, Susan was still an infant when her
parents, struggling with an unhappy marriage, moved to Tujunga, Cali-
fornia, northeast of the San Fernando Valley. Even after the move to
paradise, with the implicit promise of a chance to start afresh, Susan's
family continued to unravel. Susan was a gawky, buck-toothed little girl.
Rejected and abused, at age eight she found solace in the telephone, a
place where perfect strangers seemed happy to offer a kind word or two.
She made friends with operators, and began calling random numbers in
the telephone book, striking up a conversation with whomever she hap-
pened to catch. Sometimes she called radio disc jockeys.
    After her parents divorced, Susan dropped out of the eighth grade,
ran away to the streets of Hollywood and adopted the name Susy Thun-
der. Susan didn't make many friends, but she did know how to feed
herself. Before long, she was walking Sunset Boulevard, looking for men
in cars who would pay her for sex. She cut a conspicuous figure next to
some of the more diminutive women on the street. Barely out of puberty,
Susan was already approaching six feet.
    When she wasn't walking the streets, she was living in a hazy, drug-
filtered world as a hanger-on in the L.A. music scene, a rock-star grou-
pie. Susan was a bruised child developing into a bruised adult. Quaalude
was her medium of choice for spiriting her away from reality, and when
Quaalude was scarce, she switched to alcohol and heroin. Her mother
finally put her into a nine-month rehabilitation program; she was
abruptly thrown out midcourse. Conflicting stories of Susan's ouster were
in keeping with the blurry line between fact and myth that described her
life. As Susan was to tell it, the adulation of power she developed as a
groupie compelled her to single out the most powerful male staff member
at the treatment center and seduce him. Another story, circulated by
Susan's detractors in the phone phreaking gang, is that she was discov-
ered in the men's bathroom on her knees, servicing another patient.
    Susan found an apartment in Van Nuys and retreated once again to
the telephone, taking comfort in knowing that with the telephone she
could gain access to a world of her own conjuring and shut it out when-
ever she chose. She began calling the telephone conference lines that
were springing up all over Los Angeles in the late 1970s. By dialing a
conference-line number, Susan could connect herself to what sounded
like cross talk, except that she was heard by the others and could join in
the conversation. Some conference-line callers were teenagers who
dialed up after school; others were housewives who stayed on all day,
tuning in and out between household chores but never actually hanging
up the phone. By nightfall, many of the conference lines turned into
telephonic sex parlors, the talk switching from undirected chitchat to
explicit propositions.
    One day in early 1980 Susan discovered HOBO-UFO, one of the first
"legitimate" conference lines in Los Angeles in that its owners used their
own conferencing equipment instead of piggybacking on the phone com-
pany's facilities. Drawing hundreds of people every day, HOBO-UFO
was run from the Hollywood apartment of a young college student who
called himself Roscoe. A friend of Roscoe's named Barney financed the
setup, putting up the money for the multiple phone lines and other
equipment while Roscoe provided the technical wherewithal. Susan de-
cided she couldn't rest until she had met Roscoe, the power behind it
all. But to achieve that goal, Susan knew she would have to abandon
her disembodied telephone persona. She liked describing herself to men
over the telephone. She knew from experience that all she had to do
was mention that she was a six-foot-two blond and she wouldn't have to
wait long for a knock at the door. She was right. No sooner did she
deliver the description than Roscoe came calling.
    The woman who greeted Roscoe was exactly as she had described
herself. Susan had dressed up and made her face up carefully for the big
date. But she could not conceal certain physical oddities. Her long face
displayed a set of teeth so protrusive as to produce a slight speech imped-
18 ...   CYf;fRPLlNK
iment. And there was something incongruous about her large frame: her
upper torso was narrow and delicate, but it descended to a disproportion-
ate outcropping of hips and heavy thighs. Roscoe, for his part, was thin
and pale. His brown-framed glasses met Susan's chin. But if either Susan
or Roscoe was disappointed in the other's looks, neither showed it. They
went to dinner, and when Roscoe asked Susan about her line of work
she told him she did psychological counseling and quickly changed the
   A business student at the University of Southern California, Roscoe
was one of the best-known phone phreaks around Los Angeles. When a
reporter from a local newspaper began researching a story about confer-
ence lines, he told a few HOBO-UFO regulars that he wanted to meet
Roscoe. The next day a caller greeted him by reeling off the billing name
on his unlisted phone number, his home address, the year and make of
his car, and his driver's license number. Then the caller announced
himself: "This is Roscoe."
   When Susan and Roscoe met in 1980, phone phreaking was by no
means a new phenomenon. Phone phreaks had been cheating the Amer-
ican Telephone and Telegraph Company for years. They started out with
"blue boxes" as their primary tool. Named for the color of the original
device, blue boxes were rectangular gadgets that came in a variety of
sizes. Sometimes they were built by electronic hobbyists, at other times
by underground entrepreneurs. Occasionally they were even used by the
Mafia. One of Silicon Valley's legendary companies even has its roots
in blue box manufacturing. Stephen Wozniak and Steven Jobs, who
co-founded Apple Computer in 1976, got their start in the consumer
electronics business several years earlier, peddling blue boxes in college
    A blue box was universally useful because it could exploit a quirk in
the design of the nation's long-distance telephone system. The device
emitted a high-pitched squeal, the 2600-hertz tone that, in the heyday
of the blue boxers, controlled the AT&T long-distance switching sys-
tem. When phone company equipment detected the tone, it readied
itself for a new call. A series of special tones from the box allowed the
blue box user to dial anywhere in the world. Using these clever devices,
phone phreaks navigated through the Bell System from the palms of
their hands. Tales abounded of blue boxers who routed calls to nearby
pay phones through the long-distance lines of as many as fifteen coun-
tries, just for the satisfaction of hearing the long series of clicks and
kerchunks made by numerous phone companies releasing their circuits.

    Blue boxes were soon joined by succeeding generations of boxes in all
    colors, each serving a separate function, but all designed to skirt the
    computerized record-keeping and switching equipment that the phone
    company uses for billing calls.
       The phone phreaking movement reached its zenith in the early 1970s,
    One folk hero among phreaks was John Draper, whose alias, "Captain
    Crunch," derived from a happy coincidence: he discovered that the toy
    whistle buried in the Cap'n Crunch cereal box matched the phone
    company's 2600-hertz tone perfectly,
       Tending to be as socially maladroit as they were technically proficient,
    phone phreaks were a bizarre group, driven by a compulsive need to
    learn all they could about the object of their obsession, One famous
    blind phreak named Joe Engressia discovered the telephone as a small
    child; at age eight he could whistle in perfect pitch, easily imitating the
    2600-hertz AT&T signal. Joe's lips were his blue box, After graduating
    from college, in tireless pursuit of knowledge about the phone company,
    Joe crisscrossed the country by bus, visiting local phone company offices
    for guided tours. As he was escorted around, he would touch the equip-
    ment and learn new aspects of the phone system, Joe's ambition was not
    to steal revenue from the telephone company but to get a job there. But
    he had made a name for himself as a phreak, and despite his vast store
    of knowledge, the phone company could not be moved to hire him.
    Eventually, Mountain Bell in Denver did give him a job as a trouble-
    shooter in its network service center and his whistling stopped. All that
    he had wanted was to be part of the system.
       The Bell System needed people like Joe on its side. By the mid-1970s,
    AT& T estimated it was losing $30 million a year to telephone fraud. A
    good percentage of the illegal calls, it turned out, were being placed by
    professional white-collar criminals, and even by small businesses trying
    to cut their long-distance phone bills. But unable to redesign its entire
    signaling scheme overnight, AT&T decided to ferret out the bandits.
    Using monitoring equipment in various fraud "hot spots" throughout the
    telephone network, AT&T spent years scanning tens of millions of toll
    calls. By the early 1980s automated scanning had become routine and
    Bell Laboratories, AT&T's research arm, had devised computer pro-
    grams that could detect and locate blue box calls. Relying on increas-
    ingly sophisticated scanning equipment, detection programs embedded
    in its electronic switches and a growing network of informants, AT&T
    caught hundreds of blue boxers.
       In 1971, phone phreaking ventured briefly into the sphere of politics.

       20   •   Cyg£RP~NK

       The activist Abbie Hoffman, joined by a phone phreak who called him-
       self Al Bell, started a newsletter called Youth International Party Line-or
       YIPL for short. With its office at the Yippie headquarters on Bleecker
       Street in New York City's Greenwich Village, YIPL was meant to be the
       technical offshoot of the Yippies. Hoffman's theory was that communi-
       cations were the nerve center of any revolution; liberating communica-
       tions would be the most important phase of a mass revolt. But Al Bell's
       outlook was at odds with Hoffman's; Al saw no place for politics in what
       was essentially a technical journal. In 1973, Al abandoned YIPL and
       Hoffman and moved uptown to set up shop as TAP, the Technological
       Assistance Program.
          Much of the information contained in TAP was culled from AT&T's
       various in-house technical journals. It was information that AT&T
       would rather have kept to itself. And that was the point. Whereas the
       original phreaks like Captain Crunch got their kicks making free phone
       calls, TAP's leaders, while steering clear of a hard political line, believed
       that the newsletter's mission was to disseminate as much information
       about Ma Bell as it could. By 1975, more than thirteen hundred people
       around the world subscribed to the four-page leaflet. For the most part,
       they were loners by their own admission, steeped in private technical
       worlds. TAP was their ultimate handbook. Written in relentlessly tech-
       nical language, TAP contained tips on such topics as lock picking, the
       manipulation of vending machines, do-it-yourself pay phone slugs and
       free electricity. TAP routinely published obscure telephone numbers;
       those of the White House and Buckingham Palace were especially pop-
       ular. And in 1979, during the hostage crisis in Iran, TAP published the
       phone number of the American embassy in Tehran. Every Friday eve-
       ning, a dozen or so TAP people held a meeting at a Manhattan restau-
       rant, many still cloaked in the ties and jackets that betrayed daytime
       lives spent toiling away at white-collar jobs. After work, and inside the
       pages of TAP, they adopted such names as The Professor, The Wizard
       and Dr. Atomic.
          In the late 1970s, a phone phreak who called himself Tom Edison
       took over TAP, bringing in another telephone network enthusiast,
       Cheshire Catalyst, a self-styled "techie-loner-weirdo science fiction fa-
       natic," as one of TAP's primary contributors. Tall and dark with the
       concave, hollow-cheeked look of someone rarely exposed to sunlight,
       Cheshire had been phreaking since the sixties. He discovered the tele-
       phone at age twelve, and learned to clip the speaker leads of the family

    stereo onto a telephone plug so that he could put the handset to his ear
    and listen to the radio while doing his homework. If his mother entered
    the room, he just had to hang up the receiver. By the time he was
    nineteen, Cheshire had become a telex maven, having programmed his
    home computer to simulate a telex machine. Before long he was sending
    telex pen-pal messages around the world. In his twenties, already a
    veteran TAP reader, Cheshire moved to Manhattan, got a job at a bank
    in computer support and joined the TAP inner circle.
       TAP wasn't exactly a movement. It was an attitude, perhaps best
    described as playful contempt for the Bell System. One elderly woman
    from the Midwest sent her subscription check along with a letter to Tom
    Edison saying that although she would never do any of the things de-
    scribed in TAP, she wanted to support those people who were getting
    back at the phone company.
       As one responsible for keeping TAP unassailable, Cheshire didn't
    sully his hands with blue boxes, and he paid his telephone bills scrupu-
    lously. Out of corporate garb, he and his friends stayed busy irking Ma
    Bell through their constant wanderings inside the phone system. As
    Cheshire and his friends would explain to outsiders, they loved the
    telephone network-it was the bureaucracy behind it they hated.
       People like Cheshire lived not so much to defraud corporate behe-
    moths as to home in on their most vulnerable flaws and take full playful
    advantage of them. Beating the system was a way of life. Flying from
    New York to St. Louis, for instance, was not a simple matter of seeking
    an inexpensive fare; it meant hours of research to find the cheapest
    route, even if it meant taking advantage of a special promotional flight
    from N ew York to Los Angeles and disembarking when the flight made
    a stop in St. Louis. And getting the best of AT&T, the most blindly
    bureaucratic monopoly of all, embodied a strike against everything worth
    detesting in a large corporation.
       As private computer networks proliferated in the late 1970s, there
    came a generation of increasingly computer literate young phreaks like
    Roscoe and Kevin Mitnick. If the global telephone network could hold
    a phone phreak entranced, imagine the fascination presented by the
    proliferating networks that began to link the computers of the largest
    corporations. Using a modem, a device that converts a computer's digital
    data into audible tones that can be transmitted over phone lines, any
    clever interloper could hook into a computer network. The first require-
     ment was a valid user identification-the name of an authorized user of
22   •   CYf;ERPl1NK
the network. The next step was to produce a corresponding password.
And in the early days, anyone could root out one valid password or
   The rising computer consciousness of the phone phreaks was inevita-
ble as technology advanced. Electromechanical telephone switches were
rapidly giving way to computerized equivalents all over the world, sud-
denly transforming the ground rules for riding the telephone networks.
The arrival of computerized telephone switches dramatically increased
the risks and the sense of danger, as well as the potential payoff. The
phone company's automatic surveillance powers grew by orders of mag-
nitude, served by silent digital sentinels that sensed the telltale signals
of the electronic phone phreaks' microelectronic armory. As the peril
grew, so did the sense of adventure. Whenever an outsider gained con-
trol of a central office switch or its associated billing and maintenance
computers, by evading often inadequate security barriers, the control
was absolute. Thus the new generation of phone phreaks could go far
beyond placing free telephone calls. Anything was possible: eavesdrop-
ping, altering telephone bills, turning off an unsuspecting victim's phone
service, or even changing the class of service. In one legendary hack a
phone phreak had the computer reclassify someone's home phone as a
pay phone. When the victim picked up the telephone, he was startled
to hear a computerized voice asking him to deposit ten cents. For phone
phreaks, the temptation to step up from the simpler technology of the
telephone to the more complicated and powerful technology of comput-
ers was irresistible.
   In 1983, just as TAP was coming to symbolize the unification of
computers and telephones, the journal came to an untimely end. Tom
Edison's two-story condominium in suburban New Jersey went up in
flames, the object of simultaneous burglary and arson. The burglary was
professional: Tom's computer and disks-all the tools for publishing
TAP-were taken. But the arson job was downright amateurish. Gaso-
line was poured haphazardly and the arsonist failed to open the windows
to feed the fire. For years, Tom and Cheshire speculated that the phone
company had engineered the fire, but proving it was another matter.
Tom had a real-world name, a respectable job and a reputation to main-
tain. Cheshire rented a truck and hauled what remained of the operation
-including hundreds of back issues of TAP, all of which escaped un-
singed-over to his place in Manhattan. But in the end, TAP didn't
survive the blow. A few months after the fire, Cheshire printed its final
   Meanwhile, the Southern California phreaks had been holding their
equivalent of TAP meetings. Once a month or so, a group of phreaks,
including Roscoe and occasionally Kevin, would get together informally
at a Shakey's Pizza Parlor in Hollywood to talk and exchange informa-
tion. But the L.A. phreaks weren't a particularly sociable bunch to begin
with, and their meetings were far less organized than those on the East
Coast, with fewer political overtones.
   Though an avid TAP. reader, Roscoe shunned blue boxes and most of
the other electronic crutches of phone phreaking. They were just too
easy to trace, he thought. Roscoe preferred to exploit flukes, or holes,
that he and his friends found in the newly computerized telephone sys-
tem. Modesty was not one of Roscoe's virtues. He claimed that he had
acquired as much knowledge about the telephone system and the com-
puters that controlled it as anyone else in the country. He kept note-
books filled with the numbers of private lines to corporations like Exxon
and Ralston Purina, along with access codes to scores of computers
operated by everything from the California Department of Motor Vehi-
cles to major airlines. He boasted that he could order prepaid airline
tickets, search car registrations and even get access to the Department
of Motor of Vehicles' computer system to enter or delete police warrants.
Whereas most pure phreaks viewed their art as a clever means of bypass-
ing the phone company, Roscoe saw it as a potential weapon. With
access to phone company computers, he could change numbers, discon-
nect phones or send someone a bill for thousands of dollars. Most of the
numbers in his extensive log came from hours of patient exploration on
a computer terminal at school. And many of the special tricks he learned
from Kevin Mitnick.
   Roscoe met Kevin in 1978, over the amateur radio network. When
Roscoe was tuned in one day, he was startled to hear a nasty fight in
progress between two hams. The control operator of the machine was
accusing a fellow ham named Kevin Mitnick of making illegal long-
distance telephone calls over the radio using stolen MCI codes. At the
time, Roscoe knew nothing of telephones or computers. But given the
vituperative tone of the angry ham, either Kevin Mitnick had done
something truly terrible or he was being unjustly accused. Suspecting the
latter, Roscoe switched on his tape recorder and recorded the invectives
as they flew through the air. Then he got on the radio to tell Kevin that
he had a tape recording of the accusations if Kevin wanted it. Sensing a
potential ally, Kevin gave Roscoe a telephone number to call so that
they could speak privately. As Roscoe was to find out later, Kevin had
24 ..    CYt:£RPliNK
given him the number of a telephone company loop line, a number
hidden in the electronic crevices of the telephone network and reserved
for maintenance workers in the field who are testing circuits. Roscoe was
immediately taken with this teenager, a good three years his junior, who
evidently knew so much about telephones. He drove out to the San
Fernando Valley to meet Kevin, gave him the tape and cemented a new
friendship. Sometimes Kevin called Roscoe directly at his home in Hol-
lywood, a toll call from the San Fernando Valley, and they talked for
hours. When Roscoe asked him how a high schooler could afford it,
Kevin just laughed.
                                  .. T ..

When Susan met Roscoe in 1980, he had been phreaking for about a
year. She fell in love with him almost at once. He was the first man she
had met who displayed some intelligence and whose life didn't revolve
around drugs and the drug scene. She found Roscoe's interest in com-
puters charming, even fascinating. Roscoe was taking phone phreaking
to a new level, combining his knowledge of the phone system with his
growing knowledge of computers. Susan saw this as a brilliant next step
for someone with a phone obsession. What was more, they were both
talented at employing their voices to desired ends. They shared a faith
in how much could be accomplished with a simple phone call. As a
teenager, Susan had employed the technique she called psychological
subversion, otherwise known as social engineering, to talk her way into
backstage passes at dozens of concerts. Posing as a secretary in the office
of the head of the concert production company, she could get her name
added to any guest list. Susan prided herself on those skills. If she and
Roscoe had anything in common, both lacked the mechanism that com-
pels most people to tell the truth.
   Roscoe and Susan started to date each other. Roscoe was attending
the University of Southern California and his schedule there, he told
her, let him see her only on certain nights. But that was fine with Susan,
as she was holding down two jobs. One was as a switchboard operator at
a telephone answering service. The more lucrative job was something
she knew how to make a lot of money at: she worked for a small bordello
in Van Nuys. Her counselor story didn't last long. Roscoe made it his
business to learn all he could about people, and Susan was no exception.
When the truth emerged about Susan's profession, Roscoe found it more
amusing than scandalous.
   In her head, Susan was living out a romance of her own quirky inven-
tion. In fact, the relationship between Susan and Roscoe was oddly
businesslike, hardly distracted by passion. Often their dates consisted of
an excursion to the USC computer center, where Roscoe would set
Susan up with a computer terminal and keep her occupied with computer
games while he "worked." Susan soon realized that he was using accounts
at the university's computer center to log on to different computers
around the country. Susan lost interest in the games and turned her
attention to what Roscoe was doing. 'Before long, she became his prote-
   Susan developed her own talent for finessing her way into forbidden
computer systems. She began to specialize in military computer systems.
The information that resides in the nation's military computers isn't just
any data. It represents the nation's premier power base-the Pentagon.
And in digging for military data, as Susan saw it, she, a high school
dropout and teenage runaway, was just a silo away from the sort of
control that truly mattered. Still, she was a beginner, far short of mas-
tering the Defense Department's complex of computers and communi-
cations networks. What she couldn't supply in technical knowledge she
compensated for with other skills. One of her methods was to go out to
a military base and hang around in the officers' club, or, if she was asked
to leave, in bars near the base. She would get friendly with a high-
ranking officer, then go to bed with him. While he was sleeping, she
would search through his personal effects for computer passwords and
access codes. Roscoe could never have had such access, and Susan de-
rived some measure of satisfaction from knowing this fact. She would
report each new success to Roscoe, who praised her profusely while
making careful note of the specifics.
   From her job at the bordello, Susan was taking home about $1,200 a
week, and all of it came in handy. She invested every spare cent in
computer and phone equipment. She installed a phone line for data
transmission, and an "opinion line" she named "instant relay." Whoever
dialed the "instant relay" number got Susan's commentary on topics of
her own choosing. At the same time, she taught herself to use RSTS,
Resource Sharing Time Sharing, the standard operating system for Dig-
ital Equipment Corporation's PDP-ll minicomputers. (Operating sys-
tems are programs that control a computer's tasks the wayan orchestra
conductor controls musicians. Operating systems start and stop programs
and find and store files.) For computer intruders the fascination with a
26   .&   CYEERPLINK
computer's operating system is obvious: it is not only the master control-
ler but also the computer's gatekeeper, regulating access and limiting the
capabilities of users.
   Roscoe often employed Susan's Van Nuys apartment as a base of
operations. He was usually accompanied by his younger cohort, the
plump and bespectacled Kevin Mitnick. Kevin was the kind of kid who
would be picked last for a school team. His oversize plaid shirts were
seldom tucked in, and his pear-shaped body was so irregular that any
blue jeans would be an imperfect fit. His seventeen years hadn't been
easy. When Kevin was three, his parents separated. His mother, Shelly,
got a job as a waitress at a local delicatessen and embarked on a series of
new relationships. Every time Kevin started to get close to a new father,
the man disappeared. Kevin's real father was seldom in touch; he remar-
ried and had another son, athletic and good-looking. During Kevin's
junior high school years, just as he was getting settled into a new school,
the family moved. It wasn't surprising that Kevin looked to the tele-
phone for solace.
   Susan and Kevin didn't get along from the start. Kevin had no use for
Susan, and Susan saw him as a hulking menace with none of Roscoe's
charm. What was more, he seemed to have a malicious streak that she
didn't see in Roscoe. This curiously oafish friend of Roscoe's always
seemed to be busy carrying out revenge of one sort or another, cutting
off someone's phone service or harassing people over the amateur radio.
At the same time, Kevin was a master of the soothing voice who aimed
at inspiring trust, then cooperation. Kevin used his silken entreaties to
win over even the most skeptical keepers of passwords. And he seemed
to know even more about the phone system than Roscoe. Kevin's most
striking talent was his photographic memory. Presented with a long list
of computer passwords for a minute or two, an hour later Kevin could
recite the list verbatim.
   Roscoe and Kevin prided themselves on their social engineering skills;
they assumed respect would come if they sounded knowledgeable and
authoritative, even in subject areas they knew nothing about. Roscoe or
Kevin would call the telecommunications department of a company and
pose as an angry superior, demanding brusquely to know why a number
for dialing out wasn't working properly. Sufficiently cowed, the recipient
of the call would be more than glad to explain how to use the number in
    While Kevin's approach was more improvisational, Roscoe made
something of a science out of his talent for talking to people. He kept a

      separate notebook in which he listed the names and workplaces of var-
      ious telephone operators and their supervisors. He noted whether they
      were new or experienced, well informed or ignorant, friendly and coop-
      erative or slow and unhelpful. He kept an exhaustive list of personal
      information obtained from hours of chatting: their hobbies, their chil-
      dren's names, their ages and favorite sports and where they had just
         Roscoe and Kevin didn't phreak or break into computers for money.
      Secret information, anything at all that was hidden, was what they
      prized most highly. They seldom if ever tried to sell the information they
      obtained. Yet some of what they had was eminently marketable. Ros-
      coe's notebooks, filled with computer logins and passwords, would have
      fetched a tidy sum from any industrial spy. But phreaking to them was a
      form of high art that money would only cheapen. Roscoe especially
      thrived on the sense of power he derived from his phreaking. Presenting
      a stranger with a litany of personal facts and watching him or her come
      unhinged gave Roscoe his greatest pleasure.
         Another frequent visitor to Susan's apartment was Steve Rhoades, a
      puckish fifteen-year-old from Pasadena with straight brown hair that
      cascaded down to the middle of his back. His timid intelligence and easy
      manner had a way of catching people off guard. He was an expert phreak
      who had earned a grudging respect from the Pacific Bell security force-
      the very people he loved to taunt. So adept was he at manipulating his
      telephone service from the terminal box on the telephone pole outside
      his house that the phone company removed the footholds. Like Kevin
      and Roscoe, Steve was an amateur radio buff. Two-way radios, in fact,
      often figured in their phreaking. When foraging through phone company
      trash for manuals, discarded papers containing passwords and whatever
      else might help them in their phreaking endeavors-a method they
      called trashing-they communicated by two-way radio.
         Whatever the personality differences in the L.A. gang, their various
      talents brought results. Working together, they were able to get the
      numbers of lost or stolen telephone credit cards using highly imaginative
      methods: for instance, they would divert cardholders' toll-free calls to
      report lost cards to a pay phone of their choosing and answer with,
      "Pacific Bell, may I help you?" Their vast store of knowledge came from
      months of diligent research: they obtained manuals however they could
      get them and joined as many company tours as they could, mostly to
      familiarize themselves with building layouts. The group routinely got
      credit information from the computer at TRW's credit bureau. Some-
28   A   CygERPUNK
times it was a matter of talking an unsuspecting employee out of a
password; sometimes a job required "physical research," such as a late-
night excavation of TRW's Dumpsters.
   By brainstorming together, the group would be inspired to ever more
audacious stunts. On one occasion, Steve Rhoades figured out a way to
override directory assistance for Providence, Rhode Island, so that when
people dialed for information, they got one of the gang instead. "Is that
person white or black, sir?" was a favorite line. "You see, we have
separate directories." Or: "Yes, that number is eight-seven-five-zero and
a half. Do you know how to dial the half, ma'am?"
   A few months into her infatuation with Roscoe, Susan noticed un-
happily that he was spending less time with her. Then someone let her
in on the reason: she had been deceived. Roscoe, it turned out, was all
but engaged to someone else, a law student as prim and straight in her
ways as Susan was errant in hers. The other girlfriend apparently pro-
vided Roscoe's tie to respectable society. Susan despaired. He had been
displaying affection for Susan when what he really wanted to do was
exploit her eagerness to hack. When she confronted him, he just
laughed. So she tried a veiled threat: she told him it was quite likely that
FBI agents would come knocking on her door, and she would have to
talk to them. Roscoe feigned puzzlement. Susan refused to believe that
he wasn't at least torn between the two women. Seizing upon that
possibility, she suggested that they go to Las Vegas and get married at
once. He just smiled and said she had been misled. What Roscoe had
failed to take into account was that the person he had just crushed was
a woman who had been bruised once too often. By rejecting her so
heartlessly, Roscoe was inviting trouble. He had yet to experience Su-
san's dark side.

Eddie Rivera, a free-lance writer in Los Angeles, wasn't sure what he
had stumbled into. As he was getting out of his car one day on Sunset
Boulevard in late April of 1980, he saw a flyer that simply read: "UFO
CONFERENCE CALL NOW." His curiosity stirred, the young reporter dialed
HOBO-UFO and heard three people chatting idly, engaged in what
seemed to be an interminable telephone conversation. A television mur-
mured in the background. After listening to five more minutes of prattle,
Eddie sensed the possibility of a story far different from what he usually
produced as a rock-and-roll critic. Like Susan Thunder before him, Eddie
focused on meeting the person running the conference. "Excuse me," he
interrupted. "If you guys on this line know who runs it, have him call
me." The line went silent; he felt as if he had switched on the kitchen
light late at night and seen dozens of cockroaches running for cover. But
it worked. Within a day of putting the word out that he wanted to meet
Roscoe, Eddie received his first call. Eddie got a story assignment from
the L.A. Weekly, an alternative newspaper, and went to work.
    Their first meeting took place at an electronics store on Santa Monica
Boulevard owned by Barney, Roscoe's pudgy, disheveled benefactor.
When not in school or overseeing the HOBO-UFO line from his home,
Roscoe was often at Barney's place. With not a customer in sight, the
store was littered with old televisions in various stages of disrepair. For
his part, Barney derived no small amount of pleasure from the venture
he was financing; he used the conference line to meet adolescent girls.
    Eddie hadn't known quite what to expect. Roscoe's appearance was
surprisingly neat, but there was something amiss: his pale blue polyester
pants with a slight flare at the bottom, and his dark polyester print shirt
with an oversize collar, were already at least five years out of date. The
twenty-year-old Roscoe seemed more to resemble an electrical engineer-
ing student than a telephone outlaw.
    For the first interview, Barney put a "closed" sign in the shop window
and, in what would become a routine preamble to the interviews, the
three went out for doughnuts. Roscoe lived on junk food, as did, it
seemed, all his fellow phreaks. A patina of doughnut glaze frequently
rested on Roscoe's lips. In the afternoons, Roscoe moved on to Doritos
and cheeseburgers. And Barney's store was strewn with Winchell's Do-
nuts coffee cups whose contents suggested that Barney might be using
them as petri dishes.
    Roscoe made it clear from the start that he would be in complete
control of the nature and quantity of information he imparted to Eddie.
He enjoyed telling Eddie just how much he knew about telephones-far
beyond the body of information known to the average telephone com-
pany employee.
    There was something oddly mechanized about Roscoe's language.
Eddie was struck by the young man's formal, almost bureaucratic way of
speaking. In response to a question, Roscoe usually answered as indi-
rectly as possible. He had a curious affection for the passive construction.
Roscoe didn't simply make phone calls. Instead, telephone conversations
were initiated. Perhaps, Eddie thought, Roscoe's tangled locution re-
sulted from reading too many of those phone company manuals he kept
 talking about. In any case, his manner of speech distanced him from
30   ...   CYE£RPLlNI
whatever he was talking about. Perhaps it made him feel more impor-
   Roscoe told Eddie that he had a friend at the phone company who
could get him into the switching room that housed the powerful com-
puter controlling all the telephones in Hollywood. During one recent
late-night visit there, Roscoe told Eddie, he had walked over to a wall
that was blanketed with telephone company switching equipment and
watched as his friend flicked a switch. At the sound of a female voice,
the friend announced proudly, "That's Farrah Fawcett." Bored telephone
company employees, Roscoe's friend claimed, monitored people's calls
all the time. Roscoe told the story in such precise detail that Eddie had
no doubt about its truth.
   In the early reporting stages of Eddie's article, Roscoe was highly
secretive about his whereabouts, guarding his daily comings and goings
like a fugitive. If Eddie wanted to speak with him, he would have to wait
for him to call. After a few weeks, Roscoe gave him a number where he
could leave a message. It was a month before Roscoe invited the reporter
to his home, one of ten units in a plain two-story white stucco building
located in a shabby neighborhood on the southern edge of Hollywood
that was cluttered with similar apartment buildings. Roscoe lived with
his mother in a ground-floor two-bedroom apartment. Adult bookstores
dominated the neighborhood.
   There seemed to be little in Roscoe's life besides school and tele-
phones. Eddie heard that Roscoe had a girlfriend, but he never saw any
sign of her. Roscoe had once introduced Eddie to someone named Susan,
a bizarre and cranky young woman, eccentrically tall and with unusually
wide hips, but she and Roscoe appeared to be just friends. His world was
the telephone, and from his small bedroom in the back he operated his
HOBO-UFO conference. His phone rang constantly as people called the
line. Roscoe continuously monitored the conference through a speaker-
phone, which created a constant low level of conversation in the room,
and he could pick up his telephone any time and interrupt. Another line
attached to an answering machine rang frequently as well. Many of those
calls came from giggling teenage girls to whom Roscoe had given his
private number.
   There was something about the way Roscoe reacted to telephone
tones that made Eddie suspect he had a musical ear. He could recite a
phone number just by hearing it dialed. Spying an upright piano and a
large stereo in the modest quarters, Eddie decided he was right. Scholas-
tic awards from Belmont High School lined Roscoe's bedroom wall; now
he was attending USC on a scholarship. And from what Eddie could see
of Roscoe's interaction with his mother, an immigrant from Argentina
who appeared to speak no English, Roscoe was a model son. Eddie was
taken aback to hear Roscoe's perfect Spanish, spoken with the unself-
conscious ease of a native. Roscoe's complexion was so pale and his
English so flatly American that Eddie would have put his roots far north
of Argentina, possibly in Iowa.
    If Roscoe had a side that was less restrained and formal, he displayed
it to Eddie just once. On a drive through Hollywood, down a section of
Western A venue elevated above the freeway, Eddie and Roscoe were
stopped at a red light a few doors away from a storefront church, where
a small congregation of Hispanics lingered outside. "Slow down!" Roscoe
blurted out suddenly from the passenger's seat. "I'm going to freak these
people out!" He rolled down his window and leaned his torso out the
window. "Ay Dios mio!" he screamed in perfect Spanish, in the evan-
gelical wail of a recent convert. As they accelerated away from the
horrified worshipers, Roscoe was beside himself with laughter. "That
gets 'em every time!" he cried.
    For the most part, the people who called Roscoe's conference lived
for their telephonic encounters. Many were blind, Roscoe told Eddie, or
otherwise handicapped. Others were housewives or single mothers. The
majority were overweight. Their names-Rick the Trip, Regina Watts
Towers, Dan Dual-Phase, Mike Montage-suggested to Eddie a group of
shy folk making sad attempts to add mystery and intrigue to their lives.
Eddie's suspicions were confirmed when, several weeks into his report-
ing, he attended a phreak party at Dan Dual-Phase's house. For many
members of the group, it couldn't be a simple drop-in affair. So few of
the conference line callers could drive that the transportation logistics
alone had lent an unusual air to the party. At the event itself, the
conference line callers sat in scattered clutches of embarrassed silence,
too shy to speak to one another in person.
    As part of the education process, Roscoe presented Eddie with litera-
ture. He gave the reporter several back issues of TAP, referring to it as
if its circulation matched that of Newsweek. After a mystifying perusal
of TAP's pages, with its proclamations that no code can be completely
secure, Eddie could only conclude that it carried the voice of true out-
laws. Roscoe also showed him a legendary Esquire magazine story about
phone phreaks. One of the featured phone phreaks in the story was
Captain Crunch. Roscoe angrily denounced Crunch as an idiot whose
blue box was a crutch. Phone phreaks should be like Houdinis, able to
32   A   CygERPUNK
cruise the telephone network as if by magic, without the visible aid of
tools. Roscoe did have one hardware crutch-a Touch-Tone dialer, a
square, gray plastic box small enough to fit in one's palm. On the front
was a dialing pad with ten numbers; batteries were taped to the back.
The dialer was elegantly simple yet indispensable. Its function was to
send Touch-Tones into the mouthpiece of a rotary-dial phone. These
were precisely the tones that he needed to get access to corporate phone
networks for his free telephone calls. Once he reached a company's
private telephone system, a friendly digitized voice asked for a code.
After receiving a correct code punched into the dialer, the phone system
stood open for dialing anywhere at all.
   One day during Eddie's reporting assignment, Roscoe had with him a
slightly younger friend named Kevin. Eddie had already heard about
Kevin from Roscoe. If Eddie found Roscoe's knowledge of telephones
impressive, Roscoe had told him in a reverential tone, he should meet
Kevin. Kevin lived about forty-five minutes away from Roscoe in the
San Fernando Valley, and until Kevin got his driver's license, Roscoe
regularly retrieved him and then took him home.
   Kevin was overweight and exceedingly shy. Roscoe, in fact, was
downright garrulous compared with his laconic friend, who assiduously
avoided eye contact. Where Eddie had seen glimpses of normalcy in
Roscoe, in Kevin he saw nothing but a life steeped in telephones and
computers. When he joined Roscoe and Kevin for an afternoon of
phreaking, Eddie noticed that Roscoe frequently deferred to Kevin,
whose encyclopedic knowledge of the telephone company's computer-
ized control switches was well beyond his own. And where Roscoe was
clearly taken with the idea of a reporter trailing after him for weeks on
end, Kevin was altogether uninterested in Eddie. In fact, Roscoe too
seemed temporarily to lose interest in the reporter when Kevin was
around, so deep was his concentration on the business of phreaking with
   Eddie was struck by the patience and perseverance the two youths
displayed when seated before a computer screen. Eddie had seen Roscoe
spend an hour at a time simply scanning for dialing codes, but the display
of endurance when Roscoe and Kevin joined forces was in another
league. For five hours on one occasion, they sat in front of a computer .
terminal that was connected to a phone company computer, watching a
series of numbers scroll by. Roscoe and Kevin grew increasingly excited
over the hieroglyphics on the screen, but their excitement passed right
over Eddie. If the reporter asked them to explain what they were seeing,
he received a sidelong glance of amused condescension. If these two
phreaks were breaking the law, Eddie couldn't tell. Before long he
stopped trying to understand what they were doing and found himself
struggling to stay awake.
   Eddie had an inkling that Kevin might be even more important than
Roscoe for his story. Not only did Kevin seem to know more than Roscoe
about telephones, but Eddie got the impression that Kevin had taught
Roscoe much of what he knew. Eddie figured he should probably take
the kid out for a cheeseburger, but he decided he wanted nothing to do
with him. As he felt himself pulled further into this strange world, he
realized that the phreaks were beginning to make him nervous. In real-
ity, Roscoe's life wasn't much richer than the lives of the lonely souls
who hung on his conference line. Yet Roscoe regarded the others with a
mixture of delight and disdain: delight at their obvious respect for him,
and disdain for the emptiness in their lives. But they just depressed
Eddie. Between the desperate conference-line callers and the sinkhole of
technical language, clouded further by Roscoe's impossibly stilted
speech, Eddie was beginning to regret that he had tackled such a difficult
story. He began to yearn for an easy, more familiar assignment, perhaps
a backstage interview with Joey and the Pizzas. At the least, his stint as
an honorary member of the .gang had come to an end. It was time to
write his story.
   Eddie's cover story in the L.A. Weekly in the summer of 1980 left a
lasting impression on those who saw it. Mostly it was a profile of Roscoe,
a phone phreak who could do anything with a telephone. Shortly after
the story appeared, Eddie was at a party and overheard a conversation
about it. When he mentioned that he had written it, people said they
wanted to meet this Roscoe and learn a few of his tricks.

Meanwhile, Susan was obsessed by her desire to get something on Ros-
coe, something she could hold over him and something other people
would believe. It wouldn't be difficult, she figured, since his forays often
led him into dubious territory. And she wouldn't mind getting Kevin as
well. Kevin and Roscoe had a long-standing agreement to share infor-
mation with each other, usually to the exclusion of Susan. No, she
wouldn't mind one bit if Kevin got pulled into the undertow.
   The first opportunity for revenge came with the U.S. Leasing break-
in. With a little programming flourish, Roscoe had given himself system
privileges on a computer at U.S. Leasing, a San Francisco company with
34 •     CYE£RPLINK
subsidiaries specializing in leasing electronic equipment, railcars and
computers. In most large multiuser computers there is a hierarchy of
privilege meted out to users. The system manager has the highest level
of privilege, acting as de facto God in the computer system, while ordi-
nary users have their capabilities more stringently curtailed. Such a
pecking order works only if the lesser users cannot find ways to masquer-
ade as the system manager.
   Because its network address circulated widely throughout the phreak
community in 1980, U.S. Leasing had one of the most popular comput-
ers for phreaks to play on. Both Roscoe and Susan, in fact, enjoyed
posting the computer's network address on electronic bulletin boards,
offering guided tours of the system to phreaking neophytes. U.S. Leasing
used nothing but Digital Equipment PDP-II computers. All the com-
puters ran RSTS, a notoriously insecure operating system. RSTS had
been designed in the 1970s as the ultimate in user-friendliness. Ask the
system for a password and it would assign you one automatically. Ask for
a system status report and you were provided with the name of every user
on the system. More often than not, people chose their names as their
passwords. To make things easier for its customers, Digital even supplied
sample passwords, such as field and test for field technicians. And the
field technicians often had highly privileged accounts.
   Getting to computers such as the one at U. S. Leasing was deliciously
easy. The first step was to dial into the Telenet network. Telenet was
the first commercial network designed solely to link together computers.
Computer networks differ from telephone networks in that instead of
each conversation getting its own circuit, many computer-to-computer
conversations on a single network can share a single circuit. Because the
information being transmitted is digitized, strings of ones and zeros, it
can be broken up into small packets. Each packet contains an address
that tells the network where it's going. This is known as packet-switch-
   T elenet was structured in such a way that you could choose to com-
municate with any computer within the network simply by typing a
sequence of numbers, whereupon you were automatically connected to
the computer you chose. The computers, each with an assigned number
sequence, formed a mesh. Any system connected to Telenet, be it a
computer at Bank of America or one at General Foods, could be reached
with one local phone call.
    Gathering some physical evidence was Susan's first task. She knew
how difficult it was to track down people who broke into computers: the
only fingerprints they left were electronic, and those were nearly impos-
sible to attach beyond a reasonable doubt to an individual. Reams of
printouts logging an intruder's electronic joyride through a computer or
a network of computers were worthless unless there was stronger evi-
dence linking the specific person to the incident. Susan's first step was
to get something in Roscoe's handwriting. When he jotted the number
of the U.S. Leasing computer and several company passwords on a sheet
of paper, Susan watched as he folded it and put it in his back pocket.

The computer staff at U.S. Leasing were baffled when, one day in De-
cember 1980, the computers that ran their business began acting
strangely. The company's computers were behaving in an unusually slug-
gish manner. So it was a relief to the computer operator on duty late one
afternoon when someone called to say that he was a software trouble-
shooter working for Digital Equipment Corporation. The slowdown
problem, he told the operator, was affecting all of Digital's sites. The
situation was so widespread, he said, he wouldn't be able to come in
person to fix it; he would have to walk someone through the procedure
over the phone. The cheerful technician asked for a phone number for
the computers, a login and a password. He said he would then insert a
"fix" into the system. The computer operator at U.S. Leasing was only
too happy to oblige-it was a procedure he had gone through before
with Digital. He thanked the Digital technician, who assured him that
everything would be back to normal in the morning.
   But the next morning, the computers were as phlegmatic as ever. If
anything, the problem was worse. The computer operator called John
Whipple, U.S. Leasing's vice-president for data processing, who called
the local Digital office in San Francisco and asked for the helpful tech-
nician who had called the previous day. There was no employee by that
name in San Francisco. So Whipple called Digital headquarters in Mas-
sachusetts. Not only was there no such employee on record, but Digital
had no plans to make a universal repair. Whipple went straight to the
computer room to find the operator. "Someone has been in," he told
   Whipple's only choice, he decided, was to find and destroy any un-
authorized accounts, to call everyone who used the U.S. Leasing com-
puters and have them change their passwords. Later in the day, the
36   ~   CYGERPUNK
computer operator had a second call from the "technician." He was as
friendly as ever, explaining that the fix hadn't taken. "I can't seem to
get into your machine," he said in a concerned voice.
    This time, the operator played dumb. "Give me your number and I'll
call you back."
    "I'm not really reachable," came the response. "Let me call you back."
    When Whipple got to work the next morning, the computer operator
was beside himself. The printer connected to one of the computers had
been disgorging printouts all night. The floor was papered with them.
And every page of every printout was densely covered with type. Cov-
ering each printout, repeated hundreds of times was a spiteful message:
    Another read: "REVENGE IS OURS!"
    And finally, in neat rows of type, marching across the page: "FUCK
    Interspersed among the vulgarities were names. Roscoe was one. Mit-
nick was another. Mitnick? MITnick? Could that mean that MIT stu-
dents had done it?
    People who were attracted to the computer profession in the 1960s
and 1970s could hardly be considered flamboyant. By and large, they
were technical loners. Those who entered the field before the computer
industry mushroomed, beckoning thousands of ambitious self-starters
with the promise of fortunes to be made, were mostly men like John
Whipple. They went to schools like MIT and the California Institute of
Technology, and they latched onto computers as an extension of an
adolescent compulsion to sit in their rooms and pull radios apart. Others
had a simple fascination with math, and computers were just the logical
next step.
    Whipple had been in the computer industry for twenty years, long
enough to have witnessed many a harmless prank. MIT students were
famous for harmless pranks. But this was no practical joke. These elec-
tronic vandals might as well have broken into the room itself, sprayed
graffiti all over the walls and taken a hatchet to the computers. Not only
had they plastered printer paper with their malicious messages, they had
gone into the files in system B themselves, deleted every scrap of infor-
mation on inventory and customers and billing notices and replaced
everything with their invective. They had, that is, destroyed the com-

     purer's entire data base. Whipple ordered both of U.S. Leasing's com-
     puters to be cut off from any telephone lines and sealed from outsider
     access. His chief worry wasn't so much over the lost data, because he
     also had the information on backup tapes. His main fear was that some-
     how the intruders had devised a program that would let them back into
     U. S. Leasing even after all the passwords had been changed, a software
     contrivance called a trapdoor. That day, the company's computer oper-
     ations remained shut down while the computer staff loaded the backups,
     reconstructed the data and restored it on the machines. The entire
     restoration process took twenty-four hours. After a day, the computers
     went back on line.
        Although all the passwords had been changed, and the system seemed
     as secure as it was going to get, Whipple decided he wasn't going to rest
     until arrests had been made. His first call was to Digital headquarters.
     All he wanted was some idea of how the intruders might have broken in
     at all, and some assurance that it wouldn't happen again. Anyone at
     Digital, he figured, would sympathize immediately with what Whipple
     was going through. At the least he expected a sympathetic ear, and he
     wouldn't have been surprised if the company dispatched some eager
     young security expert to San Francisco to take care of everything. But
     the response he got was tepid if not cool. Within a few minutes, he was
     mired in a bureaucratic procedure. In order to have Digital even consider
     the matter, Whipple would have to fill out a purchase order, then have
     his supervisor authorize the expense. A purchase order? Some invisible
     creeps were erasing his data, writing "FUCK YOU" all over the place,
     threatening his other computer, and Digital wanted him to fill out a
     purchase order?
        This wasn't the reaction Whipple had expected from Digital, of all
     companies. Digital was founded in 1957 by Ken Olsen, an individualistic
     and outspoken MIT engineer. Just as the International Business Ma-
     chines Corporation was passing the billion-dollar mark in sales of ma-
     chines that sat behind glass enclosures and processed information in huge
     batches, Olsen set out with $70,000 in venture capital to build a smaller
     computer that interacted directly with the user. The idea of an interac-
     tive computer had come from a pioneering generation of computer re-
     searchers at MIT, and one of their early machines became the model for
     Digital's first computer, the PDP -1. When Digital delivered one of its
     first computers to MIT, it was installed in a room one floor above an
     IBM machine. The IBM computer was locked behind two layers of glass;
     problems were submitted on batches of IBM cards and results didn't come
38   &   CY;gERPUNK
until the next morning. The Digital computer was accessible to students
any time of the day or night; commands were typed on the keyboard and
the computer responded in a few seconds. A love affair began. Students
gathered around and worked on the new computer until 3:00, 4:00 and
5:00 in the morning. The MIT administration even considered removing
the Digital computer because people were so obsessed with it they
stopped washing, eating and studying for their classes. And from the
beginning, Digital designed its computers to be used in networks.
   Digital found its first large customer base at universities and other
research institutions. For the same reasons of accessibility and speed, it
didn't take long for Digital computers to gain acceptance among com-
mercial customers, too. The complexity that came with buying an IBM
mainframe was often overwhelming for people outside of university com-
puter science departments. Digital computers provided a package that
was as professional as IBM's, but not nearly as complex.
   Whipple had believed Digital to be a company driven by its customers'
requirements. He was surprised and dismayed by the company's response
to his request for help. Evidently, the bureaucratic tangle served to mask
the true situation: the people at Digital headquarters in Massachusetts
wanted nothing to do with the event.
   So Whipple called the FBI. Three agents arrived at his office the next
day. They showed far more concern than anyone at Digital had. But
their questions were geared toward trying to figure out whether this was
a federal case. There was no doubt that a crime had been committed.
There were no federal laws governing computer crime at the time, but if
this case fell under federal jurisdiction, it could be prosecuted under
federal wire fraud statutes. At the same time, the state of California had
a year-old law on the books prohibiting unauthorized access to computer.
systems. After some deliberation, the FBI decided that, since the break-
in didn't appear to have involved interstate telephone calls, it should be
handled by local authorities.
   Whipple's next call was to the telephone company. He asked Pacific
Bell to place traps on the lines. The phone company agreed to do the
traces. But that presumed that whoever had committed this act would
call again. Whipple had the feeling, even the hope, that these intruders
would be arrogant enough to do so. And they were. The call came in
the afternoon. This time, the computer operator's object was to keep
the "technician" on the line while a trace was completed. "I don't know
what's going on here," he moaned into the phone, "but everyone is
upset." The operator kept a dialogue going for as long as he could. He
cursed U.S. Leasing, saying it had taken away his password, so he
couldn't log on to the computers even if he wanted to. Oddly, the
friendly technician seemed in no hurry to hang up. With an eerily
professional knowledge of the system, he told the operator exactly what
to do to bring the computers back on line. Whipple and a phone com-
pany employee were listening on another line, and when the operator
put the hacker on hold, Whipple could hear two people talking in the
background. "I think they're on to us," one said to the other.
   The trace went only as far back as an outbound Sprint port in San
Francisco-the electronic doorway of GTE Sprint, then a fledgling long-
distance telecommunications company. Using several different long-dis-
tance accounts to cover their tracks was one of the phone phreaks'
favorite tricks. Someone wishing to break into a computer could make
himself more difficult to trace by first dialing through the equipment of
one or even several long-distance carriers before finally calling his target.
In every case false credit card numbers were used. Such a ruse made the
phone company's task of completing a trace considerably more difficult.
Crossing company boundaries made it easy for phreaks to hide behind
layers of bureaucracy that slowed law-enforcement officials.
    It took Whipple several phone calls before he finally reached Sprint's
security department, such as it was. The security people gave him a
number he could call around the clock. If the intruder called again, they
told him, he should call the number at once so that they could start
their own trace. A few hours later, the intruder did call. Again the
unhappy operator kept him chatting, and again the phone company
traced the call to the same Sprint telephone number. With the glee of
an angler whose hook is securely lodged in the mouth of a prizewinning
marlin, Whipple called the number the Sprint security officials had given
him. There was no answer.
    Whipple had never been much of a candidate for confrontation. He
was easygoing and low-key. But now he was on the warpath. To arrive
at work one morning and see his printer spewing out obscenities and all
of the information on one of his computers obliterated was something
Whipple had never imagined. He wanted to find out if other Digital
customers in the San Francisco area had been hit. Or did these mis-
creants have a particular reason for wanting to attack U.S. Leasing?
That afternoon, he began calling other large Digital installations. At
least a half-dozen others told him they too had had troublemakers in
their systems. One large hospital that had just bought its Digital system
didn't even have it fully installed, yet had already been getting repeated
40   A   CygERPUNK
phone calls from a "Digital technician" demanding the computer dial-
up number and passwords. When the hospital computer manager ex-
plained to the caller that the computer wasn't even up and running, the
indignant caller demanded to speak with the hospital's chief administra-
tor. Whipple asked the hospital administrator if he would be willing to
join him in pressing charges. No, came the answer. After all, the hos-
pital had been spared any damage, and the negative publicity, he told
Whipple, would far outweigh the satisfaction of seeing justice done.
   Whipple realized that if anyone was going to take up a crusade against
these electronic foes, it would have to be U.S. Leasing alone. He took
the systems down for another day to change everyone's password a sec-
ond time. He set up the computers to monitor all unsuccessful log-in
attempts. The following day, the computers were back on line. By four
o'clock that afternoon the intruders were back, too, trying without suc-
cess to break in again. Like insects smacking themselves senseless against
a screen at night to get to the light inside, they kept flying at the system,
trying password after password. Once again, U.S. Leasing immediately
shut down all its outside lines.
   By this time, the company was buzzing with talk of the intrusions.
Whipple was stunned when he met a senior executive in the hallway
who chuckled and said, "Maybe we should hire this kid." Whoever was
doing this certainly seemed to have more than a passing familiarity with
the computer's operating system. In fact, from the commands he was
typing, he seemed to know more about exploring its nooks and crannies
than Whipple and his staff. But hire such a meanspirited person? That
would be like giving the Boston Strangler a maintenance job in a nurs-
ing-school dormitory. Whipple's reply to the executive was curt: "How
many people will we have to hire to watch him?"
   As incredulous as he was disappointed, Whipple began to resign him-
self to writing this off as a colossal, embarrassing waste of energy and
time. There wasn't much left to do but count up the losses and try to
forget about it. In the end, he figured, the break-in had cost the company
a quarter of a million dollars in lost time and business. The chances of
ever catching the creeps who had caused all the headaches were slim at
best. A few weeks later, Whipple dropped the matter entirely.
   Years later, Roscoe and Kevin would claim they had been framed by
Susan Thunder, that she had plastered their names all over the U.S.
Leasing computer in order to pin the break-in on them, while Susan
would hold firm to her claim that Kevin and Roscoe had done it.
Continuing her quest for revenge, Susan decided on a frontal assault.
She started out slowly, even harmlessly. She began compiling a fact
book on Kevin, Roscoe and Steve Rhoades. She talked her way into the
telephone billing office, got copies of their phone bills and devised com-
plex frequency charts on the numbers they called. She was also able to
get a customer name and address report on each number called. By
keeping close tabs on the phone bill of [o Marie, Roscoe's fiancee, she
could track him closely.
   It didn't take Roscoe long to figure out that Susan had turned on him.
She called him at strange hours and left spiteful messages on his answer-
ing machine. One day she called Ernst & Whinney, where he worked
in the data processing department, to inform the personnel office that
one of the staff members was using the company's computer terminals
after hours. As a result of Susan's call, Roscoe was fired. To elude her,
he began changing more than just his phone number; he also changed
his "cable and pair," a specific set of wires assigned to his apartment by
the phone company switching office that ordinarily would have stayed
the same when his number changed. But Susan's solution to that was
simply to get [o Marie's bill.
   There were other ways of getting phone numbers. When Kevin Mit-
nick changed his number, as he often did, Susan would drive out to his
apartment in Panorama City, clip a test set to the phone line and call a
standard number the maintenance people used to find out which number
belonged to which cable and pair.
   It wasn't long before she realized that Kevin was someone to reckon
with. He was even more familiar with the ins and outs of the phone
company than Roscoe was. In fact, Kevin had outsmarted Susan by
logging in to a phone company computer and disabling the automatic
number identification test function for his telephone. If she stood outside
his house and entered the code used to read back his number, all she got
was a failure.
   By then Kevin had begun listening to Susan's private telephone con-
versations, by attaching a hand-held ham radio and clipping it to Susan's
phone lines in a phone terminal box under a carport a few yards from
her apartment. Anyone else who wanted to listen just needed to tune in
to a little-used frequency on an FM radio. Kevin and Roscoe began
taping her calls, most of which were long late-night dialogues with

    42   A   CYEfRPLlNK
    another phone phreak she had started to date. Kevin and Roscoe tittered
    in the background as a languorous Susan described the tricks of the
    prostitute trade to her boyfriend in full detail. And they taped her own
    taped greeting from the Leather Castle, the bordello where she plied her
    trade. "Jeanine," unmistakably Susan but uncharacteristically sweet,
    gave out the current rates for each service: $45 for a half hour "if you're
    dominant," $40 "if you're submissive" and $60 "if you'd like to wrestle."
    For a while, Kevin and Roscoe followed her around in separate cars,
    communicating with each other over their radios like two cops on a trail.
       When she discovered them, Susan escalated the war by bringing it
    into the public domain.

    Bernard Klatt was one of the tens of thousands of technical workers in
    Silicon Valley, 350 miles north of Los Angeles. These intense engineers,
    predominantly men, populated the twenty square miles south of Stanford
    University in increasing numbers after World War II. They created a
    technological paradise that spawned first the semiconductor, then the
    microprocessor and finally the personal computer.
       An electrical engineer, Klatt worked as a technician at Digital Equip-
    ment Santa Clara, servicing and troubleshooting the company's comput-
    ers. For Klatt, a tall, dark-complexioned Canadian who was aloof and
    formal with strangers, as for many others in Silicon Valley, computing
    was part vocation and part passion.
       He lived with his wife in a two-story apartment building in Santa
    Clara, several blocks off EI Camino Real, Silicon Valley's main commer-
    cial thoroughfare. The building was one of thousands built as Silicon
    Valley exploded with new industry in the 1960s: a contemporary prefab
    structure with thin walls and a kitchen separated from the living room
    by a belly-high divider that served simultaneously as wall, counter and
    breakfast table.
       Like many in Silicon Valley, Klatt had given over his spare bedroom
    to his computers. But while others tinkered with inexpensive personal
    computers, Klatt was busying himself with a surplus Digital Equipment
    minicomputer. One of the company's most successful computers in the
    1960s, the machine, known as a PDP-8, was awkward to program by
    today's standards, but it was one of the first computers that didn't require
    an industrial-strength life-support system. Klatt was able to power it by
    simply plugging it into his apartment's wall outlet, an unusual feature for
    large computers of its time.
                                      1Uv~:   Tk 7)Mi4-5~   H~     'f   43
   Klatt decided to make his computer useful to a wider audience. Work-
ing with a friend, he wrote a program in the version of the BASIC
computer language used on the PDP-8 that would allow a caller with a
modem to dial in and transmit and receive messages. The PDP-8 stored
the messages and then waited to display them for the next caller. This
turned it into a bulletin board system, or BBS for short.
   Bulletin boards like Klatt's grew in popularity in the early 1980s,
spreading around the United States as part of the personal computer
explosion. They were a little like neighborhood bulletin boards found in
laundromats or community centers, offering lawn cutting services and
free kittens, but with more vitality. Since messages could be appended
to other messages and categories could be organized to be scanned
quickly by computer, the systems subdivided into special-interest cate-
gories. Most of the discussions were about computers, but subjects ran
the gamut from science fiction to odd sexual practices. Private messages
could be left for a particular user, while public messages were accessible
to anyone who logged in. Callers could check in every day to see what
had been added and to append their own comments. A kind of electronic
stream of consciousness emerged as BBSs became the digital substitute
for a neighborhood chat on the front stoop. By 1980 there were well
over a thousand BBSs around the country. Today there are at least ten
times as many.
   They covered a remarkable variety of features. Some permitted callers
to store and retrieve software. This in tum gave rise to the world of free
software and "shareware." Some programmers made substantial sums of
money by giving their software away on approval, requesting that the
users send a payment only if they found it valuable.
   Inevitably, by the early 1980s, pirate BBSs had emerged as well.
Legendary boards like Pirate's Cove in Boston permitted callers to down-
load (using modems to transfer it to their own computers) commercial
software without paying for it, and shared advice on how to break soft-
ware copy-protection schemes. Some of the pirate boards were well
known, their telephone numbers posted widely. Other boards were more
secretive, accessible only to a limited membership.
   Klatt's bulletin board, called 8BBS, was one of the nation's first bul-
letin boards for phone phreaks. Bernard Klatt had a fanatical commit-
ment to freedom of speech. After putting his system on line in March of
1980 and publishing its number on several other systems, he insisted
that his board be a free haven for its users. The ground rules, which were
set out for new callers in an introductory computer message, were simple:

         44 ...   CYE£RPUNK
         Notice:Uncontrolled message content.
                Proceed at your own risk.

         8BBS management specifically disclaims any
         responsibility or liability for the contents
         of any message on this system. No
         representations are made concerning accuracy
         or appropriateness of message content. No
         responsibility is assumed in conjunction
         wi th message 'privacy'. 8BBS acts in the
         capacity of a 'common' carrier and cannot
         and does not control the content of messages

            As a result, within months after 8BBS first began operation, a ragtag
         collection of phone phreaks and computer aficionados had discovered it
         and established it as one of the nation's premier clandestine electronic
         meeting places. Anyone dialing in was immediately struck by the dedi-
         cation that the community of several hundred regular callers had to the
         subculture of phone phreaking. They were attracted by the notion that
         they were participating in some kind of high-technology avant-garde.
         Others called just to browse or "lurk," reading posted comments without
         making their presence known. 8BBS· was frequently busy around the
            Soon the board had national scope. Callers dialed in from as far away
         as Philadelphia. Some would place their calls using purloined telephone
         credit card numbers to avoid long-distance charges. Others would ex-
         change illegal information. Credit card numbers, computer passwords
         and technical information on telephone networks and computers were
         all stored and read on 8BBS. It was impossible to tell whether the person
         sitting at the other end of a message was a skilled professional systems
         programmer or a teenager in the family den perched in front of a Com-
         modore 64.
             It was in December of 1980 that both Roscoe and Susan, already
         locked in a bitter war, found their way to 8BBS. For several months they
         became regular callers, leaving general tips and trading information,
         while at the same time, as many phreaks are prone to do, flaunting their
         particular skills.
             Even their first messages were revealing. Susan was ever the seducer:
Message number 4375 is 14 lines from Susan
To ALL at 04:38:02 on 04-Dec-80

I am new in computer phreaking and don't
know that much about systems and access. I
have however, been a phone phreak for quite
awhile and know alot about the subject of
telephones   . . ( I think) . . . anyway, I
would very much like to chat with anyone
interested in sharing information,
especially about computers. By the way, I am
a 6 ft. 2 inch blonde female with hazel
eyes, weight 140, and I enjoy travelling
alot. If anyone can suggest any neat places
to go on weekends, let me know . .

   Roscoe was the cocky self-styled techno-wizard, breathless in his first
private message to Bernard Klatt:

Message number 4480 is 20 lines from ROSCOE
[RP] to SYSOP at 18:38:27 on 06-Dec-80.
Sub j ec t: ROSCOE

46 ...   CYEfRPliNK
(213)469-          VOICE AND LEAVE MSG ON HOW

   With the arrival of the Southern California phone phreaks on 8BBS
in late 1980, the board took a disturbing twist. The anonymity that once
was a source of comfort now led to rising paranoia. Susan and Roscoe's
war spilled out in bitter public messages, and the ominous possibility that
not only the computer underground but law-enforcement agencies were
reading 8BBS became a concern.
   Susan, who was growing more and more computer literate, was often
antagonistic and haughty in her public postings. In February 1981, she
claimed that Roscoe was collaborating with the enemy:

Message number 6706 is 16 lines from Susan
To **>ALL<** at 00:44:18 on 27-Feb-81



      HAVE FUN~

         Outside of his spat with Susan, Roscoe was a good citizen on the BBS.
      Out of concern for the well-being of the board, he told other callers he
      believed the phone company had installed a line monitor on 8BBS. If
      telephone numbers were being recorded then everyone was potentially
      in trouble, regardless of their aliases. The news cast a pall over the BBS.
      The rising paranoia also affected Roscoe. He began to contemplate get-
      ting out of the underground.

        SOON TO RET IRE . . . ROSCOE.

         Then, as if to help Roscoe out the door, Susan attacked him more
      viciously than ever, adding statutory rape to the list of crimes she wanted
      the others to believe Roscoe had committed:

      • ["Anton Chernoff," an early computer programmer, was Kevin's nickname of choice
      on 8BBS.]

        48 ...   CYG6RPlJNK
           In mid-1981 the private war that Susan had taken public began to
        wane-at least on 8BBS. After Roscoe and Susan were gone, 8BBS
        continued to operate for almost another year. The board had lost some
        of its computer underground edge and shifted back in the direction of
        less controversial computer hobbyist concerns.
           But its phone phreak roots brought 8BBS crashing down. In early
        1982 Bernard Klatt received a faster modem in the mail, a gift from some
        Philadelphia phone phreaks wishing to improve the speed of 8BBS. He
        didn't know that it was stolen, purchased by mail using a false credit
        card. In April of 1982, while he was on vacation in Canada, a team of
        law-enforcement officials from the Santa Clara County sheriffs office
        and telephone company security agents came to his apartment with a
        search warrant. A fire ax shattered his front door and the agents confis-
        cated all of the 8BBS disks and backup tapes.
           When he returned, Bernard Klatt wasn't prosecuted, as he hadn't
        known that the gift modem was stolen property. But when his employer
        found out about the search of his home and the nature of the computer
        hobby, he lost his job. The 8BBS era had ended.
                                         ... ,.. ...
        Perhaps, Susan said years later, if Roscoe hadn't filed a civil harassment
        complaint against her, dragging her into the ugly legal arena first, she
        wouldn't have used the evidence. It was early 1981 and the warring had
        been going on for several months when Susan was summoned to the Los
        Angeles city attorney's office for a hearing regarding Roscoe's complaint
        that she was disrupting his life with obscene and threatening telephone
        calls to him and to his mother. Susan nodded when the hearing officer
        admonished her to cease this behavior. Her opportunity for decisive
        revenge finally came later that year, after the COSMOS incident.
           It was at the Shakey's Pizza Parlor during Memorial Day weekend in
        1981 that Kevin and Roscoe decided to break into Pacific Bell's
        COSMOS center in downtown L.A. Understanding COSMOS was es-
        sential for efficient phreaking. COSMOS (the acronym stands for "Com-
        puter System for Mail1fI~me Operations") is a large data-base program
        used by local telepf{'6[{~ cd~panies throughout the nation for everything
        from maintaining telephone cables to keeping records and carrying out
        service orders. In 1981, most of the hundreds of COSMOS systems
        installed around the country were running on Digital Equipment com-
        puters. In order to manipulate COSMOS, it was necessary to know a
        dozen or so routine commands, all of which were outlined in telephone
company manuals. A thirty-minute foraging session in the trash bins
behind the COSMOS center could yield considerable booty: discarded
printouts, notes passed between employees and scraps of paper contain-
ing passwords.
    Though most trashing was done late at night, even those who pre-
ferred to trash at midday were seldom spotted. If they were, they usually
had a plausible enough reason for being found knee-deep in the garbage.
Susan's favorite trick was to dress in shabby clothes and rummage in
Dumpsters while muttering to herself like a deranged homeless woman.
Roscoe claimed he was hunting for recyclable material.
    Roscoe, Kevin and Mark Ross, a friend and occasional phreak, set out
from Shakey's late one night in a caravan of three cars. When they
arrived at the phone company parking lot, it was a short walk through
an open chain-link fence-why, after all, should the phone company
bother locking up its trash?-into an area containing several Dumpsters.
They clambered in and started wading through old coffee grounds, dis-
carded Styrofoam cups and other garbage in search of the occasional
    Kevin and Roscoe already knew quite a bit about COSMOS, but they
wanted to get a firm grip on the password setup. Certain passwords were
needed for certain levels of privilege in the COSMOS system. The
password to the most privileged account of all, called the root account,
still eluded them. The root account was the master key to all the other
accounts on the computer system; its password would give them the
power to do anything in COSMOS.
    The trash cans' offerings were paltry that night. Perhaps another
phreak had been there earlier in the evening. Kevin found a handbook
to a computer called the "frame" that would probably be useful, but
otherwise the pickings were thin. After taking what they had found in
the trash to their respective cars, the three phreaks huddled briefly to
discuss whether to go into the COSMOS center itself. Kevin wanted to
find room 108, the office that he knew contained the COSMOS com-
puter itself, a guaranteed lode of information.
    At first, Roscoe wanted Kevin to approach the guard by himself.
Kevin would impersonate a Pacific Bell employee, tell the guard that two
friends would be arriving in a few minutes for a tour of the facility and,
if the guard bought it, Kevin would use his hand-held walkie-talkie to
transmit a Touch-Tone signal to Roscoe's walkie-talkie telling him that
the plan had worked.
    But Kevin was nervous; he didn't want to go in by himself. They
50   A   CygE~PUNK

would all go together or not at all, he said. The three argued about the
matter for several minutes until finally the other two agreed to accom-
pany Kevin to the back entrance, where a guard sat at a reception desk.
They piled into Kevin's car and backed out of the trash area. In order to
be conspicuous in arriving this time, Kevin drove around to the main
parking lot in the back of the building and parked with his headlights
beaming in through the glass door.
    If the guard thought there was anything peculiar about a Pacific Bell
employee arriving at 1:00 A.M. on a Sunday to give a guided tour to two
friends, he didn't show it. Kevin appeared far older than his seventeen
years, and he talked a convincing line. As Roscoe and Mark stood by,
Kevin chatted up the guard. First came idle chitchat: Kevin had a report
due on Monday, he said, and was upset that he had to come in to work
on Memorial Day weekend. Apparently welcoming the distraction from
his nighttime vigil, the guard didn't ask to see Kevin's company 10, nor
did he ask which department the pudgy young man worked for. In a
friendly and inquisitive manner, Kevin strolled over to a television mon-
itor that wasn't working and asked the guard about it. They both
shrugged. "Just on the blink, I guess," Kevin suggested. Then, as if he
had done it a thousand times, he signed the name Fred Weiner in the
logbook, and Sam Holliday for Roscoe. His imagination apparently
spent, he entered M. Ross for Mark Ross.
    Once inside, after a false start that led them into the mail room, the
visitors dispersed through the halls. It didn't take long for them to find
the COSMOS computer center in room 108. A metal basket attached
to the wall held a list of dial-up numbers for calling the COSMOS
computer from outside. Kevin removed the list and put it in his pocket.
And in a paper tray on a desk they found a list of codes to the digital
door locks at nine telephone central offices.
    In the Rolodex files on several desks, the three young men planted
cards printed with the names John Draper (the real name of the infamous
Captain Crunch) and John Harris. Next to John Harris's name was the
telephone number of the Prestige Coffee Shop in Van Nuys-a number
that phreaks routinely intercepted and rerouted to a loop line; next to
Draper's name was the number of a nearby pay phone. The idea was to
create instant credibility: if one of them called claiming to be John
Draper, a Pacific Bell employee, and if as a security measure the recipient
of the phone call consulted his Rolodex file to verify the name and
number, he would find the information.
    Next to room 108 was the COSMOS manager's office. The walls were

    lined with shelves holding six large manuals. A quick look through the
    books was all they needed to see that this was perhaps the most impor-
    tant thing they would find that night; they were the COSMOS manuals,
    which contained information on everything that could be done with the
    COSMOS computer. The group perused the shelves and set aside the
    books they wanted to take. Once they had gathered what they thought
    they could use, they piled it all on one desk and sorted it so as to reduce
    the bulk and appear less suspicious to the guard. Roscoe spotted a brief-
    case next to a desk; he picked it up and put the manuals inside. Then,
    each carrying as much as he could, they walked back out to the guard
    station, where Kevin signed out and exchanged hearty farewells with the
    guard. Just as casual as he was about letting them in, the guard seemed
    thoroughly unconcerned that these three young men had arrived empty-
    handed and were leaving laden with a briefcase and armloads of books.
    Kevin drove the others to their cars and the three drove in single file to
    the nearest Winchell's Donuts to divide their plunder. It had all taken
    less than two hours.
       Unfortunately, they had been too greedy. When the COSMOS man-
    ager arrived on Monday morning, he couldn't help but notice that his
    shelves were nearly bare. He notified security immediately, and ques-
    tioned the guard who had been there over the weekend. The guard
    mentioned the threesome from Sunday morning, and said he was sure
    he could recognize them again. Then the manager found two cards in
    his Rolodex with unfamiliar handwriting, which belonged to no one in
    the office. It was time to call corporate security.
                                      .... l' ....

    Meanwhile, Susan was sitting on the evidence she had been gathering
    for nearly a year. She had intimations that the L.A. district attorney
    was preparing to indict her on a dozen or so felony charges, including
    unlawful entry into Pacific Bell headquarters and conspiring to commit
    computer fraud. So she decided it was time to visit Bob Ewen, an inves-
    tigator in the district attorney's office, to blow the whistle on her erst-
    while buddies. She told Ewen that she had come to him because she was
    concerned for her nation's security: Kevin knew she had some extremely
    sensitive information, and Susan was certain that he wanted to get his
    hands on it.
       By way of example, Susan told Ewen that she had once spent several
    days locked in her apartment, and that when she emerged she told
    Roscoe and Kevin that she had been downloading and printing out

    52   &   CygE~PllNK

    missile firing parameters and maintenance schedules for intercontinental
    ballistic missiles. She boasted that she knew the schedules of the men
    who worked in the hole (the men authorized to turn the keys), and that
    she knew when the maintenance was done and what the backup systems
    were. Susan claimed that, given the right information, one pimply ado-
    lescent scrunched into a phone booth with a terminal and a modem-
    and certainly Kevin Mitnick-could set off the necessary chain of com-
    mands to release hundreds of missiles from their silos and send them
    hurtling across the globe. She hadn't tested her theory, of course, but
    she claimed to have developed a method for fooling the computer, lead-
    ing it to believe that the necessary sequence of actions preceding a first
    or retaliatory strike had already occurred. The real vulnerability, Susan
    believed, lay in the nation's communications system. The Pentagon's
    heavy dependence on the Bell System made it a sitting duck for the likes
    of Kevin Mitnick. Susan asked for immunity from prosecution in ex-
    change for doing her civic duty and testifying against Kevin and Roscoe.
        Ewen knew enough to be leery of many of the "theories" that issued
    forth from this rara avis. For one thing, he recognized her as the same
    tall, buck-toothed blond whose face was familiar to guards at Pacific Bell
    offices throughout the county. She had been spotted in Pasadena with
    Mitnick and Rhoades, and in the San Fernando Valley with the one
    who called himself Roscoe. But Ewen wanted to see Roscoe and Mitnick
    convicted. He sent her to see the district attorney. As it happened, the
    district attorney felt exactly the same way. He offered to give her im-
    munity in exchange for her full cooperation. Susan, of course, was fully
    prepared to take the witness stand and tell her version of everything.
        When Bob Ewen came onto the COSMOS case in 1981, he had
    already been dealing with phone phreaks for some time. For years, he
    kept a collection of confiscated blue boxes and other electronic fraud
    gear in a box under his desk. Usually there wasn't much to fear from
    phone phreaks, but when he went out to serve a warrant on Mitnick,
    Ewen wasn't sure what he was dealing with. As Susan Thunder had
    described the suspect, he was potentially dangerous. Ewen went first to
    Mitnick's residence to carry out a search. Mitnick wasn't home, but his
    mother was. Reed-thin and wearing a skirt that struck Ewen as too short
    for a woman approaching forty, Shelly Jaffe, who bore some resemblance
    t()p()p~ye~ girl,Oli~e~9yG seemed confused and flustered. As Ewen
    picked his way through her son's room, Shelly stood at the doorway and
    popped her gum nervously. She reminded him of a teenager who hadn't
outgrown the habit of reading movie magazines and still harbored a
dream of becoming a Hollywood discovery.
   The search wasn't easy. Mitnick appeared to be something of a pack
rat. He had apparently saved every scrap of paper and every printout he
had ever obtained, and Ewen had to scrutinize all of it. He unearthed
printouts filled with telephone company information, computer pass-
words and material from one of the computer centers at the University
of California at Los Angeles. Shelly became increasingly flustered. She
did not, she said, know anything of her son's illegal doings. She assured
Ewen that Kevin could not be responsible for the things the police
suspected. Kevin, she insisted, was an angelic son.
    "Then what's this?" Ewen asked, holding up a printout of telephone
company information.
    "I've never seen that before," she responded.
    "When is the last time you were in your son's room?"
    "I don't remember."
    Ewen got the impression that Shelly, who displayed no hint of malice
herself, might be intimidated by her son. It was evident that she tried to
maintain a decent home. Aside from Kevin's cluttered room, which was
actually tidy compared with some Ewen had seen, Shelly kept a neat
household. But she appeared to have no control over the boy she had
raised or the situation at hand. She seldom saw Kevin. Her early hours
as a deli waitress rarely coincided with her son's more nocturnal sched-
ule. Clearly, she knew nothing about computers and had no desire to try
to learn anything about them. It was Ewen's hunch that even telephones
might overwhelm her.
    Among all the evidence Ewen gathered in Kevin's bedroom, he found
nothing linking the suspect to the COSMOS break-in. There were no
manuals, directories or other signs that Kevin had been there. Still, on
the strength of the guard's identification of Mitnick, Ewen had obtained
an arrest warrant. Although the Mitnick family hadn't been particularly
observant Jews, Kevin often went after school to the Stephen Wise
Temple in the private community of Bel Air, where he worked part-
time. Ewen drove to the synagogue with three other officers; the four
men divided into two cars and waited in the parking lot for Mitnick to
show up. When Mitnick began to pull into the lot, he apparently saw
the men sitting in their unmarked cars, because he hesitated, then drove
by. As soon as he took off, the police peeled after him, sirens ablare.
Mitnick's car accelerated. Just as Mitnick pulled onto the 405 freeway
54 ..    CYGERPl1NK
headed north, his pursuers overtook him and forced him onto the shoul-
   At first, Ewen wasn't sure what he was up against. The suspect was
very large, at least two hundred pounds, and possibly armed. Ewen feared
Kevin and his buddies might have modified the COSMOS computers
and planted a "logic bomb"-a hidden computer program designed to
destroy data at a given time. This made Ewen worry that he might even
have a terrorist on his hands. So they handled him as roughly as they
would a murder suspect: guns drawn, they pushed him onto the hood of
his car and handcuffed his hands behind his back. It wasn't until the
officers felt not muscle under his shirt but soft and pliable flab that they
realized this young man was no physical threat and loosened their grip.
Mitnick complained that he had to go to the bathroom immediately.
Then he began to cry. "You scared me. I thought I was going to die."
   "Why did you think you were going to die?" Ewen asked.
   "I thought they were after me."
   "Who?" asked Ewen, bewildered.
   "You know, there are a lot of people that don't like me." Kevin didn't
mention it to Ewen, but a few months earlier he had been hunted down
by an irate ham operator who was fed up with Kevin's high jinks. Ewen
told Kevin he was sorry to hear that people didn't like him. He was just
doing his job, he said, and would have to take the teenager to jail in
handcuffs. In the car, Kevin began to talk. He admitted to knowing
Roscoe and Susan and Steve Rhoades. Then, as if he suddenly realized
he was talking to an officer, he stopped talking. When Ewen asked him
if he had put a logic bomb in the phone company computers, Kevin
looked as if he might start to cry again.~NQ.,~heinslsl~d".h~:wgllldJ1eyer
do anything t()h~na c()I1lputer.                                          .
   Recovering the COSMOS manuals proved more difficult than finding
the suspects. Once Kevin had been taken into custody and his mother
and grandmother had hired a lawyer for him, Ewen began to press the
lawyer for the missing manuals. Ewen was particularly worried that the
manuals would be copied and distributed throughout the phreaking com-
munity. He shuddered at the possibility. With passwords and dial-up
numbers, phreaks could shut down much of the phone service in the
greater Los Angeles area. Kevin's attorney insisted that Kevin didn't
have them, but that he might know where to find them. A few days
later, the lawyer delivered all of the stolen manuals.
   The prosecutor decided to fold the cases against Kevin and Roscoe
together. Both were charged with burglary and grand theft, and with
conspiring to commit computer fraud-felonies under California law.
The charges would be for the U. S. Leasing incident, already more than
a year old, and the COSMOS break-in. U.S. Leasing's John Whipple,
who had long since dismissed the entire incident as a bad dream, was
surprised to hear that the vandals who had crippled and defiled
his company's computer had been apprehended and charged with
having "maliciously accessed, altered, deleted, damaged and destroyed
the U.S. Leasing system." Mark Ross was charged with burglary and
conspiring to commit computer fraud in connection with the COSMOS
   After Roscoe was arrested, his mother gave him a choice. He could
either use her money to hire her lawyer, an old friend from Argentina
who knew nothing about computers or computer crime, or draw on his
own meager resources and hire whomever he pleased. Although he had
found a good criminal lawyer who had eX£Q"ience in cases such as his
and would charge less than his mother': fritnd\ Roscoe c~~ld do nothing
to persuade her to pay. He ended.u P W~lt.l!t e c'e purer-illiterate Argen-
tine.                              J
   While the defendants were ouf on a-tf            e prosecution was pre-
paring its case, Ewen decided it Imr ht be a goo idea to eavesdrop on
one of the Shakey's meetings. So one night             the trial, Ewen, the
juvenile prosecutor, and Pacific Bell and   gIg,   investigators dressed in
the most casual clothing their closets could produce and convened at the
Hollywood Shakey's. They seated themselves at a table about thirty feet
away from the group. Neither Roscoe nor Mitnick was in attendance.
Rhoades was also absent. Instead, seated at the phreaks' table, the center
of attention, was Susan. The group of law enforcers sat in disbelief as
their star witness took over the meeting, trading phone information with
the others, oblivious to the presence of the men she was working for.
When she and another phreak got up to leave, Ewen followed her to the
parking lot and watched as she retrieved a printout from the other
phreak's car. "Susan," Ewen called to her as she was returning to the
restaurant, "what do you think you're doing?" It took Susan a few sec-
onds to focus on Ewen; indulging her vanity, she seldom wore her
glasses. When she recognized Ewen, she grew defensive. "Hey, I thought
I could get some stuff you guys could use."
   Ewen looked at her sternly.
   "Okay," she conceded, "so I still like to do this stuff."
56   A   CYf:ERPlINK
Shortly before the hearing, Kevin Mitnick pled guilty to one count of
computer fraud and one burglary count. In exchange for the dismissal of
two additional charges and in the hope of being sentenced to probation
instead of a term in the California Youth Authority prison system, Kevin
agreed to testify for the prosecution, even if it meant betraying Roscoe.
    On December 16, 1981, a pretrial hearing was held. The prosecution's
first witness was Susan, who cut an imposing figure as she strode to the
witness stand and swore to truthfulness. Susan had been given immu-
nity. Under assistant district attorney Clifton Garrott's gentle question-
ing, she recounted Roscoe's trespass into the U.S. Leasing system. He
had arrived one evening the previous December to take her out on a
date, and as she was getting ready he sat down at her computer terminal
and dialed into U.S. Leasing. When she asked him what he was doing,
Susan told the court, he said he was using a program called god to get
into a privileged account, in order to run a program called MONEY to
print out the passwords on the system. Susan went on to explain in
detail how the MONEY program could only be called up using a certain
    Unable to follow the exchange between Garrott and his witness, the
judge interrupted. "Slow down a minute," he demanded. "This is a
whole new education for me over here."
    The proceeding digressed for a few minutes while Susan presented an
introductory course in computer terminology for the uninitiated in the
room. She explained passwords, logins and account numbers. When
Roscoe got the MONEY program to run, she continued, she saw him
start to write down passwords and account numbers. And when she
looked at the screen a few minutes later, she recalled, the words delete
and zeroed had appeared.
    At that point, Garrott produced a ragged eight-and-a-half-by-eleven-
inch sheet torn from a three-ring binder and held it up for the judge to
see. Roscoe was stunned. When he had arrived home that night after
this so-called date with Susan, he had reached into his back pocket and
realized the piece of paper was gone. Little had he suspected that she
would pick his pocket!
    Garrott continued his questioning. He asked Susan if she had asked
Roscoe what he was doing at her terminal. "Yes," she answered. He said
he was "taking care of business . . . getting even." Then, she said, he
began, "laughing hysterically, crazylike." At the prosecutor's gentle
prodding, she said she was upset by Roscoe's behavior that night. "The
vibes weren't right. There was a lot of tension between us," she told the
court. "I said, 'Why did you do that? You know it was destructive.' "
    "What did he say?" asked Garrott.
    "I got no response."
    Jose Mariano Castillo, Roscoe's attorney, did his best to discredit
Susan as a witness.
    "When in December of 1980 did this incident take place?" he asked
    "I believe it was the twelfth or thirteenth."
    "What day of the week was that?"
    "Thursday or Friday."
    "Were you employed?"
    "Where were you employed?"
    "The Leather Castle."
    "What was your occupation at the Leather Castle?"
    Garrott leapt to his feet to object. The question was irrelevant and
immaterial. "What is the relevance of the type of work that she was
doing there?" the judge asked Castillo.
     "Well, it goes to credibility, Your Honor."
     "Well, let me ask her specifically," and he turned back to Susan.
"Were you a prostitute?"
     "No," Susan replied hotly, expressing indignation at the very sugges-
     Garrott objected to the question and the judge sustained his objec-
     Castillo continued. "Have you ever been prosecuted for prostitution?"
     This time, Garrott objected before Susan could answer.
     Try as he might, Castillo could pursue no line of questioning that
 might help to impugn her testimony. Each time Castillo asked a question
 of Susan, Garrott objected to it as irrelevant. The judge sustained all but
 a few of the objections. When the prosecutor didn't object, Susan was
 coy with Castillo, often craning forward and asking him to repeat the
 question, as if she were hard of hearing, or confused.
     Castillo asked her if she had been in love with Roscoe.
     "Infatuated, perhaps," Susan responded. "In love? No."
     Castillo asked if Roscoe had been rejecting her advances. "Yes" was
 Susan's reply.
58   •   CygERPUNK
   "And you felt upset because of the rejection?"
   "No. He wasn't the only man I was going out with."
   "How many people were you going out with, dating, at that time?"
   This time, the judge overruled Garrott's objection and instructed
Susan to answer the question.
   "Three or four."
   When Castillo began to ask her about her drug problems, Garrott was
quick to object again. Castillo persisted. Susan's drug problems, he
claimed, had some effect on her emotional state. Moreover, he said, she
had sexual problems. The judge rebuffed him and sustained Garrott's
   The hearing dragged on this way into the afternoon. Every time
Castillo tried to impute vengeful motives to Susan, to sully her credibil-
ity or to introduce other names-Kevin Mitnick, Steven Rhoades-as
possible suspects, Garrott leapt to his feet and the judge ruled in his
favor. All told, Roscoe's prospects were beginning to look bleak. If this
pretrial hearing was the legal equivalent of a movie's rough cut, Roscoe
could easily imagine how the finished film would turn out.
   At the end of the afternoon Kevin was called to testify. He began by
explaining the term phone phreak to the court. While refraining from
placing himself in any category in particular, Kevin described different
"types" of phreaks. "One type is a person that likes to manipulate tele-
phone lines and computers," he lectured. "Another one is the type that
just likes talking with other people on conference lines. And stuff like
that. "
   Kevin recounted the visit to the COSMOS center in precise detail.
The Shakey's meeting. The trashing. The unsuspecting guard. The for-
aging through room 108. The stealing of the manuals. Garrott showed
particular interest in what transpired later at Winchell's Donuts. Kevin
told the prosecutor that the manuals were left in the cars but the door
code list was brought inside, and that he later made a copy of the door
codes for himself. Afterward, Kevin said, he and Roscoe divided up the
manuals from behind Kevin's car.
   Despite his best efforts, Roscoe's attorney wasn't making much head-
way in building up a convincing case for his client's innocence. Rather
than take chances with a trial and risk fumbling the defense in court,
Roscoe's attorney decided to work out a plea bargain with the prosecutor.
On April 2, 1982, a day before his twenty-second birthday, Roscoe pled
no contest to the conspiracy charge and to the charge of computer fraud
against U.S. Leasing. Two months later, he was sentenced to 150 days

    in jail. Ross got thirty days. Kevin had considerably more luck than his
    confederates: after a ninety-day diagnostic study mandated by the juve-
    nile court system, he got a year's probation.

    While Roscoe served out his sentence, Susan dropped her alias and
    refined her new entrepreneurial persona. She became a security consul-
    tant. In April of 1982 she appeared on the television news show
    "20120." Susan wore a purple blouse beneath a wide brown jumper. The
    outfit was set off by a pair of oversize star-shaped earrings that swung
    conspicuously amid limp, straight blond hair. Why, asked an inquisitive
    Geraldo Rivera, had she and her friends wrought such havoc with the
    phone company? "I was getting off on the power trip," she replied. "It
    was neat to think that I could screw up the Bell System." Hassling
    someone-disconnecting his phone or canceling his car insurance-was
    all standard practice. "Is there any system that can't be gotten into?"
    she queried herself. Then she answered her own question. "God, I
    live by an old saying: if there's a will, there's a way. There is always a
       In the next two years, Susan became a peripatetic Cassandra, travel-
    ing around Los Angeles, painting doomsday scenarios for her clients.
    She developed a standard set of these grisly prognoses and trotted them
    out for reporters and lawmakers. A year after her "20/20" appearance,
    under the auspices of an FBI agent, she flew to Washington to expound
    her theories before the U.S. Senate. During her testimony, Susan ex-
    plained about phreaking and trashing, or "garbology," as she put it. She
    recalled with some nostalgia the talents and feats of the L.A. gang. She
    talked about posing as repair personnel and altering credit ratings.
    She described the practice of using the U. S. Leasing computers.
       William Cohen, the chairman of the Senate subcommittee holding
    the hearings, then asked, "There was no indication they felt there was
    anything wrong about going in and using U.S. Leasing's program? In
    other words, if you walked into somebody else's house and decided you
    wanted to watch television for a couple of weeks or a year and just walk
    in and turn on somebody's set, there would be nothing wrong with that?"
       "I guess if people were stupid enough to leave their door unlocked,"
    Susan replied. "Hey, if people are stupid enough to leave the system
    passwords around that way-that's how the group felt."
       Upon learning that Susan's formal education had stopped at the
    eighth grade, the incredulous senator remarked, "It seems it would be so
60   •   eyg£RPUNK
easy for one of our major adversaries to secure the services of people like
yourself who have no advanced degree or training."
   "I agree," Susan responded. "It poses a very serious national concern,
a very serious threat to national security. I wanted to find out how
everything worked. I have got to admit to a certain interest in something
that was supposedly classified as top secret."
   "Journalists have the same fascination with classified information,"
Cohen remarked.
   "I studied to be a journalist when I was in junior high school."
   "I was afraid you were going to say that."

While in Washington, Susan got a chance to demonstrate her "social
engineering" skills. As Susan later told the story, a team of military brass
-colonels and generals from three service branches-sat at a long con-
ference table with a computer terminal, a modem and a telephone.
When Susan entered the room, they handed her a sealed envelope
containing the name of a computer system and told her to use any
abilities or resources that she had to get into that system. Without
missing a beat, she logged on to an easily accessible military computer
directory to find out where the computer system was. Once she found
the system in the directory, she could see what operating system it ran
and the name of the officer in charge of that machine. Next, she called
the base and put her knowledge of military terminology to work to find
out who the commanding officer was at the SCIF, a secret compart-
mentalized information facility. Oh, yes, Major Hastings. She was
chatty, even kittenish. Casually, she told the person she was talking to
that she couldn't think of Major Hastings's secretary's name. "Oh,"
came the reply. "You mean Specialist Buchanan." With that, she called
the data center and, switching from nonchalant to authoritative, said,
"This is Specialist Buchanan calling on behalf of Major Hastings. He's
been trying to access his account on this system and hasn't been able to
get through and he'd like to know why." When the data center operator
balked and started reciting from the procedures manual, her temper
flared and her voice dropped in pitch. "Okay, look, I'm not going to
screw around here. What is your name, rank and serial number?" Within
twenty minutes she had what she later claimed was classified data up on
the screen of the computer on the table. A colonel rose from his seat,
said, "That will be enough, thank you very much," and pulled the plug.
   Computer security experts had been suspecting for years what Susan

    was proving time and again: the weakest link in any system is the human
    one. Susan liked to illustrate her belief with the following scenario: Take
    a computer and put it in a bank vault with ten-foot-thick walls. Power
    it up with an independent source, with a second independent source for
    backup. Install a combination lock on the door, along with an electronic
    beam security system. Give one person access to the vault. Then give
    one more person access to that system and security is cut in half. With a
    second person in the picture, Susan said, she could play the two against
    each other. She could call posing as the secretary of one person, or as a
    technician in for repair at the request of the other. She could conjure
    dozens of ruses for using one set of human foibles against another. And
    the more people with access the better. In the military, hundreds of
    people have access. At corporations, thousands do. "I don't care how
    many millions of dollars you spend on hardware," Susan would say. "If
    you don't have the people trained properly I'm going to get in if I want
    to get in."
      The list of cohorts with whom Kevin was ordered not to associate
was comprehensive. It included Roscoe, Susan, Steven Rhoades and
Mark Ross. Kevin also had strict orders to stay away from all phone
phreaks in general. In effect, his circle of acquaintances was pared con-
siderably. But there was always Lenny.
    Lenny DiCicco had never been attracted to phone phreaking, or
trashing, or even social engineering. His was a fascination with buttons.
As a little boy in Oak Park, outside of Chicago, Lenny wouldn't bother
with a toy unless it had a knob to turn or a lever to push. Even as a
toddler, he had a knack for finding the switch to stop an elevator be-
tween floors, or the button that would halt an escalator. When he was
five, he managed to activate the fire alarm of one of Chicago's largest
hospitals, sending nurses and physicians scurrying through the halls. But
to his parents, Lenny's compulsion to play with gadgets was an enigma.
Gilbert DiCicco (pronounced "duh-ssa-ko") was an illustrator at the
Chicago Tribune; Vera DiCicco was an artist, too. Their only child es-
chewed crayons in favor of toys with moving parts and motors.
    When Gil DiCicco took an illustrating job with a San Fernando
Valley newspaper in 1977, the family moved to Los Angeles. In his very
first minutes in Los Angeles, twelve-year-old Lenny exhibited surprising

enterprise. As Gil and Lenny waited for Gil's brother to pick them up at
the airport, Lenny discovered that a quarter was refunded for each rental
cart returned to the automated dispatcher. For half an hour, Lenny ran
furiously around inside the terminal, rounding up stray carts and collect-
ing the refunds. An auspicious omen, thought Gil, for starting their new
    Lenny enrolled at Sepulveda Junior High School in Mission Hills. His
formal introduction to computers came at Monroe High School, in John
Christ's introductory computer class. Lenny demonstrated an intuitive
understanding of the labyrinthine integrated circuits around which com-
puters were built and an aptitude for programming. It was the start of
the personal computer era and he developed a passion for computing.
And like many his age, the. gangly teenager also displayed a special
interest in communicating with remote systems over telephone lines.
When Lenny managed to log on one day to the school district's central
computer, Christ could only chuckle. Two years earlier, a student of
Christ's had demonstrated the same talent by using the simple classroom
terminal to access the school district's Digital Equipment computer sys-
tem. His name was Kevin Mitnick. He had dropped out of high school
and taken an equivalency test for his diploma. "Oh, no," Christ groaned
and smiled at Lenny, "not another Kevin!"
    Nothing could have pleased Lenny more.
    Lenny knew of Kevin Mitnick, who was something of a legend in Los
Angeles high school computing circles. As a student at Monroe, Mitnick
had absorbed information like no student before him. His telephone
company exploits had been described at length in the Los Angeles Times.
By Mitnick's own account to the FBI, the paper reported, he and his
friends had gained unauthorized access to computers all over the United
States-systems supposedly impenetrable to all but selected people who
know the passwords. Mitnick also told the FBI that he had obtained
sensitive data from "the Ark," one of the main systems in Digital's
software development group. In suburban Southern California, where
high school heroes traditionally emerged on the gridiron, Kevin had
been the perfect antihero: a nerd who used technical wizardry to befuddle
the authorities. There seemed to be no end to his clever tricks. While
at Monroe, Kevin had reconfigured a modem line so that he could use it
to dial out of the school's main computer and into others. The modifi-
cation gave him a perfect cover for his activities. Anybody who at-
tempted to trace the phone calls would inevitably run into a cold trail at
the computer center.
,-----   -----   ..   _-----------------------------

          64          A   CYE6RPVNK
             Lenny first met Mitnick in 1980 when a mutual friend introduced
          them. Lenny and his friend were on their way to a conference of an
          organization for Digital Equipment customers called DECUS. They had
          stopped by Kevin's house to invite him to come along. He declined, but
          asked them to keep their eyes out for interesting literature. National
          DECUS meetings usually attract as many as twenty thousand Digital
          users from around the country. For high school students to be intrigued
          by Digital computers was unusual enough, but for them to go out of their
          way to attend a DECUS convention was a sign of extreme dedication.
          Mingling with the attendees, they could be as anonymous as they
          pleased. The convention hall was a sea of terminals clustered within
          exhibition booths that demonstrated the latest in Digital computers,
          software and product literature.
             The day after his trip to the DECUS meeting, Lenny got a call from
          Kevin, who asked if he could get copies of the manuals Lenny had picked
          up. Lenny had finally met someone whose unusual passion matched his
          own. Like Lenny, Kevin Mitnick displayed little respect for his elders
          and even less for the institutions over which they presided. Someone
          had told Lenny that when Kevin was an adolescent, his mother had so
          much trouble controlling him she sent him to a disciplinary camp for
          incorrigible children. On top of the much-publicized COSMOS inci-
          dent, Mitnick had gained notoriety for breaking into computers at col-
          lege campuses around Los Angeles. Lenny knew he was dealing with a
          veteran, and although Lenny was learning his way around the computer
          underground quickly, he knew he still had a long way to go.
             Lessons from Kevin were far more interesting than anything Monroe
          High School had to offer. Lenny had always found school a useless bore.
          His attendance record at school was already spotty, and once he met
          Kevin it got worse. Kevin had a lot of time on his hands. He was working
          as a delivery boy at Fromin's, a delicatessen in the heart of the San
          Fernando Valley owned by Arnold Fromin, who was living with Kevin's
          mother in Panorama City. Hungry for a terminal to use, the pair discov-
          ered the San Fernando Valley's Radio Shack stores. Each store had a
          demonstration model of a personal computer called the TRS-80. Because
          the computers were also used to send inventory updates each evening to
          Radio Shack's Fort Worth headquarters, they were equipped with mo-
          dems as well. Lenny and Kevin merely had to supply a communications
          software program to transform one of these computers into a full-Hedged
          terminal. Using stolen MCl codes, they could dial long-distance into
          any computer they could find. Kevin's initial method was to talk some-

    one out of a password. Once he was in, he no longer needed to employ
    his verbal skills.
       What were supposed to be five-minute demonstrations became mara-
    thon six-hour reconnaissance sessions. Kevin sometimes pushed his luck,
    imploring a manager to keep his store open well past closing time. They
    could last for about a month at each store. When a manager got irked
    enough by their constant presence to ask them to take their hobby
    elsewhere, they simply combed the Yellow Pages for new possibilities.
    They became techno-nomads.
       The two teenagers turned their attention to the University of South-
    ern California. Kevin had already been caught fiddling with computers
    on the USC campus a couple of years earlier, but, as usually happened
    on the college campuses where he was caught, he was "counseled and
    released." When he returned with Lenny at his side in 1982, no one
    recognized him. Kevin already knew the locations of campus terminal
    rooms that stayed open twenty-four hours a day, available to anyone who
    could pass as a student.
       Both Lenny and Kevin were largely self-taught and they were still
    learning. Most of what they picked up about computers came not from
    time spent in classrooms or computer labs but from hours spent with
    whatever system they could steal some time on. Most of the time, the
    computers were Digital systems, the most popular computers on college
    campuses. So it wasn't surprising that the computers they came to know
    best were Digital.
       By the early 1980s, Digital had augmented its PDP series of minicom-
    puters with the VAX line (VAX stands for "Virtual Address Extension,"
    which refers to a way to expand memory that speeds a computer's perfor-
    mance). These computers were designed in such a way that all models
    could use the same software and share data over computer networks.
    With the VAX, Digital strengthened its position in the commercial
    market and started competing successfully in the banking, insurance and
    accounting markets, traditional IBM strongholds. But Digital still had
    its roots in the technical and university community. And as its loyal
    customers at universities such as USC migrated to the VAX, so did
    Kevin and Lenny.
       Phreaking as an end in itself had never done much for Lenny. His
    affinity for the telephone began and ended with his frequent lengthy
    conversations with friends. He felt uneasy about misrepresenting himself
    in order to get information or coerce a password out of someone. And
    Lenny didn't have Kevin's appetite for electronic revenge. Kevin used
66   A   CYG£RPliNK
phone company computers to wage his private wars against whoever he
thought had crossed him. Lenny watched one evening as Kevin attached
a local hospital's $30,000 telephone bill to the home phone of a fellow
ham radio buff whom Kevin disliked. But there was something that drew
him to Kevin. His own reluctance notwithstanding, Lenny was intrigued
by Kevin's ability to feign and cajole his way around almost any phone
company office, or even a police department. Kevin could call someone
at a switching office, for example, and convince him to drive fifteen
miles just to turn a computer on. The secret lay in convincing the person
on the other end of the line that he was the supervisor of that person's
supervisor. He would gain that person's trust by speaking the same lan-
guage. It was often remarkably simple. Kevin's use of telephone company
vernacular snowed most customer service operators, whose full coopera-
tion he needed to, say, pull a victim's toll records.
   And some of Kevin's phreaking tools came in handy for Lenny. The
telephone company's loop lines-the special numbers it used for testing
customer service-were ideal for certain tasks. For example, when
Lenny and Kevin applied for jobs and needed to supply a reference, they
could give out a test line number, sit on the line waiting for the call,
and give the reference themselves. And they made frequent use of phone
company test numbers that provided a constant busy signal, or simply
rang and rang. To them, it seemed more like harmless hoodwinking
than fraud. Free long-distance calls using credit card numbers were per-
haps the best part of Kevin's phreak repertoire. But when Lenny tried to
show his father how to cash in on the free service, he got a stiff repri-
   Computers were Lenny's abiding passion. At fifteen, he was still too
young for his driver's license, so on weekends and school holidays his
father drove him to the nearby Northridge campus of California State
University to use the school's computers. Lenny usually asked to be
picked up six or seven hours later. The senior DiCicco understood noth-
ing of Lenny's fascination with computers and things technical. Occa-
sionally Lenny tried to explain things to him, but his words sailed over
his father's head. Gil DiCicco had grown up in the 1940s and 1950s,
when relatives would shout into the telephone to make themselves heard
on long-distance calls. Gil didn't see much to celebrate in this so-called
computer revolution of the 1980s. If anything, he lamented the loss of
communities as he understood them, where people met in person and
not electronically, and where teenagers did more than sit in front of
computer screens. He was sad to see that young computer enthusiasts
felt more comfortable interacting with machines than with people. He
was a humanist who prided himself on having nothing to do with these
machines. Still, he knew enough to guess that Lenny's computer fixation
would land him a well-paying job some day, so he was glad for the time
Lenny spent inside the Northridge computer center. Lenny had told him
that he had made many friends there, and although he spent a large
portion of the day playing computer games, it was keeping him off the
    But when Gil got a call from the campus security office one Sunday
afternoon, saying he should come retrieve his son, who had just been
caught illegally logging on to the highly restricted administrative com-
puter used by the school, among other things, for recording grades, he
was sure Kevin Mitnick had something to do with it.
    Kevin Mitnick was in fact Gil DiCicco's worst nightmare. Lenny had
never shown much interest in school, but he usually managed to bring
home fair grades. But when Lenny started hanging around with Kevin,
his grades took a dramatic plunge. Gil had only met him once, but there
was something unsettling about the overweight teenager. He was always
whispering things to Lenny in front of others. This guy wasn't just a bad
influence. He was hurting Lenny's chances of ever making good on the
future Gil was trying to give his son.
    First, Gil admonished Lenny, trying to impress on him how lucky he
was that Cal State-Northridge hadn't pressed charges. Furthermore, he
told him, he didn't want to see him spending time with Mitnick. The
senior DiCicco then decided to take a rash step to see to it that his son's
unwholesome friendship with Kevin Mitnick ended: he went to Fromin's
and confronted Kevin. "Look," he said, "just stay away from Lenny. It's
doing both of you no good." Gil didn't want to see Kevin's face around
the DiCicco house again, he told him flatly. From the dull expression on
Kevin's face, Gil figured he might as well have delivered his speech to
the nearby bowl of half-sour pickles. Kevin shrugged and smiled, avoid-
ing eye contact with Gil. "Uh, okay, I've gotta get back to work now,"
he replied, and turned away. Gil had no illusions that this little chat
had made any impression.
    Now that Northridge was off limits, Kevin and Lenny spent weeks at
USC, building up their own small empire of purloined accounts. They
had managed to get accounts on all the university's computers. When
 they arrived one evening as usual, they saw that of their six accounts,
all but one had been disabled. It was obvious that someone in the
computer lab was on to them. Lenny was worried, and he warned Kevin
that it was probably a setup. He wanted to leave, but Kevin wouldn't
budge, insisting that they restore their lost accounts and that they stay
on campus to use the university's own high-speed links. In the end,
Lenny shared Kevin's delusion that what they were doing was undercover
work worthy of a Hollywood spy thriller. Brazenly walking onto the
campus was the kind of thing Robert Redford would do as the hunted
CIA researcher in the movie Three Days of the Condor. Kevin's favorite
movie, it told the story of a former English literature graduate student
hired by the CIA to extract plots from novels, who stumbles onto a
conspiracy within the agency. In one dramatic scene Redford masquer-
ades as a telephone lineman and throws the agency's high-tech surveil-
lance gear off the track by crossing some wires. And Lenny's own code
name was Falcon, after Christopher Boyce, the young TRW employee
and protagonist of the book The Falcon and the Snowman who in the mid-
seventies fed the Soviets secret military technical data that traveled over
TRW's satellite communications system.

For all Mark Brown knew, it could have been a faculty member who was
monkeying around with the system. He wouldn't have been surprised if
it turned out to be a student looking for a way to make trouble. Four
years earlier, in 1978, two USC students, unhappy with their grades and
financial aid, had made news when they were arrested for trying to tap
into the university's computer to improve their standing. Those two
failed, but years later, in 1985, a year-long investigation would uncover
a ring of some thirty USC students in league with someone in the
school's records and registration office, who routinely changed grades
and created fraudulent degrees. They charged handsomely for the ser-
vice, selling doctoral degrees for as much as $25,000.
    Brown, a young assistant in the computer lab, hadn't encountered
any malicious computer attacks. In 1982, USC's attitude toward com-
puting was typical of most universities'; it wanted to keep its computers
open and accessible to its users. In the late 1970s and early 1980s,
security had yet to become a serious issue on college campuses. Setting
up a reliable password system was just about the only security barrier at
places like USC. Because so many of Digital's customers had told the
company that they were more interested in convenience than in secu-
rity, the operating system software was shipped with many of the built-
in security features disabled. While customers could, by reading the

    instruction manuals, activate those security features, most customers
    chose not to, since increased security meant increased inconvenience.
       In 1982, computers and computer terminals were proliferating on
    college campuses so quickly that it was almost impossible for computer
    managers with their limited resources to pay much attention to security.
    Besides, placing undue restrictions on computer usage would have been
    like sectioning off the university library for selective access. Librarians
    weren't there to keep people from the books, but to lead them as easily
    as possible to the books they were seeking. So it was with computers as
       Still, a modest hierarchy had to be imposed on the USC computer
    systems; the people who watched over the system needed a master key
    to everyone's account, just in case they needed to retrieve a lost file or
    fix a student's account. Like any other computing center, USC had
    system managers with full access to everyone's account, and Mark Brown
    was one of them. Part of his job was to monitor the system's activity, to
    make sure that the machines were running correctly and to provide
    assistance where needed. One day, he noticed that some strange things
    were happening. Someone was logging in to privileged accounts nor-
    mally used only by system managers. So, taking the role of detective, he
    started trying to track the trespasser down. His first step was to place
    some secret traps in the operating system to catch anomalous behavior.
    Next, he wrote a program that watched for unusual activity on the
    accounts the trespasser used most frequently. Finally, he began printing
    out the records of the break-in attempts.
       After a few days, it was clear that the USC interloper was exploiting
    a loophole in the campus computers' operating systems to get access to
    the most privileged accounts on the system and then create his own
    account names and passwords. Alternatively, he used the accounts of
    others to break into privileged accounts. The printouts showed that he
    was coming in over off-campus phone lines. Brown's first move was to
    call the phone company and have the line traced back.
       To do that, he learned, he would need to notify the police and obtain
    a court order. So he decided to take it on himself. Rather than shut the
    intruder out of the system entirely and give up the chance to track him
    down, Brown limited his access and minimized the amount of damage
    he could do. Judging from the intruder's actions, his motives seemed
    relatively benign-that is, until the day one of the log scripts showed
    him downloading university accounting files. These are files that keep
70 ..    CYEfRPLlNK
records of the establishment and deletion of accounts, and they often
include password entries. Not only did the intruder have the accounting
files, he also had managed to use a privileged account to steal personal
mail files from systems people like Brown, presumably in order to glean
further information about the computer system. That was stepping over
the line. Gentlemen do not read other people's mail. Brown began
taking countermeasures.
    Brown considered himself a hacker in the hallowed, traditional sense
of the word. To him, hacking was the honorable pursuit of perfection in
computer programming. It did not mean breaking into someone else's
system. Brown decided it would take someone who thought like an
intruder to catch one. He called for help from a couple of his fellow
programmers, including Jon Solomon, a former phone phreak, and
started the chase.
    After a week of intensive monitoring, they realized that on certain
days the phantom visitor was entering the school's computers at much
higher speeds than usual. He could do that, they reasoned, only from a
specific terminal room, where the only high-speed links could be found.
Brown had been around the computer community long enough to see a
lot of cockiness, but this was incredible!
    They knew that the next time the high-speed connection came, it
would take a bit of maneuvering to isolate where the intruder was in the
computer lab, but it could be done. To make things more efficient, they
decided to risk scaring him off with their obvious tampering: they limited
his access to one machine.
    One night when Brown went out to dinner, he left Solomon to watch
for the trespasser. No sooner had he left than the intruder was on line-
and his connection was coming from inside the campus. It took Solomon
only a few minutes to trace the terminal. It was spooky: the interloper
was at work in the very lab where Solomon was sitting. Solomon walked
across the room and stood a few feet behind an overweight young man
with a familiar face who was sifting through a stack of papers he seemed
to be using as a reference. He was oblivious to anything else in the room.
Solomon stepped a couple of feet closer to look over the intruder's
shoulder. The papers were copies of the stolen accounting files. Next to
the intruder sat another man who seemed much younger than the first.
They appeared to be together. Suddenly Solomon placed the familiar
face: he had met the chunky hacker weeks earlier at a DEeUS confer-
ence. The hacker had struck Solomon as pretty cocky, boasting about
his exploits. Solomon knew that his name was Mitnick, and that he had
been arrested in the past for breaking into computers. Neither Mitnick
nor his friend was enrolled at usc. Hurriedly, Solomon stepped away
and called the campus police.
   In the few minutes it took for the police to arrive, Solomon watched
as the pair kept typing, logging in and trying out different passwords.
Solomon was beginning to get the feeling that he could have watched
them for an hour without their noticing him, so engrossed were they in
the task at hand. When the two officers appeared in the doorway,
Brown's assistant pointed to the older of the two strangers and they
flanked his chair. When the plump young man looked up, his face
expressed none of the guilt or remorse that Solomon expected to see.
Here was a known computer criminal who wasn't merely breaking into
the computers of a private university with which he had no legitimate
connection-he strolled onto campus to do it. Solomon expected to see
some sign of self-rebuke-at least some surprise-that he had been
caught in the act. Instead, when the police questioned him, Mitnick
took offense. "I'm not doing anything wrong," he said. The papers he
was working with, he claimed, belonged to him. His friend sat nearby,
discernibly more nervous. When the officers asked Mitnick to step out-
side, he picked up his thick stack of computer printouts, tucked the sheaf
under his arm and obliged. The officers took both of them to the campus
police station.
   At the security office, the police took their names: Kevin David Mit-
nick and Leonard Mitchell DiCicco. When Brown and Solomon ques-
tioned Mitnick, his attitude was supercilious, as if under other
circumstances he wouldn't deign to answer. He taunted Brown for what
he claimed were unsophisticated sleuthing tactics. They never would
have caught him, he claimed, if he hadn't made himself such an easy
target by coming onto campus.
   In a records search, the campus police discovered that Mitnick was
on juvenile probation for breaking into other computers and that Di-
Cicco, barely seventeen, had been caught six months earlier doing the
same thing at Cal State-Northridge. They handcuffed Mitnick and
DiCicco to a bench and called the L.A. police to come fetch them. In
spite of Kevin's cool reaction to Brown, his fear of getting caught was so
profound that it gave him heart palpitations so severe that he would soon
have to depend on heart medication usually taken by people three times
his age. Nonetheless, he seemed to realize the inevitability of arrest. He
had once told Lenny that he knew he would be caught again after the
COSMOS stunt. He had been lucky that time, getting probation instead
72   ...   CYEERPl1NK
of jail. Nevertheless, his compulsive side had won out over his fear of
the consequences, and now here he sat, handcuffed to a bench. It was a
conflict that would play itself out for years to come as Kevin's obsession
intensified. And, at his side, not so much because he sought it but rather
because he couldn't avoid it, would be Lenny, a faithful but not always
willing companion.
   The bench was in a narrow hall next to the door, twenty feet away
from a security officer working at a desk. By now Lenny was familiar with
the deep fear that came with getting busted, but as a juvenile he knew
he was better off than Kevin. When he had been caught at Cal State-
Northridge, he had been detained and released after a few hours. His
parents hadn't even punished him. Kevin, on the other hand, was now
nineteen and could get this put on his adult record. Lenny prodded
Kevin. "Hey, Kev," he said, reaching into his back pocket for his wallet.
He grinned as he produced a handcuff key. Lenny unlocked his own
handcuffs, then Kevin's. The two sat speechless for a moment. Lenny
broke the silence. "You go ahead. You've got a lot more to risk. Take
off." Both looked over at the officer, who appeared to be paying no
attention to his wards. The door next to the bench was unlocked; Kev-
in's car was in a parking lot just a few yards away.
   Lenny's mind was perhaps too full of comic strips and spy thrillers. He
had a madcap side to him that tended to surface just as a situation was
spinning out of control. His escape plan for Kevin came straight from
fiction. Lenny's plan would have brought Kevin closer to being a true
criminal than Kevin could have imagined for himself. Kevin rejected
Lenny's plan not out of reason but out of fear. He was too scared to try
to flee by himself. Kevin would try it only if Lenny came too.
    Lenny sighed and snapped his own cuffs back into place. He dropped
the key onto the floor and pushed it under the bench with his foot.
When Kevin tried to lock his own cuffs again, he couldn't. The noise
Kevin was making in his efforts to engage the hardware caught the
attention of their guard. He got up from his seat and checked Lenny's
cuffs, then turned to Kevin. As he pulled on Kevin's arm, it swung
free. "So we've got a Houdini over here," the officer remarked. He
searched both of them thoroughly and shackled Kevin back to the
bench. "Do this again and I'm going to handcuff you to the toilet," he
    When Solomon and Brown returned to the computer room to exam-
ine the terminals where Mitnick and DiCicco had been seated, they saw
that Mitnick's terminal had just logged out of a computer called Elex-

    Wash. Solomon recognized it as a Defense Department computer, but
    he couldn't tell what Mitnick had been doing with the account. Brown
    looked over the stack of printouts that had been confiscated from Mit-
    nick, and he saw that it had a list of passwords to all the local accounts
    that had been created in the last two months. The stack also contained
    account names and passwords of companies Brown knew to be affiliated
    with the Defense Department, and what appeared to be secret informa-
    tion about genetic research by a company called Intelligenetics. All of
    the computers were connected to the Arpanet, the computer network
    funded by the military that connected nonclassified military installa-
    tions, military contractors, universities and research centers around the
       At first, Lenny thought he was going to get out of this one without a
    scratch. For some reason, USC decided to drop the charges, and as
    punishment at home, his parents grounded him for a week. Then, in a
    puzzling reversal, six months later USC refiled the charges and Lenny
    was summoned to appear in court. He pled guilty to a felony as a juvenile
    and got a year's probation.
       Dominick Domino, the detective in charge of the Los Angeles Police
    Department's young computer crime unit, wanted to see Kevin somehow
    rehabilitated. In his police report, Domino wrote a brief summary of the
    case, ending it with an ominous observation: "This defendant is expert
    at computers and apparently enjoys the challenge of breaking computer
    codes. He will undoubtedly continue to be a police problem in this area
    unless maturity rechannels his energy and ambition."
       So Kevin went to jail: six months at the California Youth Authority's
    Karl Holton Training School, a juvenile prison in Stockton, California,
    with about 450 inmates. Karl Holton was one of the more secure prisons
    in the state; violence-prone teenagers who were difficult to control were
    sent there for crimes ranging from armed robbery to murder. Kevin was
    doubtless the only inmate convicted of breaking into computer systems.
    Living conditions at Holton were harsh: the prison was overcrowded,
    there was minimal personal privacy and there was a great deal of vio-
    lence. But Kevin tried to make productive use of his time there. He
    became something of a jailhouse lawyer, and he developed a computer
    program for tracking fellow wards of the court. He also worked with the
I   Los Angeles police to put together an instructional videotape on com-
    puter security. He was released in late 1983.
74 •     CYEERPUNK
Richard Cooper wanted to know why Kevin Mitnick was using the
computer so much. Why was he helping himself to Cooper's phone line
all the time? And how was it that he could magically reconnect lines
that had been disconnected?
   When Cooper presented himself at the L.A. district attorney's major
fraud section in October 1984 for a talk with investigator Bob Ewen, he
described himself as a sales consultant at Video Therapy, a curiously
named enterprise on Ventura Boulevard in Sherman Oaks. Cooper said
he was a partner with Donald Wilson, the owner of National GSC, a
merchandising company with a mystifying product line consisting of
gourmet desserts and "solar barbeque" franchises. Donald Wilson, a
friend of the Mitnick family, had agreed to hire Kevin soon after his
release by the California Youth Authority to work for Great American
Merchandising, one of several National GSC subsidiaries. These grandly
named subsidiaries, it seemed, sprang full grown from Wilson's head and
usually fizzled into bankruptcy shortly thereafter. Cooper ran one of the
ventures, called Video Therapy, but he had had about enough of Wil-
son's bizarre businesses and was unnerved by the presence of the young
man Wilson hired to handle clerical duties and work with the computer.
    Before Mitnick arrived, the only person who touched the office com-
puter was a secretary, who used it for three to four hours a day to write
letters. Mitnick was now working on the computer all day. When
Cooper came at 9:00 each morning, he usually saw Mitnick's black
Nissan, conspicuous for its vanity plate that read "X HACKER" and its
mobile radio antenna, already in the lot. And Mitnick often stayed at
least until Cooper left at 6:00 each night. When Cooper asked Wilson
why Mitnick was glued to the computer so assiduously, Wilson replied
that he was working on a number of special projects.
    But the answer didn't satisfy Cooper. Every time he passed Mitnick's
desk and looked at the screen, he saw that Mitnick was in the middle of
patching in to the computer of TRW's credit bureau to inquire into
credit standings. Cooper knew that neither Great American Merchan-
dising nor its parent company would be making such inquiries. And
there was no reason to run credit checks through TRW, as Wilson's
franchisees generally did not apply for credit.
   Then the telephone nonsense started. In early October, Cooper told
Ewen, he moved his office into the "client relations" room, where Mit-
nick and the computer sat. Every time Cooper saw the phone line for
his Video Therapy light up, Mitnick was using it.
    All day long, Cooper told Ewen, he heard Mitnick on the telephone.
He appeared to be talking with Pacific Bell employees, because he heard
Mitnick refer to such things as COSMOS, satellite operators and work
orders. He introduced himself variously as Gillie, Paul, Peter and Steve
to whomever he called. And there was one apparently Scottish person
Mitnick often made passing reference to during these telephone calls: R.
C. Mac. Ewen had to smile at that. Cooper had no idea that RC MAC
was the Recent Change Machine Administration Center, an internal
telephone company department that processed orders and changes in
service via computer.
    Coincidentally, Bob Ewen had already opened an investigation two
weeks before Cooper's visit, on an allegation that Steve Rhoades and
Kevin Mitnick were illegally getting into telephone repair and billing
computers, and fraudulently using an access code from Satellite Business
Systems, a long-distance company, to make toll calls. Mitnick and
Rhoades also appeared to be harassing people on MIT computers, where
they had guest accounts. And six months earlier, a rash of complaints
had come from the ham radio community, claiming that Mitnick had
been provoking hams over the air once again. To avoid getting sent back
to prison, Mitnick had surrendered his ham radio license to his parole
officer. Based on what Cooper was now telling him, Ewen decided there
was enough cause to issue a search warrant for the offices of National
GSC and an arrest warrant for Mitnick.
    Ewen took with him three other investigators and Terry Atchley, a
Pacific Bell security official. From the Erector-set feel of the tiny office,
Ewen got the impression that this was a fly-by-night operation at best.
Donald Wilson told the group that the reason he had hired Kevin Mit-
nick was that he felt sorry for him and he wanted to give him a break.
Wilson said he wasn't aware of any calls made to TRW, but when Ewen
questioned the secretary, she said she had received two calls for Mitnick
from a woman saying she was from TRW. When Ewen asked where
Kevin might be found, Wilson said he didn't know. Kevin had left for
lunch just fifteen minutes earlier. He added as an afterthought that he
had once overheard Kevin say he would flee to Israel before going back
to jail.
    When Ewen showed up at Shelly Jaffe's Panorama City condominium
an hour later, Shelly seemed calmer but even less cooperative than the
first time Ewen had paid her a visit three years earlier. It was as if Kevin
had coached her on what to say in case the police arrived at the door, or
so it seemed to Ewen. Kevin was living with her, she told the investiga-
 tor, but she hadn't seen him for a few days and refused even to venture
76 A     Cyg£RPUNK
a guess as to his whereabouts. So Ewen checked out some of Kevin's
usual haunts including Warner-Elektra-Atlantic Records in Burbank,
where Lenny DiCicco was working as a computer operator. Did Lenny
have any idea where Mitnick might be? Israel, perhaps? No, Lenny told
Ewen, he had heard that Mitnick might be in Las Vegas.
   Ewen's next call was to TRW Information Services. The company's
security auditor checked the credit bureau's records and confirmed that
dozens of inquiries on Mitnick and Rhoades had been made in recent
months. They had come from William Pitt Jewelers and Security Pacific
Bank. The inquiries appeared to be something of a prank. The jewelry
store alone had made 350 of them into such alleged credit applicants as
Steven Rhoades, Steve's grandmother Juanita, Kevin Mitnick, Lenny
DiCicco and Gretchen Dog-the name of Juanita Rhoades's Doberman
pinscher. Mitnick and Rhoades were merely playing around with a pur-
loined TRW account number. It wasn't surprising that Mitnick and
Rhoades hadn't altered any of the credit ratings, as that would have been
far more .difficult to do.
   During the search at Donald Wilson's office, Ewen and his men seized
a Xerox personal computer, a printer, a disk drive, a modem, a monitor
and various floppy disks. But he missed Kevin. Somehow, Kevin must
have found out he was in trouble. He was well-known for running fre-
quent warrant searches on himself.
   That might explain the strange phone call that came in to the warrant
section of the Los Angeles Police Department on October 24, the day
before Ewen served the search warrant. At 6:00 that evening, a man
who identified himself as Detective Jim Schaffer from the LAPD's West
Valley office asked if there was a probation violation warrant out on one
Kevin Mitnick. The computer operator who took the call said yes, there
was a fresh warrant in 'the computer. The detective thanked him and
gave him a number where he could be reached. About an hour later, a
second call came from a male with a different voice who also identified
himself as Detective Schaffer. He said he had Mitnick in custody and
wanted to confirm a parole warrant. Suddenly suspicious, the operator
put the second caller on hold and dialed the number he had been given
by the first caller. "West Valley detectives," came the salutation from a
woman, or at least a female-sounding voice, who confirmed Detective
Schaffer's existence. When the operator came back to the caller, there
was no one on the line. When he called the L.A. number again, an
answering machine picked up the line. "Hello, you have reached Roto-
Rooter. We're closed . . ."
   Kevin may have fancied himself an aspiring Condor, but when things
got rough, the bravado shattered. It turned out that while walking to his
car at lunchtime, Kevin saw Bob Ewen, the security investigator for
Pacific Bell and three other men headed for the elevator of his building.
Kevin called Lenny from a pay phone across the street from his office.
He was in a panic. He told Lenny he knew there was a warrant out for
him, and he was going to leave Los Angeles.
    Lenny was sure Kevin would be in touch at least once a day, as always.
But he didn't hear from his friend after he went into hiding.
    Instead, once a week or so, Kevin put in a call to Roscoe. Without
disclosing his whereabouts, he asked for gossip about the amateur radio
scene. Roscoe had long since given up his HOBO-UFO conference line.
He was now married to [o Marie, the woman whose presence in Roscoe's
life had crushed Susan's pride four years earlier. Jo Marie had completed
law school and Roscoe had gone relatively straight. He had dropped out
of USC after only a year, opting instead for a quick diploma from a local
vocational school called the Computer Learning Center of Los Angeles.
Rather than parlay his weeks of fame from the cover of the L.A. Weekly
into a security consulting job, as he had hoped, he settled for a position
as data processing manager at an auto parts importer north of Long
Beach. It wasn't exactly rarefied work, but it suited Roscoe's orderly
style. Now that he had settled down, Roscoe was beginning to think
about using his wife's legal expertise to have his 1982 conviction set
aside by the court, thus putting the whole unpleasant mess behind him.
A weekly check-in call from a fugitive was the last thing he needed.
    Kevin wouldn't tell Roscoe where he was, and with each telephone
call, Kevin seemed more paranoid. Fleeing the warrant and probation
was, perhaps; Kevin's own point of no return; he knew that what he was
doing was wrong as defined by the law. This was by no means the first
time he had been in a legal scrape, but it is unlikely that he had ever
previously imagined himself a criminal.
    Once, Kevin called Roscoe and told him he needed legal advice from
[o Marie. He had been attending a small two-year college in Northern
California, and the credits were under an assumed name. He wanted to
change his name back to Kevin Mitnick so that he could get the credits
transferred to an L.A. school.
    During his year-long absence, Kevin did leave a hint or two that he
was still in business. One day in early 1985, Roscoe came across the
phone number of Ronnie Schnell, an old bulletin board buddy from his
8BBS days, and decided to call. When Roscoe reminded him of who he
78 •     CYEERPLINK
was, Ronnie sounded surprised. "Oh, I have Kevin Mitnick on the other
line. He wants me to get him an Arpanet account." Roscoe was amused,
and he counted the seconds before his phone rang. "How did you know
I was calling Ronnie?" Kevin demanded when he called Roscoe ten
seconds later. It was a mite odd that four years had passed since either
Kevin or Roscoe had talked to Ronnie, Kevin was on the lam, deter-
mined to keep his whereabouts secret, and suddenly Roscoe happened to
be on the line at the same time. However much Roscoe tried to tell him
it was a coincidence, Kevin remained unconvinced.
   In another strange incident that seemed to suggest Kevin was still
playing around with phones, Steve Rhoades was fooling around one day
with the toll-free number for lost telephone calling cards, and when he
called the number, the person who answered with "Pacific Bell, may I
help you?" was unmistakably Mitnick. Rhoades was so amused he re-
corded the greeting and used it as the outgoing message on his answering

In the summer of 1985, Kevin resurfaced, nearly a year after his disap-
pearance and just a few weeks after his arrest warrant expired. Bob Ewen
was shocked to learn that the juvenile probation department had
dropped the warrant from its books with no explanation. If Ewen had
known that the warrant was going to dissolve, he would have insisted
on an extension. The investigator knew that if he had been working the
case steadily instead of spottily, he would have found Kevin no matter
where he had gone.
   Lenny didn't find out Kevin was back in L.A. until he got a call one
day at work. When he picked up the receiver, all he heard was someone
pushing Touch-Tone keys. The caller was spelling out his own name
using a private code that Lenny immediately recognized-the kid who
had escaped the law was back in town, possibly up to his old tricks.
Kevin finally spoke up and the second chapter of the DiCicco-Mitnick
partnership began.
   Kevin said very little to Lenny about his time spent underground.
From what he did say, Lenny figured that Kevin had cobbled together a
false ID or two, gathered his bar mitzvah money, and caught the first
plane out of L.A. But even when Lenny tried to goad him years later
into telling him where he had been, Kevin wouldn't say.
   When Kevin returned, Lenny was working on the four-to-midnight
shift as a computer operator for Hughes Aircraft's Radar Systems Group
in El Segundo. It was his fifth job in two years. Well-spoken and relaxed
during interviews, Lenny was a master at talking himself into any posi-
tion. His habit of abusing his employee privileges, however, would sur-
face soon enough, and what his family referred to as "Lenny's computer
shenanigans" usually drove his supervisors to fire him after only a few
    Whenever Lenny got a new job, Kevin wanted to know what com-
puters were there. A Digital computer was a good start; a dial-up line
that allowed access to the computer from the outside was even better.
And if Lenny was working a job with night hours, and could get Kevin
in when everyone else had left for the day, then Lenny heard from him
just about every day.
    Kevin had a new project in mind: establishing accounts on the score
of minicomputers that Pacific Bell operated in Los Angeles for order
entry. Looking back several years later, Lenny decided that when they
were dabbling in the "minis," as telephone company personnel referred
to them, they gathered more power than they had ever had or would
ever see again. Someone with access to these computers, which connect
to switches controlling all Los Angeles telephone service, could enter
commands and see them take effect immediately. The switches accessi-
ble through the minis did everything from discontinuing service on a
line to initiating a trace. It was Lenny's job to put computer equipment
at Kevin's disposal. Without a badge, Kevin couldn't come onto the
premises at Hughes. Once, however, in April of 1986, he managed to
talk the guard into letting him in. During this, Kevin's first and last visit
to Hughes, he logged on to Dockmaster, a computer run by the National
Computer Security Center, a division of the National Security Agency,
the nation's highly secretive intelligence agency. Despite its obvious
appeal to the adventuresome pair, Dockmaster didn't harbor any deep
national security secrets. It was simply NSA's public bridge to the outside
    To get onto Dockmaster, Kevin had found the name of someone
outside of the NSA with a guest account. Posing as a technician at an
NSA computer center, Kevin had telephoned the legitimate user and
said he was issuing new passwords and needed some information: name,
 telephone and current password. It was an old trick that Kevin and
Roscoe had refined together, and it usually worked like a charm.
    But Dockmaster was incidental, a diversion from the Pacific Bell
"minis" project. Since he couldn't come to Hughes, Kevin telephoned
his instructions to Lenny. He called constantly. "It's the wife," Lenny's
80   .&   CYEfRPlJNK
partner in the computer room would joke when he handed Lenny the
phone. And when they needed to devote twenty-four hours at a stretch
to their work, they checked in to one of the dozens of cut-rate motels
along Sepulveda Boulevard in the heart of the San Fernando Valley. It
was pure cyberpunk: the mile-long strip of seedy shelters was a magnet
for local prostitution and drug rings, but neither Kevin nor Lenny cared,
as long as the motel room had a telephone line that could be transformed
into a data communications center. Once that criterion was met, Lenny
preferred motels with pools in case he got the urge for exercise, and both
were partial to places next to a 7-Eleven or other suitable convenience
store for frequent junk-food runs. Kevin usually plunked down the
$19.95 for the night onto the registration desk in coins filched from
Shelly's tip money from waitressing. Once inside the room, they went
straight to work: Kevin unpacked the terminal while Lenny popped open
his tool set to rig the telephone jack so that it would accommodate their
modem. They worked through the night, taking little note of some of
the seamier activity around them. They often pushed the checkout time
to the limit, until the manager came around to throw them out, making
them personae non gratae at that establishment.
    It took six months for them to get fully privileged accounts on every
phone company mini in the Los Angeles area. And with each computer
they conquered, their potential power increased. It would not have been
difficult for Kevin and Lenny to take down the phone service for the
entire metropolitan area; but neither of them was interested in that. The
idea of power was more seductive than actually wielding it.
    When Hughes got word from the NSA that someone from the EI
Segundo facility had been into the Dockmaster computer, Hughes man-
agement went on a witch-hunt. The night of the penetration was linked
to Kevin's visit, Mitnick was linked to Lenny and Lenny was summarily
dismissed. The company's corporate security department spent the days
after Lenny's departure behind closed doors, conducting intensive inter-
views with anyone Lenny had associated with.
    After Lenny was fired from Hughes, he got a job as a delivery man for
a flower shop. The only vital credential was a perfect driving record.
Though only twenty, Lenny already had a formidable record of outstand-
ing traffic warrants. But that was Leonard Mitchell DiCicco's problem;
the florist hired Robert Andrew Bollinger, a model motorist. As Lenny
would later tell the story, with Kevin's help he had created a new iden-
tity for himself. He even went so far as to rent an apartment in the San
Fernando Valley as Mr. Bollinger.
   But a florist doesn't have much need for a VAX minicomputer, the
mainstay of Digital's product lines and Kevin's target of choice. Now
Lenny hardly heard from Kevin at all. The intensity of this friendship,
Lenny realized, waxed and waned depending on his job. The degree of
Kevin's loyalty was a function of the kind of equipment Lenny could
   Bollinger quit the flower delivery business after one month. Lenny
DiCicco resurfaced as computer operator for a shipping company. For
the time being Kevin kept a low profile. He seemed to have become
more paranoid since his return. Kevin had always been secretive about
disclosing personal data-it wasn't until months after Lenny met him
that he was even willing to give Lenny his home telephone number-
but now he was acting strangely. He was taken with the idea that people
should be able to contact him at any time, but without his having to
reveal his whereabouts, so he started carrying a pocket pager everywhere
he went. And both Lenny and Roscoe noticed that the more paranoid
Kevin became, the more he ate.
   Perhaps in an effort to follow Roscoe's path into the world of conven-
tional computing, in September of 1985 Kevin enrolled at the Computer
Learning Center of Los Angeles. Roscoe had graduated a few years earlier
and spoke well of it. Kevin's previous experience in higher education
had been dismal. In 1982 he was expelled from Pierce College, a two-
year community college in the heart of the San Fernando Valley, for
tampering with the school's computers. Maybe this time he could keep
his mind on the coursework.
   The Computer Learning Center was started in the 1960s by a group
of computer executives who, astutely enough, predicted that the nascent
computer industry would see explosive growth in years to come and
would need tens of thousands of people qualified to fill the jobs that were
emerging. By the late 1970s the computer industry was a major source of
white-collar employment, not unlike the consumer electronics industry
when it mushroomed in the 1950s and 1960s. And just like any other
technical school advertised on matchbook covers and city buses, the
Computer Learning Center wanted to attract a student body hungry for
   To gain admission to CLC, a prospective student needed a high school
diploma or equivalency degree. The applicant then took a basic aptitude
test designed to measure ability in symbolic logic and mathematical
reasoning, the foundations of computer science. The $3,000 tuition for
the nine-month program was paid in advance and the student was on
82   &   (]YE6~PUNK

the way to a fresh career. CLC graduates, many of whom were retooled
English majors or auto mechanics, joined the data processing ranks in-
side Security Pacific Bank and the dozens of giant aerospace corporations
that dotted the Los Angeles basin. Polaroid snapshots of these successful
graduates were enshrined in a glass display case upstairs at CLC. And
next to the first-floor admissions office, letters of glowing commendation
from such satisfied employers as First Interstate Bank, Continental Air-
lines and Agfa-Gevaert lined the walls.
   Although the school failed to offer its graduates a formal degree, as a
technical school it could certify would-be technicians and programmers
as proficient in using the center's computers. For the most part corporate
America didn't need computer science Ph. D.'s fresh out of UCLA, Stan-
ford or Cal Tech as much as it needed young men and women willing to
work a lobster shift, changing tapes to back up the day's business for a
starting salary of $20,000. The Computer Learning Center was a steady
source of entry-level programmers, technicians, computer operators and
data-entry clerks for the region. If not exactly the glamour posts of the
computer industry, these jobs at least paid better than other clerical
positions that required less training.
   The Computer Learning Center was a resolutely straight, no-nonsense
place. Its dress code forbade all but the most businesslike attire. Like so
many scrub-faced Mormon missionaries, the men were required to wear
jackets and ties, the women skirts or dress pants.
   Larry Gehr, who taught an introductory class at CLC in the COBOL
computer language, was vaguely aware of Kevin Mitnick's past problems,
but he tried to treat him like any other student. Kevin was inquisitive,
and his mind always seemed to be working at least a lecture's worth
ahead of the class.
   Kevin had an unusual mix of computer expertise: there were some
fundamentals he didn't know, but in other areas he was well ahead of
his fellow students. He displayed so much ability, in fact, that Gehr was
concerned that an entry-level job wouldn't challenge him. Gehr adopted
the role of teacher-counselor to this erratic student. He tried to encour-
age Kevin, telling him that if he could get a good entry-level job and
show what he could do, he would rise swiftly.

Even in the days when Susan was around, dating her way through the
phreaks, Kevin had seemed entirely oblivious to women. So it was a bit
of a surprise to friends when, in the summer of 1987, Kevin casually
mentioned that he had just gotten married. Everyone developed his own
theory about Kevin's nuptials. That the bride in question worked at
GTE, one of the two telephone companies servicing the L. A. area,
seemed no coincidence. At first, word went out that she was a program-
mer there. For someone like Kevin, his friends agreed, there was nothing
more tantalizing than the promise of an inside source. And by the time
Susan Thunder, now in the Downey area in south Los Angeles training
for a career in professional poker, found out about it, Mrs. Mitnick was
being described as a senior executive at the telephone company.
   Bonnie Vitello was slight and dark, her round face and brown eyes
framed by a thick mass of long chestnut hair. Bonnie's expression in
repose approached a frown, but the suggestion of a sour temper would
suddenly disappear with an unexpected explosion of glistening white
teeth that cast a disarming spell over strangers. Bonnie's smile was her
built-in edge.
    Born in New Jersey, Bonnie was six when her parents moved to
Monrovia, about twenty miles northeast of downtown Los Angeles.
When the Vitello family moved there in the mid-1960s, the predomi-
nantly white, mostly conservative city was on the verge of change: gang
fights between the recently arrived Hispanics and the entrenched whites
were starting to plague the community. Nine years after their arrival,
unhappy with the growing racial tension, the Vitello family of six moved
south to Rowland Heights, a quiet, arid strip of a town on the eastern
fringes of Los Angeles County. Twenty minutes north of Disneyland,
Rowland Heights is a community of new condominium subdivisions
interrupted occasionally by clusters of single-family homes-an ideal
place to start over. A listless student back in Monrovia, Bonnie suddenly
became interested in school in Rowland Heights. At her new high
school, the teachers weren't busy breaking up gang fights and she no
longer had to guard her back. She graduated early and enrolled at a small
junior college not far from home. Then interruptions set in, not the least
of which was her first marriage, at age eighteen, which lasted just six
    Bonnie's impulsive jump into marriage exacted its price. She dropped
out of college and put her academic pursuits on hold indefinitely. At
twenty-one she was doing temporary office work when she got an offer
to work full-time as a secretary for GTE at its Monrovia headquarters.
Reluctantly, she went back to the place she had left so gladly. When the
84   .&   C'!EERPLINK
phone company moved its headquarters from Santa Monica to Thousand
Oaks, a suburb on the northwestern edge of the Los Angeles area, Bon-
nie transferred there.
    Bonnie hated the monotony of clerical work. After several years of
filing and typing, with no real prospects for advancement, she put her
mind to learning about computers. It was an easy transition for her.
Personal computers had made their way onto desks throughout the com-
pany. When Bonnie was given an IBM Personal Computer, it was with
the expectation that she would use it for nothing more than word pro-
cessing. But the little machine captivated her. She pored over the man-
ual and learned the machine's every feature. Word of Bonnie's computer
prowess spread and she became the department's PC troubleshooter.
With GTE paying the way, Bonnie enrolled as an evening student at
Computer Learning Center.
    One night, Bonnie was at the main console learning how to con-
trol one of the central training computers. The student operators
sat at computer consoles, simulating the activity of a large corporate
data processing operation: computer jobs were running, the system
was monitored and data files were regularly backed up. Suddenly
Bonnie saw messages flashing on her screen from someone sitting
across the room in a spot reserved for people with full privileges on the
system. "Don't flush my file!" came the message, then, "Let my job go
    "Who is that guy?" Bonnie asked one of the instructors.
    "Oh, that's Kevin Mitnick," he replied. "Don't piss him off. If he
doesn't want you to flush his job, then don't." Bonnie looked across the
room at the overweight stranger and smiled.
    She sent him a reply. "What else do you do besides tell people not to
flush jobs?"
    "I like to go out to eat," came the response. "Would you like to go
    "I can't. I'm engaged," Bonnie replied.
    "That's too bad. You have a nice smile," he wrote back.
    Bonnie had been engaged for six months to an engineer who was a
great deal older than she, but the excitement of the relationship's early
days had gradually diminished. She considered this for some minutes,
then wrote back to him: "Well, I'm not so happy with the relationship."
    He answered within seconds: "Then maybe you'll go out with me."
    She was tempted. "Maybe next week," she typed back.
    Kevin Mitnick's electronic entreaties persisted. For several weeks after
that, each time he flashed her a message asking her to have dinner with
him, she politely declined. Late one evening, he abandoned the com-
puter messages and approached her desk, eating something. "Well, ob-
viously you're not seeing your fiance tonight," he said midchew. "Do you
like Thai food?" At that point, although Bonnie had never considered
dating anyone quite so bulky, she finally said yes.
   Over dinner, Kevin asked her where she worked. When she replied
that she worked for a phone company, he began to laugh so hard he
nearly choked. But he wouldn't explain why. Bonnie found him charm-
ing and interesting. It surprised her to discover that he was just twenty-
three, six years her junior; and he professed to be equally surprised that
she wasn't his age, as he had at first assumed. He seemed more mature,
particularly in comparison with some of the other, younger CLC students
who were just out of high school. What was more, he could explain
difficult computer and mathematical concepts as no one else seemed able
to do. He wasn't boastful, and yet it was clear that his knowledge of
computers ran deep.
    They began to see each other often, sometimes taking a bottle of wine
to the beach in Santa Monica on a Friday evening. Not much of a
drinker, Kevin gamely swirled his wine in his glass as they talked. And
although Kevin had never done much dancing, Bonnie managed to coax
him into local discos. When Kevin was around Bonnie, in the early
stages of their relationship at least, another side of him emerged. In
contrast to the young man who reacted against a difficult and isolated
childhood by lashing out at hams, monkeying with people's phone ser-
vice or reaching for a computer-the one thing that gave him a feeling
of control and power-around Bonnie he was easygoing and engaging.
Love had its salutary effects as well: Kevin began to drop his excess
pounds like so much ballast, and Bonnie watched his body transform
 into a more appealing shape.
    Within a few weeks, Bonnie officially dissolved her engagement, and
 Kevin all but moved into her apartment in Thousand Oaks. It was a
small one-bedroom place, but there was enough space for the two of
 them. Before long, marriage became Kevin's obsession. Every week or
so, he asked her to marry him. Something about it felt right to Bonnie.
She was ready to accept. When Bonnie joked to Kevin that she was his
shiksa-the Yiddish word for "non-Jewish woman"-he asked her what
 the word meant.
86   ..   CYEERPlJNK
Steph Marr had been around computers long enough to know that a
good system administrator knows how to read his machine. And some-
thing was wrong with the system at Santa Cruz Operation.
    A blind church organist can tell how many people are in the church
by the way his music sounds, but he probably can't explain how he does
it. A good jockey can tell what kind of mood his horse is in, and can
often tell how well the horse is going to run by the way it walks to the
starting line. A good computer system administrator develops a feel for
the patterns and sounds of his computer. When they are aberrant he
always investigates. Computers are supposed to do the same thing over
and over again, and when they do something different there is always a
reason. The internal rhythm of a computer is seen in the delays in
getting a response to typed keys, the staccato sound of a disk arm rat-
tling, the flickering lights of modems and disk controllers, or the daily
routine of log and journal file entries.
    Named for the Northern California coastal town where the company
had its headquarters, Santa Cruz Operation got its start selling a version
of the UNIX operating system to run on personal computers. In eight
years, Santa Cruz Operation had grown from a father-son consulting
start-up to a multimillion-dollar company.
    Marr was one of the people who worked to keep Santa Cruz Opera-
tion's network of computers up and running. He had been there for a
year, long enough to know that certain users not only had certain privi-
leges on the system but also had individual habits. Engineers logged on
from their homes late at night; secretaries logged on only from work and
only during working hours.
    Steph was sufficiently tuned in to his computer's own circadian
rhythms-the times of heavy use and the periodic lulls-to notice an
aberration in late May of 1987. Not only did he feel the trouble, but the
system was telling him in no uncertain terms that something was going
awry: one of the secretaries who used the computer was acting out of
character. She was logging in after hours, cruising the system and trying
to peek into other people's directories. She was accessing files that had
been dormant for many months, including an out-of-date "help" system.
When Steph asked her about it, she professed ignorance. Apparently
someone had ferreted out her password and was blundering around inside
SCO's computers.
    When Steph realized that an intruder had entered the system, instead
of immediately trying to throw him out, he set up an alarm system, and
put limits on what the stranger could do. The trespasser must have felt
the presence of someone electronically peering over his shoulder, watch-
ing his every move, because a few days after the monitoring started, he
typed onto the screen, "Why are you watching me?"
   "Because it's my job," Steph typed back.
   Once he had engaged the company in an electronic conversation, the
hacker made a request that signified so much audacity that the system
administrator knew this was no ordinary intruder. The uninvited guest
told Steph that he wanted an account that gave him unlimited privileges
throughout the system. By this time, a group of onlookers had formed a
half-moon around Steph's desk, and a lively debate ensued over whether
or not Steph should give him an account on the company computer.
   Steph believed his own experience as a hacker on the edge of the law
gave him a better understanding of just what motivated this person. In
his younger days, Steph had done his share of system cracking. As a
college student, his favorite pastime was to slide past the security on
university computers, then call the system administrators to inform them
of what he had done before they discovered it. He, too, believed in free
access to information. But Steph also knew when to stop. He believed
in exercising restraint and respecting someone else's right to run a com-
puter system without having to build elaborate safeguards against out-
siders. There was something about this person's attitude that made Steph
sense that he knew he was breaking the law, and was perhaps even proud
of his trespass.
   A high-level account was out of the question. If Steph gave him such
power over other users, and if he was clever enough, the intruder would
be able to change the machine's operating system, or even do something
as patently malicious as depriving others of the use of the computer. If
he wanted to, he could become the electronic equivalent of a mad
gunman in a bank, holding the computer hostage. Finally, Steph decided
to give him an account that masqueraded as something more powerful
than it actually was. The account was called Hacker, a label of the
intruder's own choosing. In setting up an account for him, to which
Steph also had full access, Steph figured he was erecting an aquarium,
in which he could watch every move of this felonious fish. Moreover,
letting him lounge around on the computers would make it easier to
trace his calls. Within forty-eight hours, Pacific Bell security was doing
just that. But when Pacific Bell security investigators started to trace one
call, they discovered that somehow the trespasser was wily enough to get
access to the phone company computers themselves and block certain
commands requesting information about the line he was using.
88   l.   CygfRPl1NI<
   Despite the long hours the intruder was dedicating to wandering
through the SCO computer, he appeared to have no discernible quest.
He seemed merely to like to prowl through the system, checking out
directories but seldom opening files themselves.
   But after a week or so, the hacker's probing seemed more directed.
He fastened onto programs that would allow someone to modify the
operating system, but he didn't have the necessary privileges to do so. It
was clear after a while that he seemed to have a peculiar goal of modify-
ing XENIX, a derivative of the UNIX operating system and the crux of
the company's business. More disturbing was his apparent goal of trans-
mitting a copy of XENIX to his own computer.
   But the owner of the Hacker account gave himself away. The simple
oversight that exposed him lay in his use of MCI to gain access. A special
MCI feature automatically identified the telephone from which the tres-
passer's call to the local MCI access number was placed. It was a tele-
phone in Thousand Oaks, California.

When investigators from the Santa Cruz Police Department flew to Los
Angeles on the morning of June 1, 1987, to conduct a search at apart-
ment number 404, 1387 East Hillcrest Drive in Thousand Oaks, years of
experience investigating fraud couldn't have prepared them for the odd
collection of evidence they encountered. The officers had been told what
to look for-computer printouts, a computer, a modem and notes of
phone numbers and access codes-but they weren't sure why. When the
officers knocked on the front door there was no response, so they let
themselves in with a pass key borrowed from the manager of the com-
plex. Detective Patricia Reedy of the Santa Cruz Police dutifully noted
in her report: "On the dining room table was a computer with a black
box with multiple small red lights on it. Detective Nagel advised the
black box was a modum [sic]. The 'MC' light on the modum was lit.
Lying next to the computer was a beige touchtone phone with the front
plate removed. The computer, the modum and the phone were all
hooked together." To prevent anyone from calling the apartment and
erasing possible evidence from the computer, Detective Reedy followed
what she must have thought was the proper procedure: she removed the
receiver from the phone. As she noted in her report, the receiver then
"made a loud screeching noise." Any computer crime investigator would
have warned her to keep the phone in place, as the suspect might well
have been in the process of transmitting data.
    As they made their way through the small dwelling into the bedroom,
they first encountered large piles of clothing on the floor. On a bedside
table they counted fifty-five computer disks, in boxes and scattered on
the tabletop. From underneath the bed they retrieved "a book marked
'OS Utilities,' numerous loose sheets of printed materials on computers
and a plastic bag containing a large quantity of hand-written notes and
computer printouts." They gathered all of it as evidence. Also under the
bed was a loaded Charter Arms. 38 special two-inch revolver. Inside the
bedroom closet the detective found a second, much larger firearm-a
Remington .87 shotgun. On the bedroom floor they found two small
plastic bags containing what appeared to be marijuana, and a glass bong.
They took the items as evidence. When Detective Reedy checked the
clothing in the closet, she found $3, 000 in $100 bills in the pocket of a
man's suit jacket. She initialed each of the bills and left them on the
kitchen table.
    The detectives knew they didn't understand enough about computers
to analyze what they had found and decided to call in a computer spe-
cialist from the Ventura County sheriff's office. "The specialist advised
us that what we had in front of us was a computer terminal that had no
way of storing information. He explained the black box on top of this
terminal was in fact a modum. He was able to put the receiver back on
the phone . . . and access into the terminal. He was able to bring up on
the screen information that said 'abort.' He advised we probably inter-
rupted the computer accessing into somewhere else."
    The search ended at 4:00 P.M.
    When Bonnie came home that afternoon, she thought they had been
burglarized. The computer terminal was gone, as well as the modem, the
floppy disks and all of their computer books. But when she saw a stack
of $100 bills on the table, money Kevin had been saving for their wed-
ding celebration, next to a document bearing an official seal, she knew
 they had been searched. She packed a bag and went out to find Kevin.
When she told him who had just been to the apartment, he flew into a
    One of Kevin's first calls was to Roscoe, demanding to know whether
Roscoe had informed on him. Roscoe tried to calm Kevin down and ask
him questions. But Kevin was too excited to listen; he began to sound as
 if he were talking to himself. Should he get out of town? Without
 waiting for an answer, he kept talking. Then again, he rambled on, he
 didn't want to leave Bonnie. Should she go too? Roscoe cut him off and
 told him to get in touch with an attorney.
90   A   CygfRPUNK
   The officers returned to the apartment at 8:30 the next morning.
They knocked on the door and got no answer, so they went to interview
the manager, Alice Landry, who told them that the tenant in number
404 was Bonnie Vitello, a nice-looking woman in her twenties with a
thin build and dark hair who worked at GTE. She said Vitello had had
"her brother" staying with her. He was tall, heavyset, nice-looking and
clean-shaven. Overall, the manager told the inquiring officers, Ms. Vi-
tello had been a good tenant, although there had been a couple of
complaints about noise from inside the apartment in the evening, the
sound of people arguing. The brother, she said, was at home a lot.
   Later that morning, the officers called GTE's security manager, who
verified that Vitello had worked there for several years. That morning
she had called in and asked for vacation time. She had said she was
moving and would be back the following Monday. Detective Reedy
asked the security manager if he recognized the name Kevin Mitnick.
Not only did the manager recognize the name, but he rattled off a list of
agencies he knew had investigated Mitnick in the past. Mitnick, it
seemed, was something of a household name around GTE.
   Reedy's next call was to the L.A. County district attorney's office,
where she was directed to Bob Ewen. Ewen told Reedy that Mitnick had
been a suspect in several major computer crimes in Southern California,
and that he himself had worked a case involving Mitnick several years
before. At one time, Ewen said, there had been a warrant for Mitnick
in the office's computer, that Mitnick had found out about it and that
he had then fled to Israel. There were no current warrants out for Mit-
nick, but Ewen believed the FBI was working some cases involving him.
In any case, Ewen warned, Mitnick was extremely dangerous, capable of
destroying computer systems remotely using "logic bombs."
   Logic bombs? Israel? Southern California was bizarre enough, but this
sounded absurd. That afternoon, the officers called Santa Cruz Opera-
tion to say they believed they had found the intruder. The arrest war-
rants that were issued from the Santa Cruz County Court charged both
Kevin David Mitnick and Bonnie Lynne Vitello with unauthorized ac-
cess to a computer, a felony under California law, with bail set at $5,000
each. The officers who had conducted the search described Vitello to
the court as "dangerous and bright." And twenty-three-year-old Mitnick
was a known criminal with a long list of previous offenses. Three days
after the warrant was issued, the suspects surrendered voluntarily at the
West Hollywood police station. Once it was established that Mitnick
had acted alone, Santa Cruz Operation dropped the charges against
   For all the trouble he had been in through the years, Kevin still had
a clean adult record and he wasn't about to let that change; he refused
to plead guilty to a felony charge. His attorney asked that, in exchange
for Kevin's full cooperation in explaining how he had cracked the Santa
Cruz system, the charge be reduced to a misdemeanor. So a misdemeanor
it was, with a small fine, thirty-six months' probation and a three-hour
meeting between the Santa Cruz Operation computer staff and Kevin
Mitnick in the presence of their respective attorneys.
   When Steph Marr, the system administrator at Santa Cruz Operation,
met Mitnick for the first time, he thought he should offer some words of
praise. After all, Mitnick had managed to muscle his way into the Santa
Cruz Operation computers and, for a time at least, had eluded detection.
"Well played, well met," Marr said as he greeted him. But Mitnick
barely responded. When Marr asked him a technical question, he re-
sponded to the Santa Cruz Operation attorney instead. And his attitude
as he described his methods was annoyingly condescending. It was hardly
the hacker-to-hacker session Marr had hoped for. When Marr got a call
a year later from Mitnick asking about a job, it only confirmed that this
was indeed a young man with extraordinary gall.
   Kevin and Bonnie were married that summer, while the Santa Cruz
charge was still hanging over Kevin's head. It was a trying time, espe-
cially the frequent trips to Santa Cruz for court appearances. It wasn't
exactly the way Bonnie might have chosen to spend her thirtieth birth-
day, which arrived just a month after their apartment had been ran-
sacked by the police. But for all the problems Kevin was causing her,
Bonnie still wanted to marry him. She had already been through one big
Catholic wedding and she didn't want another extravagant affair, and
Kevin didn't seem to mind one way or the other, so they went to City
Hall and emerged fifteen minutes later with their vows in place. To
appease Bonnie's mother, the newlyweds dressed in wedding garb and
went to her house for a celebration party. The couple stood smiling for
the obligatory photographs. Bonnie was happy. Kevin had given her a
ring and his heartfelt word that his computer shenanigans were forever
behind him.
   If Kevin was planning to make a clean break with his past, it didn't
help that he became the primary subject of a Pacific Bell security memo
a month or so later. A ham with whom Kevin had a less than friendly
92   A   CygERPUNK
relationship happened to work at the telephone company, and read the
memorandum over the air. Written by a security manager, the memo
detailed the contents of the computer disks found during the search after
the Santa Cruz break-in. Describing the events surrounding Kevin's case
as "alarming," the manager listed what had been found in the Thousand
Oaks apartment: "The commands for testing and seizing trunk testing
lines and channels; ... the commands and logins for COSMOS wire
centers for Northern and Southern California; ... the commands for
line monitoring and the seizure of dial tone; . . . references to the im-
personation of Southern California security agents ... to obtain infor-
mation. . . ." The list went on.
   The memo concluded that computer hackers were becoming more
sophisticated in their attacks on phone company computers. The author
suggested that it was possible that hackers could incapacitate an entire
central office switch by overloading it or tampering with the computers
that controlled it. The memo also voiced concerns that terrorists or
organized crime groups might get their hands on "underground computer
technology. "
   When Kevin heard this, he panicked. He had to see the memo. He
called Lenny, then Roscoe, who may have stopped his extralegal activi-
ties but didn't mind being pulled in on the occasional clever hack. The
three worked out a plan for getting the memo. Kevin called the secretary
in the San Francisco office of the manager who had written the memo
and, posing as another security manager, told her he had never received
his copy of the security memo. Would she mind faxing it to him? Of
course not, she replied, she would be glad to; she even had the number
programmed into her fax machine's speed dialer. When she pushed the
button to send the memo on its way, she had no way of knowing that
Kevin had programmed the number her fax machine was calling to
forward the memo to Roscoe's fax machine at work. Roscoe had repro-
grammed his fax machine so that when it responded to the secretary's
machine, it looked as if the proper fax machine was responding. Even
after Roscoe received the memo and read it aloud to Kevin, Kevin
wanted to see it immediately. He didn't want to wait and pick it up after
work. So he had Roscoe fax it to him at a copy shop in the San Fernando
    Roscoe later leaked a copy of the memo to reporters and a story
describing its contents landed on the front page of The New York Times.
He couldn't resist embellishing slightly the story of how the memo was
obtained. He told the reporters that it had been intercepted by tapping
a phone line between two Pacific Bell fax machines. Company officials
confirmed the memo's authenticity, but said they were mystified by how
it had landed in the hackers' hands.
                                  ... T ...
The first call from Pierce College came in to the Los Angeles Police
Department on the afternoon of February 17, 1988. A Pierce College
security officer was on the phone to report that since January 13 two
young men had apparently been making illegal copies of software. On
that January evening, Pete Schleppenbach, a computer science instruc-
tor, had walked into the computer science lab and seen a tall, slender
man in his late teens or early twenties hovering studiously over one of
the system terminals. Schleppenbach was taken aback by the sight of a
complete stranger standing so authoritatively in a place clearly off limits.
"No Admittance-Authorized Employees Only" warned a sign sus-
pended directly above the stranger's head. And one would have to be
unable to read English to be oblivious to the warning notes Schleppen-
bach had taped to the terminal, obscuring the screen: "Do not turn this
terminal off. Leave it on!" and "Students: Don't use this terminal unless
all others are in use!" To see what was on the screen, the young man
had taken Schleppenbach's handcrafted monitions and flipped them
onto the top of the monitor. Schleppenbach stood and watched the
young man-a student? a Digital technician?-as he reached behind
the terminal and turned it off, then on. Then the stranger sat in front of
the machine as if he owned it and began to type.
   Schleppenbach approached him. "What are you doing?" he asked.
   The young man barely looked up. "Just looking," he mumbled. His
speech had the tone of someone convinced that his actions were beyond
   "What are you doing on the terminal? Didn't you see the signs? You're
not a student here, are you?"
   He shrugged at Schleppenbach. "No, I'm not, but she said it was all
right," and, without turning around, he thrust his shoulder in the direc-
tion of a student employee working in the computer lab. Unsatisfied,
Schleppenbach told the stranger to leave. The teacher then asked the
student worker if she had given the stranger permission to use the ter-
minal. She said she hadn't.
   When he turned around, Schleppenbach saw that the trespasser had
not left the room, but was seated at another terminal with someone who
appeared to be his friend, a pudgy complement to the rangy, cocksure
94   A   CYE£RPUNK
stranger. Might he be a Digital technician? He looked old enough to be
out in the world of gainful employ. The friend seemed engrossed in what
he was seeing on the computer screen, and he typed in short, feverish
bursts. As Schleppenbach began to approach the two, he saw the slender
one tap the shoulder of his friend, who turned around to face Schleppen-
bach. He turned back to the computer and, as if in a big rush, typed
something quickly and got up to meet Schleppenbach halfway across the
room. The pudgy stranger spoke first. Friendly and inquisitive, he told
Schleppenbach that he and his friend were there to find out about the
course Schleppenbach was teaching in office computer networks. Cau-
tious despite the young man's friendly, almost engaging demeanor,
Schleppenbach explained the course's prerequisites, and told them how
to register for classes. He told them they shouldn't be using the system
unless they were enrolled at the school. "Okay, we were just leaving,"
replied the heavy one, at once defensive and vaguely arrogant. "We just
know a little about Digital computers."
   When they were gone, Schleppenbach walked over to the computer
where the two had been sitting. He saw that a tape was inside the tape
drive, and that a light on the drive was flickering to indicate that some-
thing was being written on the tape. The only person aside from Schlep-
penbach who was authorized to mount tapes for backup was another
instructor. Schleppenbach went straight to the classroom next door,
where the other teacher was holding class, and asked his colleague if
he had been doing work on the Micro VAX II, a small Digital Equip-
ment computer. The other instructor shook his head. Schleppenbach
rushed back into the lab and saw that the light next to the tape drive
was still on.
   Schleppenbach enlisted the help of one of his students to try to figure
out what was going on. They went to a terminal and typed, "show
system"; their screen displayed the name of every job running on the
computer. When Schleppenbach saw a program called CP.COM, he
began to get nervous. Schleppenbach knew that program hadn't been
there before, and when he displayed it on the screen, he saw it was a
simple seven-line program, a command procedure for making a complete
tape backup of every program on the college's system. At first, Schlep-
penbach decided the only thing to do was to abort the work of the tape
whirring inside the tape drive, but he thought better of it. The backup
process lasted forty minutes. When Schleppenbach removed the reel of
tape from the drive, he saw right away that it did not belong to Pierce
College. When he reloaded the strange tape into the tape drive, he
asked the computer for a listing of its contents. The tape contained a
copy of every file in the system. If he hadn't intercepted those two young
men in the middle of their task, they would have walked out of the room
with a copy of software worth $20,000. It wouldn't have deprived the
college of the software, but in Schleppenbach's view it was theft none-
theless. Schleppenbach figured he had seen the last of them, but for
safekeeping he put the tape in a locked room accessible only to faculty.
The next morning, he called the chairman of the computer science
department to tell him what had happened.
   When Schleppenbach walked into the lecture room at 7:00 P.M. on
February 9 for the first class of Computer Science 64, he was shocked to
see the same two youths seated in the back of the classroom, grinning
and tapping pens against their desks. They had enrolled.
   Two days later, Schleppenbach ran into Anne Delaney, the former
computer science chairwoman, now a professor in the department. He
had hardly finished telling her of the incident with the tape drive when
she interrupted him. "It isn't Kevin Mitnick, is it?" He looked down at
his class roster. Yes. Kevin Mitnick was one. The other was Lenny
DiCicco. Delaney looked stricken. She told him that in 1982 Mitnick
had been expelled from the school's computer science program for tam-
pering with the school's computers. He was trouble. She told Schleppen-
bach to alert everyone that Kevin Mitnick was back on campus.

This was not the first that Jim Black had heard of Kevin Mitnick. Tall
and wiry with an appeal suggestive of Montgomery Clift, the forty-seven-
year-old LAPD computer crime detective had loosely kept track of Mit-
nick for years. When the call came from Pierce College, Black's hunch
was that this was more than a simple matter of copying some software
from the computers of a junior college. He had heard enough about
Mitnick through the years to suspect that he might be up to something
big. This time, he wanted to see Mitnick and his friend spend some real
time in prison. Black had heard that Mitnick didn't like jail one bit.
The detective suspected there wasn't much that would disrupt the young
man's impulse to seek control of electronic devices other than a healthy
dispensation of justice. He dropped everything to work full-time on the
   Black had begun to specialize in computer crime in 1982, when two
embittered employees of Collins Food, a restaurant chain, were accused
of planting two "logic bombs" in the company's computers. The insidi-
96   •   Cyg£RPUNK
ous software was designed to destroy payroll, inventory and sales records,
and it was only a lucky accident that led an employee to discover the
potent little pieces of code before they did their damage. Black had
begun to handle other aspects of fraud eight years earlier. He was work-
ing auto repair fraud when he was asked if he'd like to join in the Collins
Food case investigation. He spent several years on the case, and although
the prosecution couldn't gather enough evidence to convict the suspects,
the challenge of working a case in which someone had the ability to do
something so destructive without leaving a trace was such an interesting
departure from routine investigative work that he moved into the de-
partment's fledgling computer crime unit. By 1988, Black's division was
one of a dozen or so units around the country devoted to computer fraud.
   Black saw people like Mitnick not just as a general threat to computer
systems, but as a personal threat. Mitnick was known to take direct
revenge on members of the law-enforcement community. Black spoke
with one of Mitnick's former probation officers, who said the telephone
service at her home had simply gone dead one day, but when she called
Pacific Bell to report the problem, the company told her that according
to its computer her service was just fine. It had taken her days to con-
vince the phone company that her line was really dead, and still longer
to get it fixed. Black didn't want to take any chances. He made special
arrangements with TRW so that anyone attempting to see his credit
rating would have to go through extra steps. The telephone company
made similar provisions for Black's home telephone service.
   The day after the initial call, Black was at Pierce, talking to the staff
and administration. That afternoon, he ran a search on both Mitnick
and DiCicco. DiCicco's record showed several outstanding traffic war-
rants. Curiously enough, Mitnick came up clean. Black put in a call to
Bob Ewen in the DA's office, and Ewen told him about the Santa Cruz
case. When Black called the Santa Cruz Police Department, he learned
that Mitnick had pled guilty to a misdemeanor eight months earlier, and
was currently on probation.
   Black pulled driver's licenses for Mitnick and DiCicco, and called the
letter carrier to see who received mail at 8933 Willis #13 in Panorama
City. The mailman said that Mitnick got mail there. So did Bonnie
Vitello, Mitnick's wife, and Shelly Jaffe, Mitnick's mother. Black's next
call was to the local FBI office. An agent there said he had gotten a call
from the FBI office in Baltimore a few months earlier, linking Mitnick
to the penetration of a National Security Agency computer from the
Hughes Radar Systems Group in EI Segundo, but as far as he knew the
· L.A. office had no open case on Mitnick or DiCicco. A few days later,
  Black and a deputy district attorney met with Schleppenbach at Pierce.
  They told Schleppenbach that before and after each class he should
  watch both suspects and jot down anything that looked suspicious.
  Black's next step was to call the local Digital office and describe the
  problems at Pierce. An engineer from Digital's Los Angeles office took
  on the job of analyzing the tape DiCicco had left in the college's com-
  puter and monitoring the duo's computer activity at the school.
     On March 3, surveillance began. Black wanted their every movement
  accounted for. First, the campus police took up the watch. At 6:30 that
  evening, plainclothes officer Kenneth Kurtz watched Lenny DiCicco
  arrive at the computer science lab, sit down at a computer and begin to
  type. Thirty minutes later, Mitnick arrived. Once Schleppenbach ar-
  rived, the students in the lab moved to the lecture room; DiCicco spot-
  ted Mitnick and sat next to him. For the next hour, the two students
  watched the instructor, occasionally leaning over to whisper to one
  another. At 8:00 P.M., both Mitnick and DiCicco bolted from their
  chairs and beat the rest of the students to the adjacent lab. The officer
  sat nearby and watched as Mitnick began helping his stumped peers with
  their assignments. DiCicco got into a conversation with Schleppenbach
  and, by way of explaining his thorough knowledge of VMS, the operat-
  ing system for Digital's VAX computers, told him that he worked at
  TRW on a VMS system. Kurtz left a few minutes before class was to be
  dismissed and waited outside the building for Mitnick and DiCicco to
  appear. At 9:45 P.M., Mitnick came out and began to walk the perimeter
  of the computer science building. Kurtz climbed onto a roof and watched
  as Mitnick reentered the building. About three minutes later, Mitnick
  and DiCicco emerged together. Kurtz hopped from roof to roof of adjoin-
  ing buildings. He scrambled down in time to see DiCicco get into his
  small brown Toyota and drive away. Mitnick drove in his car from a
  separate exit.
     Black and his partner picked up the surveillance from there. They
  followed Mitnick's black N issan Pulsar as it traveled a seven-mile stretch
  of the western edge of the San Fernando Valley. By the time Mitnick
  reached a smaller, winding road in Calabasas, the detectives noticed
  that he seemed to be following another car. Both cars turned into a
  parking structure beneath a Home Federal Savings and Loan building.
  The cars parked out of sight, but Black saw Mitnick walk toward the
  front. He looked up and saw a light go on in a second-story office. A
  man with dark hair stood in front of the window. The light went out.
98 •     CygERPUNK
The officers left just before midnight, making a quick swing through the
parking area. The Toyota and the Nissan were the only cars there. For
the next month, Black and his partner continued the surveillance. On
occasion the two detectives had help from the department's Special
Investigative Section, or SIS, which carried out most of the sophisti-
cated surveillance needed by the L.A. Police Department. Officers in
SIS usually busied themselves with violent criminals, and although these
two suspects posed no physical threat, SIS was needed to carry out the
fine art of successful tailing. With ten to twelve men, two to each
unmarked car-Camaros, Volkswagen bugs, and pickup trucks-the of-
ficers carried out a tag-team approach to its surveillance. As one' car
peeled off, another picked up, sometimes accelerating to one hundred
miles per hour to catch up to the suspects, but always staying at least a
couple of blocks behind them.
   The two suspected software pirates stuck to something of a routine.
After leaving the campus on Tuesday and Wednesday nights at about
10:00 P.M., they would drive to the Calabasas office building. They
frequently stopped along the way at a Fatburger, an L.A. fast-food
chain famous for its tacky decor and gigantic burgers. This Fatburger
outlet was squeezed into an L-shaped pink-stucco minimall between a
7-Eleven and a taco restaurant. The two would emerge from Fat-
burger laden with bags. Judging from the sheer volume of food the two
carried upstairs, Black figured they were packing in for a long night.
   It would have come as a surprise to the two police officers to learn
that Mitnick was supposed to be on a strict diet. As far as his wife,
Bonnie, knew, her husband subsisted on a low-calorie menu consisting
of oatmeal for breakfast, two ounces of turkey for lunch, and a salad for
dinner. It was a sensible diet for someone with heart palpitations who
frequently checked himself into nearby emergency rooms with chest
pains and kept a drawer at home filled with medical insurance claims.
Not surprisingly, the regular pit stops at Fatburger had been the downfall
of many Mitnick diet plans. Computers and eating went together for
Kevin, which was one reason he never managed to fall below 240 pounds
for very long.
   The two police officers got permission to use a residential driveway
across the street. They tucked their car as far back in the driveway as
they could while still keeping a clear view of the insurance office. Occa-
sionally, one of the suspects would venture outside, look to his left and
right as if preparing to cross the street, then slip back inside. The vigils
livened up on the occasions that Mitnick came outside, crossed the
street, walked a few paces to the Hotel Country Inn and began using the
pay phone in the lobby. Mitnick's pay phone sessions would last about
twenty minutes. Whatever DiCicco was doing inside seemed to be re-
lated to Mitnick's time on the pay phone: DiCicco would walk to the
landing outside the office and peer down the street toward the hotel.
   One of the more difficult things to establish was just what office door
of the Calabasas building the two suspects were entering. The manager
of the Home Federal Savings and Loan branch told Black that neither
Mitnick nor DiCicco worked at the bank. There were at least a half
dozen other businesses that rented space there. Two members of the
surveillance team were dispatched to the roof of the building, where they
lay until they figured out that it was the door to suite 101, the offices of
a company called VPA, which stood for Voluntary Plan Administrators.
   When Black did some research into VPA, he found that it was a
company that administered disability programs for larger companies. The
company used a MicroVAX computer made by Digital Equipment. But
notifying VPA that it might have computer criminals in its midst was
out of the question. Black didn't know if the company was involved with
whatever Mitnick and DiCicco were doing.
   On March 17, when Black's partner got to work, he received a frantic
phone call from Pete Schleppenbach, the Pierce teacher, who wanted
to report a bizarre and annoying incident. Five hours earlier, at 3:00
A. M., Schleppenbach had been awakened by a telephone call from a
man identifying himself as Bob Bright, an officer with campus security.
The officer told the bleary-eyed Schleppenbach that he had just appre-
hended two male suspects who had been caught wheeling heavy equip-
ment out of the computer science lab. He described one as tall and
slender, the other as shorter and fat. "That's Mitnick and DiCicco,"
Schleppenbach said. "You've really hit the jackpot with these guys."
Schleppenbach said he would come right over. The officer told Schlep-
penbach that that was a good idea, and that he had just put in a similar
call to two other Pierce officials who were also on their way. When
Schleppenbach and two other Pierce faculty members arrived at the
campus security office, there was no one in sight but one lonely security
guard, who said it had been an uneventful evening. There had been no
burglary and he had no colleague named Bob Bright. It was a mean and
childish trick, but for Lenny and Kevin it had served a useful purpose:
now they knew they were being investigated.
   There wasn't much the police could do except log a mischievous
phone call. It was clear that Mitnick and DiCicco weren't going to make
100   A   CYC£RPUNK
this easy. Black and his partner gathered as much background as they
could on the two pranksters. DiCicco, at least, appeared to have a full-
time job at VPA, as his car would usually stay parked in the garage
beneath the building all day. Mitnick, it appeared, was unemployed.
   In March, Black got a disturbing tip. Someone from inside the police
department had heard that Kevin Mitnick was getting ready to start a
job in computer security at Security Pacific Bank.

For anyone with Kevin's record, the Security Pacific job should have
been out of the question. When he applied in early March for the
opening as an electronic funds transfer consultant in the bank's audit
department, he knew the chances of getting it were remote. His reputa-
tion as a computer criminal was more difficult to shake loose than he
had thought it would be. A few months earlier, he had taken a job at
GTE as a COBOL programmer. He had worked there for only a week
before the security department checked his background. A security offi-
cer then approached him one day, escorted him to his car and waited to
see that he left the premises. It was a humiliating episode. Most of his
other jobs had opened for him through family connections. Had the
bank managers at Security Pacific known that a computer criminal would
be in constant contact with the computer system that executed, moni-
tored and logged hundreds of millions of dollars of transactions every
day, they would have been aghast.
   Kevin diligently completed the job application. For the standard
query, "Have you every been convicted of or are you pending trial for a
criminal offense?" he placed a small, bold check next to "No." An
unusual collection of references included Donald Wilson, his former
employer at National GSC; Arnold Fromin, owner of Fromin's Delica-
tessen and his mother's boyfriend; and Roscoe. Then, in a surprising
concession to the cloak of secrecy he had maintained for the past three
years about his time as a fugitive, Kevin tipped his hand. Perhaps to
boost his otherwise anemic educational background, he noted on the
application that in the winter and spring of 1985, just when the police
had him on a kibbutz in Israel, he was a student at Butte College, a two-
year community college in Oroville, a small town in Northern California
about 250 miles northeast of San Francisco. Kevin had apparently re-
mained anonymous by enrolling under an assumed name.
   Kevin told Lenny that if he got the job at Security Pacific, he would
stop breaking into computers for good. And it seemed that he might get
the job. With a little embellishment here and there, on paper Kevin
didn't look at all bad. He listed his job at National GSC, where he had
spent so much time on the telephone, as programmer/analyst. At Fro-
min's Delicatessen, where his primary job was that of a delivery man
with a little computer work on the side, he was also a programmer/
analyst. Both businesses were owned by friends of the family who could
be expected to give good references.
   When he received the letter from the bank's personnel office confirm-
ing an annual salary of $34,000, Kevin was ecstatic. His first day of
work, the letter informed him, would be March 25, when he was to
report to the training center downtown for an orientation. The last
paragraph of the letter was boilerplate: "As discussed, your employment
is contingent upon satisfactory reference checks." When Bonnie got
home from work, a smiling Kevin told her he had been offered the job.
They went out to dinner to celebrate.
   Black got the tip about Mitnick's new job on March 23. He called the
bank's security department immediately. The next day, Black got a call
from Peter Kiefer, a Security Pacific vice-president. Yes, he told Black,
the bank had just offered Mitnick a job in its electronic funds transfer
section. Mitnick was to start work the following day.
   Late that afternoon, Kiefer and a colleague named Barry Himel ar-
rived at police headquarters to talk with Black. They presented Black
with Mitnick's application and his list of references. They also showed
him an article that had been clipped from the Los Angeles Times, dating
back seven years to 1981 and telling of three young men, a Kevin Mit-
nick among them, who had been arrested for stealing manuals from
Pacific Bell's downtown computer center in order to wreak havoc on the
telephone company and its computer systems. The story had been
brought to their attention by a bank employee who knew of Mitnick
from his involvement in amateur radio. The worried executives asked
Black if this happened to be the same Kevin Mitnick who was about to
start work as a security consultant at Security Pacific. By that point,
Black's confirmation was hardly necessary. The two security officers had
already established a strong link: as one of his references, Kevin Mitnick
had listed the true name of Roscoe, one of the three arrested in 1981 for
breaking into the Pacific Bell COSMOS center.
   It was an awkward situation, to be sure. A decade earlier, Security
Pacific had been the victim of a historic heist. The thief, Stanley Rifkin,
a plump and soft-spoken thirty-two-year-old computer expert, had
worked there as a computer security consultant and walked away one
102 ...   CYEERPliNK
afternoon with the day's code for the bank's electronic funds transfer
system. Later that same day, Rifkin phoned the wire transfer room and,
using a fictitious name, said he was with the bank's international divi-
sion. He rattled off a few security codes and his $10 million withdrawal
sailed through. Rifkin was eventually caught, but the bad publicity sur-
rounding the bank's security system had stuck with the bank over the
years. No one wanted a repeat performance.
    When Barry Himel called Black the following day, he said Mitnick
had been informed in person that the employment offer had been with-
drawn. Mitnick's response, Himel reported, was simply to smile. There
was a pause, then Himel asked: Did Jim Black think Mitnick was the
type of person to seek revenge in any way? If he were to seek revenge,
Black answered, it would most likely be through his knowledge of com-
puters and telecommunications.
   Two weeks later, Black got another call from Himel. He told Black
that one of the officers of the bank had just had a call from a news service
in San Francisco seeking confirmation and more details about a press
release from Security Pacific that had come over the wire earlier that
day. The release stated that for the first quarter of 1988, Security Pacific
was going to show an earnings loss of $400 million. The only clue Himel
had that the release was a forgery was the absence of a customary sign-
off identifying the source of the story. Of course, Himel told Black, the
story was completely false. Apprised of the fraudulent press release,
the bank's corporate officers were horrified. The potential damage to the
bank if such a release got into the newspapers was incalculable. In
plunging stock value and account closures alone, it could exceed the
$400 million reported in the fictitious press release. Fortunately, the
hoax was stopped in time. Again, Himel paused. Could this egregious
act possibly have been committed by Mitnick? It was possible, Black
replied, but he couldn't verify it. Nobody ever could.
    Computer crime cases were notoriously difficult to investigate, and
that was part of the appeal for Black. The little evidence that could be
gathered was difficult to tie directly to a suspect. Companies and univer-
sities whose computers had uninvited nocturnal visitors could produce
dozens of pages of computer printouts covered with blatant evidence of
the intrusion, but they didn't necessarily add up to much in the way of
evidence. Telephone traps were useful only as far as they went: intruders
such as Mitnick who had started out as phone phreaks were expert at
covering their tracks. When someone like Mitnick got into a telephone
company computerized switch, he could treat the vast telephone net-
                                     1Uv~:   Tk 7)~-5~    H~       T    103

work like a series of disappearing stepping-stones, manipulating the ma-
chines to create fictitious billing numbers and forward calls to
nonexistent telephones.
   Black saw computer crime as the ultimate challenge. Not only was
the technology interesting, but he enjoyed thinking about a computer
criminal's mind-set. Computer criminals hardly fit the common profile
of an outlaw. The old investigator's axiom, "We catch only the dumb
ones," seemed to break down when it came to computer crime. There
were no dumb ones. More often, they were caught because a friend or
an associate snitched. Black believed that Mitnick and his circle had
been snitching on one another for years, stretching all the way back to
the COSMOS incident, but the original gang Mitnick was involved with
appeared to have long since split up. Susan Thunder hadn't surfaced for
a few years. After a prostitution arrest in 1982, followed by a bizarre
incident in 1984 in which she and a friend tried to spring their friend
Steven Rhoades from jail by impersonating a deputy district attorney,
Susan had apparently given up her computer security consulting in favor
of beating the odds at professional poker. Both Roscoe and Rhoades
seemed to have gone straight by this time, and as far as Black could tell,
neither of them was involved in this incident.
   Black had never worked directly on a case involving Mitnick before,
but Mitnick appeared to be one of the few old phone phreaks, perhaps
the only one, who had kept his skills honed. From experience, Black
knew that this was a case that would require a lot of close cooperation
from Pacific Bell and Digital Equipment. Not only was Pierce College
using Digital equipment, but this VPA outfit was as well. Black had little
doubt that Mitnick had devoted hundreds of hours to refining his talents
on Digital computers.
   Black knew that Mitnick's technical ability far exceeded that of most
investigators, especially those unschooled in the subtle art of exploring
computer crime. The Santa Cruz Operation incident was a case in point.
Black and his colleagues had been stunned to hear of the careless way in
which the Santa Cruz police detectives had conducted the search of
Vitello's apartment. Why hadn't anyone told them not to take the
phone off the hook?
   Now that Black knew where the two suspects were working from, he
asked Pacific Bell to put a dialed number recorder on the VPA phone to
register outgoing calls. After eleven days of bureaucracy bashing, Black
was able to get the tap. It yielded a number of surprising calls: several to
nonworking, unassigned phone numbers; others made to the MCl local
104   •   C'lE6RPliNK
access using pirated authorization codes. Among the most interesting
calls were those made to a New Jersey Bell COSMOS computer similar
to the one Mitnick had been caught breaking into at Pacific Bell seven
years earlier. There was also a call to aNew York City-based subsidiary
of Security Pacific. Called Precision Business Systems, this company
managed data communications for the West Coast bank. The FBI's New
York office had recently been investigating a possible wire theft case
involving Security Pacific; in early April someone had attempted to get
into the bank's secured data communications lines.
   Although Black thought he had enough evidence from the Pierce
incident to nail Mitnick and DiCicco on computer fraud charges, he
kept holding out for more. While intriguing, the Precision Business
Systems lead was still too sketchy. Black wanted enough evidence to
warrant a substantial prison sentence and he hoped to get it from the
VPA surveillance. But in late summer, another case forced him to put
the investigation on hold and end the surveillance temporarily.
   In the meantime, Pierce College had brought its own case against
Mitnick and DiCicco. The college held a disciplinary hearing for which
the two suspects put together their own defense, exhibiting enough legal
acumen to draw the hearing out to an exhausting twenty-one hours.
When Black set aside the criminal case, they were appealing their ex-
   It had been a frustrating exercise for Black. These investigations
hinged on the close cooperation of law-enforcement agencies and com-
mercial institutions working together against a common foe. The detec-
tive had hoped for instant help from both Pacific Bell and Digital
Equipment. Instead, Pacific Bell's bureaucracy had stonewalled his ef-
forts. And although he had received immediate help from the local
Digital office, Black couldn't seem to get the attention of security people
at headquarters in Massachusetts. He hadn't been able to trace any
of Mitnick and DiCicco's outgoing calls to a major computer system.
Still, he was sure that somewhere out there a computer system admin-
istrator was having Black's problem in reverse, wondering who the
invader could be.

When Lenny began his job at Voluntary Plan Administrators in May of
1987, Kevin had no reason to visit: there were no modems. And even
though Lenny kept the news from Kevin when the company bought
modems that summer, Kevin somehow figured it out. Just as a well-fed
                                      fUv,*:   Ttt- 1)~-5~   H~     T   105

cat learns to run to the kitchen at the sound of a can opener, Kevin
started showing up regularly at VPA. The company, it turned out, was
an ideal place from which to practice his craft. As one of the principal
computer operators there, Lenny had the run of the place once everyone
had left for the day.
   But things were beginning to sour between them. Lenny knew Kevin's
capacity for threatening people. He had already threatened to turn
Lenny in for creating the fraudulent identity that got Lenny the flower
delivery job. But there was more to it. It was also a classic form of folie a
deux: Kevin could appeal to criminal aspects of Lenny's nature that
might have remained unrealized if the two had never met. Also, in
Lenny Kevin found someone with the flashes of intuition that Kevin
lacked and that were so necessary for their work. Lenny too had become
addicted to the excitement of breaking in to computers.
   Kevin had stolen, or at least tried to steal, software in the past. Now
he had a major project in mind: the acquisition of Digital Equipment's
most important software, the latest version of the company's VMS op-
erating system. Lenny became project assistant. They started out from
VPA in a low-key fashion. The first thing they established was a way
into the Arpanet, the vast research and military computer network. The
duo found an account at Patuxent Naval Air Station in Maryland. For a
few months in the summer of 1988, Kevin and Lenny used the Patuxent
computer as a convenient storage locker for the software they were
stealing electronically. When system managers at Patuxent noticed
something was going on and closed the electronic doors, the two looked
for another place to stash their data. They found it a few weeks later,
when they sneaked back into USC's computers.

Mark Brown noticed immediately that someone was in the system. Ex-
cept for the highly publicized degrees-for-sale incident three years earlier
in 1985, nothing very damaging had happened to the USC computer
system in the past few years. Brown was now manager of research and
development for USC's computing services, and he had all but forgotten
about the two teenagers who had so boldly waltzed onto campus and
broken into the system six years earlier.
   But now, someone was dialing into the USC system from off campus.
Having found a bug in the system program, the intruder was able to
modify the VMS operating system subprogram that acted as the com-
puter's gatekeeper. Brown had to admit the trespasser was extremely
106 •     CYEERPLINIC
clever. Somehow, the electronic interloper had altered the program that
supervised user log-ins so that every time someone logged in to the
computer, a copy of the password was slightly jumbled and stored in an
innocuously named place inside a file. And he had altered the program
so it left open a "back door" that allowed him to return any time to
harvest passwords.
    But there was a flaw. In trying to cover his tracks at the same time
that he was installing his rogue code in the system, the thief occasionally
crashed some of the USC computers accidentally. Users around the
campus, of course, simply assumed that the system had crashed. It was
the sort of annoyance that users expected from time to time. But Brown
could tell that these crashes were connected to the break-in. From what
he could gather, he had fallen victim to the infamous West German
Chaos Computer Club, which had achieved international notoriety that
fall after claiming responsibility for a summer of poking around inside
NASA's SPAN computer network, an international web of computers
used by scientists for space and physics research. The Chaos Computer
Club had attacked the NASA VAX computers using the very same
software trick. They called their program "the loginout patch." Who-
ever was breaking into USC wasn't just installing the loginout patch, but
was using the school's computer as a launchpad to break into other
computers on the Arpanet. That was precisely what Chaos had done on
the SPAN network. And since most of the activity seemed to occur in
the late afternoon and early evening, it made sense that West German
troublemakers would be at work late into the night from Hamburg or
Hannover or Berlin, or wherever they lived.
    A second odd occurrence left Brown even more concerned. Soon after
the break-ins started, he noticed that disk space was disappearing from a
USC computer with thousands of user accounts dedicated to coursework
in physics and chemistry. Huge chunks of storage capacity, forty mega-
bytes at a time-the equivalent of dozens of textbooks-were being
eaten up with no corresponding files to account for the missing space.
    After a few days of pulling apart the operating system in a hunt for
the source of the mysterious problem, Brown finally figured out that the
intruder was creating files and disguising them as system index files,
which are directories that describe other files-the last place anyone
would think to look. When Brown opened the files to examine them,
he was amazed. Someone was salting away the source code-closely
guarded original programs-for Digital Equipment's proprietary VMS
operating system.
   Brown didn't want simply to lock out a trespasser who had entered
privileged accounts and seemed to know the operating system. There
was no telling what might happen if USC locked him out. It might make
him angry enough to find his way in again and start doing some real
damage-deleting files or crashing systems.
   Software designers write source code in what are called high-level
languages. A translator, known as a compiler, then converts the high-
level code to binary form-ones and zeroes that can be understood by a
digital computer but are difficult for humans to decipher. Computer
companies guard their "human readable" source-code files as if they were
the crown jewels. Only the binary code, which cannot easily be under-
stood or modified, is distributed to customers. In this, computer compa-
nies are like a master chef who serves a six-course meal without giving
away the recipe. A competitor can easily copy a computer's hardware,
but recreating the operating system that runs on that computer is a far
more difficult and expensive undertaking. Companies guard their source
code not only because they're worried about competition, but also be-
cause access to source code makes it easier for a saboteur to open a secret
back door into the computer, known as a Trojan horse. A Trojan horse
is a seemingly innocent program planted inside a computer that is de-
signed for a special purpose, such as capturing passwords or even destroy-
ing data. It is often difficult to tell that a change has been made. Access
to source code can make the planning and execution of Trojan horses
   The thief was stealing Digital's lifeblood, millions of lines of software
that run on most of the VAX computers in the world, and electronically
stashing it in the USC computers. Furthermore, whoever was doing this
wasn't just taking any old VMS code; he was copying the very latest
version, called VMS Version 5.0. And the only place from which he
could have been lifting this software was a group of development com-
puters at Digital's laboratory in New Hampshire. Version 5.0 was so new
that Digital customers themselves didn't have it yet and would even-
tually get only parts of it on microfiche. Perhaps the intruder had man-
aged to break into Easynet, Digital's internal network, which connected
tens of thousands of Digital computers around the world. From Easynet,
the interloper must have found a gateway to the USC machine on an
academic network.
   Brown guessed that this thief was keeping his loot on ice at USC
because he didn't have enough storage capacity of his own or he didn't
want to be caught with the stolen goods. Every time the pilferer logged
108 •     CygERPUNK
out, Brown opened the files to see what was in them. He watched in
amazement as Digital's trade secrets scrolled by on the screen. Within
just a few weeks, many megabytes of disk space had been consumed in
the process of stashing stolen material in the illicit treasure chest.
   But when Brown called Digital's security department about the theft
taking place seemingly under the company's nose, he was surprised
again, this time by the lukewarm reception he received. He could un-
derstand that the world's second-largest computer manufacturer, with
tens of thousands of customers around the world, including a growing
number of banks and government agencies, wouldn't be eager to have it
widely known that some thief was breaking into its computers and mak-
ing off with its most prized software. After all, what did that say about
the safety of information stored on the customers' computers? But Brown
wanted to stop this thief and assumed Digital would rush to offer its full
support. He was hoping that the company would arm him with some
high-tech monitoring technology that would enable him to peek elec-
tronically over the invader's shoulder while he was in the act of burying
his plunder. He had heard that Digital had special programs that allowed
a system operator secretly to watch people who were on the system in
real time-that is, while they typed.
   When he called the local Digital office, he explained the situation.
"Look, we've got this guy breaking in here," Brown said, "and it's pretty
big time. He's got some of your sources." Brown asked if he could get
one of the company's monitoring programs.
   He was told to wait for a return call. The call came from Chuck
Bushey, the chief security investigator at Digital headquarters in Massa-
chusetts. But instead of offering to send a team out to Los Angeles to
investigate the situation, Bushey asked for an account on the USC
system so that his experts could look things over themselves. He prom-
ised he would send Brown the monitoring programs he was asking for.
But the software never came.
   From that point forward, Brown felt that the matter had drifted out
of his hands. Digital didn't seem exactly to be denying that someone had
penetrated the very core of the company, but the company wouldn't
acknowledge, to Brdwn anyway, the gravity of the situation. From his
conversation with Bushey, Brown got the impression that Digital was
somehow trying to brush the matter off. It was like dealing with the
Pentagon. Brown surmised that while the security people appeared
phlegmatic, software engineers back in Massachusetts were madly scamp-
ering to plug their security holes.
   The Digital security experts seemed much more concerned with the
lax security all over Easynet-it wasn't called Easynet for nothing-
than with the fact that the company's source code was being stolen.
Brown could have built his own VMS monitor, a program for watching
the intruder's keystrokes while he was on the USC system, but it would
have been time-consuming and tricky. VMS was an esoteric, user-un-
friendly operating system; it wasn't built for customers who liked to
tinker under the hood. It was built for large commercial and scientific
customers who were content never to touch it because they expected
Digital to handle their problems.
   So after a handful of unsatisfying encounters with the people in Mas-
sachusetts, Brown threw up his hands and decided to let the intruder
keep his millions of characters of disk space and have at it. After all, he
didn't seem to be hurting anything on the USC computer. And although
this was a lot of space to be giving to some uninvited guest, it wasn't
more than an engineering graduate student might use up in one night.
All he could do, Brown decided, was to keep careful logs of what he was
able to see.
    The intruder became bolder. He was beginning to use the USC system
as a way into other computers more frequently. So Brown shut down
access to his privileged accounts on the system. And no sooner had he
done that than the Brian Reid hoax happened.
    Chris Ho, who worked with Mark Brown at USC, got a telephone
call one afternoon in August from someone who identified himself as
"Brian Reid from Stanford."
    "We're having a break-in here, and it looks like he's coming from
USC," said the caller. "We need a privileged account on your system so
we can track him down."
    "Sure," Ho replied. "Just give me a number where I can call you back
and we'll set it up."
    "I'm not in my office right now."
    "Oh, then give me a time when I can reach you and I'll call you back
    "I'll have to get back to you." With that, the caller hung up.
    Chris Ho was more than a little skeptical of the call he had just
received. He called Stanford at once to check on Brian Reid, a Stanford
computer science professor. A secretary in the computer science depart-
 ment told him that Reid had left Stanford to work at Digital, in the
company's Western Research Lab in Palo Alto. Ho's suspicion con-
firmed, his curiosity led him to follow through. He called the Digital
110   •   Cyg£RPUNK
office in Palo Alto, and after several days of leaving messages and im-
ploring secretaries to get Reid to the telephone, he finally succeeded.
Reid needed to utter no more than two words for Ho to detect the
difference between Reid and the impostor. Where the timbre of Reid's
voice was deep and resonant, the earlier caller had spoken in a much
higher pitch, as if each word got trapped in his larynx before escaping
from his mouth. No, Reid assured the anxious USC worker, he had not
called Chris Ho, and yes, he had been aware of the break-ins at Digital
for months.
   After the phone hoax drew new attention to USC's plight, coopera-
tion from Digital improved considerably. In October, Bushey flew out to
Los Angeles and met with Brown and Ho. But Bushey seemed interested
only in logging the intruder's every move. And Brown knew what that
would mean: late-night vigils for the next two months, or until whoever
was using USC as an electronic warehouse was caught. He wasn't exactly

                                 .,. .
eager to lose weeks of sleep to help Digital, but he agreed to keep a
watchful eye on the situation.

Lenny had always enjoyed the aspect of traveling through computer
systems that made him feel like a fearless explorer. He liked the idea of
having computers throughout the world at his fingertips. The most ex-
citing thing about playing around on the Arpanet military network
hadn't been so much the information it contained; it was the act of
roving itself. Lenny's hacking may have kept him cooped up for days at
a time in the VPA office, or in the Hiway Host Motor Inn, but at the
same time it broadened his world far beyond the San Fernando Valley.
   The Arpanet was the granddaddy of all computer networks. Started
with United States military research funding in the late 1960s, the
network had served as the technology testing ground for the commercial
computer networks such as Tymnet and Telenet that were to follow.
The Arpanet originally linked universities with corporate and military
research facilities. By 1988 the Arpanet had largely been subsumed in a
growing thicket of commercial, academic, scientific, government and
military networks known collectively as the Internet. The Internet
joined individual networks together via computerized gateways so that it
was possible to travel electronically almost anywhere in the industrialized
   With the advent of computer networks, the traditional sense of geo-
graphic space as it was known to explorers of earlier times was becoming
                                     Kw~:   Tk 7)MJ.-S~   H~      T    111

obsolete. It had been replaced by a different notion, the idea of cyber-
space. Traveling from a computer in suburban Los Angeles to a computer
in Singapore was a matter of typing one command on a keyboard. It
happened instantly. In fact, distance on the Internet was so transparent
that a computer located in Southeast Asia would appear no different
than a computer in the next building or in the next county. They would
merely be different numbers on a computer host table, a listing of com-
puters on a network.
   As it happened, Kevin and Lenny's favorite computer manufacturer
was also a pioneer in computer networking. At Digital, networking really
meant remote computing-s-the freedom to move work around the net-
work from one computer to another. The company also tried to make it
easy for an engineer in Massachusetts to work on a set of data in Califor-
nia. Networking among Digital computers was built around simplicity,
uniformity and ease of use.
    In 1984, Digital built its own internal corporate network, the Easynet.
Eventually the Easynet would connect thirty-four thousand Digital com-
puters in more than twenty-five nations, giving direct access to nearly
two-thirds of the company's 120,000 employees. Easynet has made Dig-
ital the world's best-networked corporation. Engineers in Germany,
Japan and the United States share design work and rally support for
project proposals. Employees also send messages directly to Digital pres-
ident Ken Olsen. So it wasn't surprising that Easynet became a favorite
playground for Kevin and Lenny. Since it was strictly a VMS system, as
opposed to the UNIX-heavy Internet, Easynet spoke their language.
And for two young network explorers with an eye toward procuring
Digital's proprietary software, Easynet was the perfect transportation me-
dium. Once they were on one computer in the network, they could
connect to any other.
    When Kevin and Lenny penetrated Easynet, Lenny was as excited as
he had ever been. Later, Lenny would say that his breaking into com-
puters with Kevin was "like we were boldly going where no hackers had
gone before." But to Kevin "it was just a task, like coming to work every
day, just to get the job done." While Lenny seemed content to scamper
from one computer on the Easynet to another like a puppy, sniffing at
each new one, Kevin considered their computer system cracking a seri-
ous endeavor with a series of discrete goals. In fact, for as long as Lenny
could remember,Kevin had always approached his illicit computing as a
serious project. When he sought revenge on someone, it was if he were
taking on an assignment from some invisible employer. Once an assign-
112   ..   CYE£RPUNK
ment was complete, he reported back to Lenny or Roscoe with news of
his triumph. As early as 1981, Kevin would call Roscoe to report the
results of a frontal assault on the telephone service of someone he was
out to harass.
   Kevin's project for 1988 was downloading Digital's VMS source code.
It wasn't so that he could make pirate copies and sell them. Rather, it
was both the challenge of the hack itself and his intellectual curiosity
about such a complex and advanced program.
    Kevin's heavy phone use had just gotten him fired from a small firm
near Thousand Oaks that made electronic testing equipment. He was
sleeping until at least 11:00 every morning and would usually put in his
first call to Lenny around noon. Around 3:00 in the afternoon, Kevin
would instruct him to get started with a particular task. Then, at about
7:00, on the days they didn't have class at Pierce, Kevin would show up
at VP A, and after dinner they would return for a night of network
exploring. As the night wore on, Kevin's pocket pager sounded every
hour. It was Bonnie, wondering where her husband was. He lied and
said he was taking an evening class at UCLA. Kevin's lack of account-
ability to his wife didn't seem to matter to him.
    One of the first orders of business on these evenings was to go into a
large telephone wiring closet on the ground floor of the building and
connect VPA's modem line to another tenant's phone. To someone
checking toll records, it then seemed that calls were coming from an-
other business. By now, Kevin had refined this technique, prompted in
part by his increasing knowledge of how the police and telephone secu-
rity forces worked, and by a growing paranoia. He regularly checked to
make certain there were no dialed number recorders on the line he was
using. And to be extra safe, Kevin and Lenny looked for an untraceable
way to make long-distance calls. Like many other hackers and phone
phreaks, they frequently used illegal MCl calling card numbers. Most
phreaks would "scan" for them-that is, they would program their com-
puters to have their modems keep dialing a local MCl port and enter
different codes until a successful code was found. Purloined codes were
traded on electronic bulletin boards or "code lines," toll-free voice-mail
numbers that had been discovered by hackers and modified to dispense
pirated calling card numbers. But Kevin, in his inimitable way, had
perfected a method for obtaining telephone credit card numbers with
great efficiency. He later told the FBI that he had obtained the password
for the network security account on MCl's electronic mail system from
an electronic bulletin board. But Lenny knew that Kevin had talked
someone out of the password, using his excellent social engineering
   Once logged on to the MCI network security account, Kevin and
Lenny were privy to confidential information about new accounts, stolen
accounts and accounts that were in a state of limbo until MCI actually
discontinued them. They could also read electronic messages concerning
security breaches. It was a fertile field from which to harvest account
numbers, as well as an early warning system to subvert the security types.
Kevin guarded the MCI network security account from Lenny as if he
were in possession of the Hope Diamond. "This is something I could
make money with," he would tell him.
   Slipping into Digital's Easynet was a little like discovering the mother
lode. All of the company's private discussions were carried out over the
worldwide network. Using relatively standard tricks, Kevin and Lenny
collected passwords. For instance, when a machine is first set up, there
are a few preestablished accounts with preset passwords for the conve-
nience of Digital's field service people. Each such password gives access
to that account's electronic mail. Once Kevin and Lenny had found
these passwords, they used Kevin's gift for intuiting how organizations
functioned and who was important in an organization. An untrained but
instinctively expert social scientist, Kevin could look at patterns of com-
munication in stored electronic mail and figure out who had power and
who had valuable information. In that way, they found the mail that
was worthy of their attention.
   Andy Goldstein's mail was by far the most informative. Goldstein was
regarded by many as the most brilliant technical expert in the VMS
engineering ranks. He was also a VMS security expert. As a result, most
messages about VMS security problems eventually ended up in Gold-
stein's mailbox. One of his correspondents was Neill Clift, a researcher
at Leeds University in England. From what Lenny and Kevin could tell,
VMS security was a hobby of Clift's; he seemed to spend hours plumbing
the depths of operating system arcana in search of security flaws. Clift
was a prolific electronic correspondent. Most of his messages to Gold-
stein concerned problems with VMS security. And every time he de-
scribed a new flaw to Goldstein, Kevin and Lenny drank in every word.
   It was while snooping through Andy Goldstein's electronic mail that
Kevin and Lenny found the infamous Chaos Computer Club loginout
patch. From what Kevin and Lenny could decipher from the mail to
Goldstein, it seemed that someone in Europe had broken into a Euro-
pean institution using the Chaos patch. The victimized company had
114 ..    CYEERPUNK
made a copy of its entire system and sent it to Digital to analyze. Gold-
stein, it appeared, had analyzed the program, extracted the portion con-
taining the rogue code and reverse-engineered what the Chaos Club had
inserted. After stealing Goldstein's analysis from his mailbox, Lenny and
Kevin puzzled for days over the unexpected prize that had landed in their
hands. From what they could tell, someone from Chaos had changed
loginout, the program for logging in and out of the computer. Chaos had
modified, or patched, the program so that each time a password was
entered a copy of the password was sent to a spot in a remote corner of
the system, where it sat unnoticed until someone came to retrieve it.
Lenny and Kevin couldn't believe their luck. A program to steal any
VMS password, courtesy of Digital! Not only was the loginout patch small
and smooth, but it had an added feature that made the user's presence
difficult for the system to detect.
   To further confound Digital's computer security experts, the loginout
patch had been written in such a way that it successfully eluded ordinary
techniques to detect software that has been tampered with. Thus no
standard computer security alarms were set off by the program when the
patch modified the computer's operating system. A system operator
would never realize something was amiss. In fact, the patch was written
so well that Digital officials had little notion that their machines had
been so thoroughly compromised. Kevin and Lenny were delighted by
this bit of serendipitous international cooperation. More than that, they
were in awe of the Chaos Club for its ingenious hack. They had been
calling themselves "the Best," but now that they had seen this, they
amended their epithet to "the Best in the West. "
   So USC's Mark Brown had been right after all. However, it wasn't
the work of the Chaos Club itself he was seeing, but that of a pair of the
club's admirers who had discovered and were using one of the club's
cleverest hacks.
   The Chaos Club's famous hack into NASA's worldwide SPAN com-
puter network in the summer of 1987 had been a public-relations calam-
ity, both for NASA and for Digital, whose computers were the
foundation of the SPAN network. For months, members of the Chaos
Club had foraged through hundreds of computers on SPAN. By exploit-
ing an embarrassingly obvious hole in VMS, the cocky group of young
computer anarchists had first broken in to the CERN physics laboratory
in Switzerland, then electronically hopped over the Atlantic to Fermilab
in Illinois, and continued on to hundreds of computers on SPAN. When
the incident first surfaced after Chaos held a press conference to an-
nounce its accomplishments, it revived some concerns that were first
expressed following the 1983 release of the movie WarGames. The film,
which depicted a teenager who played havoc with a North American
Air Defense Command computer, roused widespread speculation that
bright kids could somehow compromise U. S. national security. In what
seemed to be a similar situation, the infiltration of SPAN by Chaos
implied to the public that SPAN was a sensitive military network. This
proved not to be the case; nevertheless, it was more than an embarrass-
ing incident for NASA and for Digital. The company suffered a corpo-
rate black eye. Nothing like this had ever happened to computers made
by International Business Machines, Digital's principal competitor.
   Digital could ill afford bad publicity. The company was under fierce
competitive pressure, moving thousands of employees from assembly
lines and corporate offices into the sales force in an effort to slash costs
and boost sales. For nearly a decade Digital's strength had been its sturdy,
steady seller, the VAX, a potent force against IBM. But by 1988, the
pace of technology was forcing Digital to seek out new markets while
continuing to satisfy its traditional customers. Digital was locked into a
race, scrambling to bring out new products and rebuild the VAX line.
The computer industry was no longer growing exponentially, particularly
in the minicomputer business that Digital had long dominated. Instead,
the markets for personal computers and workstations were still expand-
ing, but the company had stumbled badly in the PC business. To get
back on course, Digital was going to have to go after a more commercial
market than it had served in the past. Financial institutions were a new
major target and they would hardly tolerate potential flaws in the security
of the machines they were buying. So the company remained as quiet as
possible about the NASA incident. It patched the loopholes and tried
to tighten security. As it turned out, its efforts were less than successful.
   The loginout patch was written so cleanly that Lenny and Kevin had
to modify it only slightly to install it on the newer version of VMS used
in the United States. Once they had done that, they were ready to insert
their new Trojan horse. In the ensuing months it was to become their
most valuable electronic crowbar. Kevin and Lenny dubbed it their
"super-duper password scooper."

But even before Lenny and Kevin discovered the loginout patch, they
had used Kevin's wiles to get access to the company's VMS development
116   ..   CYEERPliNK
   Digital's largest software development facility is a complex of buildings
in Nashua, New Hampshire. Three of the Nashua buildings are at 110
Spitbrook Road and are code-named ZK01, ZK02, and ZK03. The
ZKO complex is the software capital of the company. It employs about
two thousand people-more than half of them programmers-and has
more than three thousand computers on-site, thirty of them mainframes.
There is a heliport outside that employees can use to travel to other
major Digital facilities or to Boston's Logan Airport. ZKO is set in the
middle of a stand of hardwoods, near a pond. It's at least a quarter of a
mile to walk indoors all the way from one end of the complex to the
other, and as you walk down the hall you look out at landscape that
probably hasn't changed much since the days of the American Revolu-
tion. It's an intriguing mixture of moods, and the tloor-to-ceiling win-
dows along the hallways manage to blur the boundary between modern
high-tech and hardwood forest.
   Managing three thousand of anything can be tricky. People who man-
age large collections of computers typically organize them into groups,
and into groups of groups. Fred Brooks, the chief designer of IBM's OS/
360 operating system, made a shrewd observation some time ago: the
structure of a computer system almost always mirrors the structure of the
human organization that created it. As a result, the VMS computers in
Digital are largely organized into groups that correspond to the manage-
ment structure: people who work together cluster their computers to-
gether under a single administration. However, technical peers work
across the grain of the organization and often give one another privileges
on their own machines.
   The Star development cluster of VAXes in the ZKO complex was
where all of Digital's VMS development work took place. Late one
night, soon after Lenny started at VPA, Kevin stationed Lenny at VPA
and, from a pay phone, called a graveyard-shift operator in the Star
control room. Posing as a technician in the field, Kevin had her walk
over to the console and type in a single command that seemed harmless
enough to someone without an intimate knowledge of VMS. Without
hesitation, she logged in to the computer, thereby spawning a new
process. A dollar-sign prompt appeared on Lenny's screen, the signal
that he could now type in the privileged commands reserved for a Digital
operator. But the full import of where Kevin and Lenny had landed
eluded them at first. Kevin wasn't so interested in hanging around to
check out the development cluster. It looked a lot like any other VAX
system. Instead, once they were in, Kevin invoked his personal battle
cry: "Get priv'ed, set up, move on." Setting up was a matter of creating
an account for themselves so that they could get in at some future time.
Who knows, the reasoning went, when access to the company's devel-
opment computers might be useful?
   Only gradually did it dawn on Kevin that they had achieved their
ultimate goal. Breaking into the development cluster could give them
unlimited access to the VMS source code. But even though they were
inside the temple, there were still obstacles. They needed to transport
the software to some safer haven, some secret location.
   In order to acquire the source code in such volume, one of the things
the two data hijackers needed was a high-speed gateway through which
they could transfer the programs. Relying on their relatively low-speed
modems at VP A would have been like siphoning the ocean through a
straw. Just as important, they needed to find a link from Easynet to the
outside world. Digital's computer network was not, as a rule, connected
to the outside world. But there were some exceptions. Digital's Western
Research Lab in Palo Alto, it turned out, had just the gateway they were
looking for. Established in 1982 by a group of defectors from Xerox's
Palo Alto Research Center, DECWRL, as it was called (pronounced
DECK-whirl), represented Digital's future, a computer science labora-
tory where researchers experimented with the most advanced ideas in
    The Palo Alto lab was composed of about twenty-five people in a four-
story streamlined brick building that also served as Digital's western
headquarters in the downtown section of Palo Alto, an upper-middle-
class community next door to Stanford University. The Western Re-
search Lab seemed more like a Silicon Valley start-up than a corporate
outpost. The lab was also one of the few UNIX havens in Digital's
otherwise VMS-dominated universe. It was logical that the researchers
at DECWRL would want to work with UNIX, specifically Digital's ver-
sion called Ultrix; not only had they grown up with UNIX, but in their
frequent cooperative research with university scientists, UNIX was the
common denominator. And there are other things that make UNIX
attractive: it is highly "portable," which means it can run on almost any
type of computer, from an IBM PC to a Cray supercomputer. VMS, on
the other hand, is confined to VAX computers. UNIX, which was in-
vented by two computer researchers at AT&T's Bell Laboratories, has
gradually become a standard for scientists, engineers and universities.
Scientists at the Western Research Lab were experts in the UNIX oper-
ating system, though they held a grudging respect for VMS.
118   ...   CYEERPLJNK
   Also at Palo Alto was the Digital Workstation Systems Engineering
Group, a small group of designers inside Digital who were working
around the clock on the next-generation workstation, Digital's first ma-
chine ever to be designed solely for the UNIX operating system. It was
being built outside Digital's traditional development process and the
company wanted to get the machine out as soon as possible in order to
remain competitive with smaller, younger companies, such as Sun Mi-
crosystems. The new workstation's code name was PMAX.
   High-speed network gateways may have been common at large uni-
versities and research laboratories, but they were found less frequently at
large corporations. Most IBM sites have no direct connections to the
outside world, mainly for reasons of corporate concern about security,
and those with connections can only dial out to other computers-they
cannot accept incoming calls. The Digital gateway in Palo Alto, with
links throughout the world, was extraordinary in its very existence. It
could crank data in and out at 56,000 bits per second, which is the
equivalent of transmitting all of Moby-Dick in less than two minutes. Its
purpose was to put the research of Digital's computer scientists closer to
research that was going on at places like Stanford and Berkeley. Having
good network connections mattered to a computer scientist in 1988 the
way a properly equipped kitchen would matter to Julia Child. As part of
the international research community, sharing ideas and papers and
ongoing research with counterparts around the world, the Palo Alto
computer scientists believed it was necessary to have direct connections
to the global Internet. And in recognition of the open-rnindedness back
at corporate headquarters, the computer scientists in Palo Alto took
great care to operate their precious gateway responsibly. To give the best
possible oversight both for maintenance and security, Ph. D. 's in com-
puter science took turns poring over daily log files, a job usually per-
formed by administrative staff.
   So it was only a matter of hours after the intrusions into the Palo Alto
computers began that the gateway watchers there noticed something
amiss. The Palo Alto scientists who monitored the gateway-Brian Reid
and Paul Vixie-decided they were dealing with experts. Sometimes the
intruders came into Digital over a phone line, as an employee might dial
into the company's computers from home. From there, they knew how
to get access to Easynet. From Easynet, they could connect to any of the
small VMS computers in the Palo Alto building, often a desktop work-
station, where they experimented with passwords until they managed to
log on. Once they had logged on, they exploited weaknesses in VMS to
                                     Kw~: T~ 7)~-5~ H~            l'   119

  gain privileges on the small system. And once they had done that, they
  would commandeer the small computer, forcing it to masquerade as a
  larger system on the network, some users of which would have expanded
. access rights. This is a method known as "network spoofing." At the
  same time, the intruders would plant their Trojan horse, a modification
  of the loginout program, and capture user passwords, and then take ad-
  vantage of the natural tendency for people to use the same password on
  all the systems on which they worked-whether VMS or UNIX.

 The intruders' objective was to log on to the gateway computer-an
 Ultrix machine called Gatekeeper, named for a character in the film
 Ghostbusters. Gatekeeper was the only means of contact with the Arpa-
 net. Getting into Ultrix, then, was necessary if they wanted to take
 software out of the company. Once the intruders had gathered their crop
 of passwords from a VMS machine, they could try each one of them on
 an Ultrix machine. And that is how they got into Gatekeeper. Long
 after midnight one night, Paul Vixie noticed activity in the account of
 one of the secretaries, who appeared to be using Gatekeeper in a very
 unusual and sophisticated manner. The next day he had her change her
 password, and disabled her account on the Gatekeeper computer.
    It was a daily battle, and even the best computer scientists at the
 research laboratory began to feel beleaguered. The intruders were creat-
 ing too much extra work. Reid and Vixie were copying all recently
 changed files to special backup disks in case something went wrong. The
 intruders' clandestine copying activities were creating an unusually large
 number of recently changed files, which occasionally overflowed the
 backup disks.
    The Palo Alto scientists felt that the intruders were as clever as any
 they had ever seen. They were experts who didn't make mistakes. They
 broke into almost any computer in Digital at will. It was evident from
 the way they were exploiting bugs that they must have source code,
 some pieces of which were so proprietary that even Reid and Vixie didn't
 have access to them. Frustrated by his own lack of access to the source
 code, Reid wrote in a message to Digital's security administrators:
 "Clearly the intruders have the sources. When sources are outlawed,
 only outlaws will have sources." One thing they did to limit the loss of
 information was to alter the data just before the intruders transmitted it
 out of the Palo Alto computer, coding it in such a way that it was useless
 after it was stolen.
120   A    CYG6RPUNK
    By August, at least three people in Palo Alto were taking turns mon-
itoring the laboratory network. It usually made for an interesting eve-
ning. One thing they noticed was that unlike a typical intruder in a
computer, whoever this was had a fairly sophisticated method of sifting
through directories for what he wanted. Ordinarily, a meddler entered a
directory that appeared to contain something worth looking at and read
the first twenty lines or so of several files. But this intruder was consid-
erably more efficient. He had an uncanny way of finding a computer
with vast amounts of data and within half an hour he would have the
one thing up on the screen that was worth stealing. He did it by execut-
ing a sophisticated sweep of the entire system, looking at access times
and looking to see which people had been using which directories. His
search strategy, curiously people-oriented, revealed an ingenious insight
into corporate hierarchies. That is, he seemed to be able to figure out
which people were important, then he would go look at what they had
been doing recently. The kicker was that he kept opening the safe and
looking inside, but he never took anything other than source code.
Perhaps he was just casing the joint.
    Among those involved in chasing down the interlopers, frustration
grew. Sometimes, just to get some work done, or to get a decent night's
sleep, the Palo Alto crew would break the network links between Palo
Alto computers and the outside world, disconnecting from both Easynet
and Internet by pulling the plug on the modems. If the computer scien-
tists at DECWRL felt that they were under a sustained attack, some
system managers at Digital's commercial offices were terrified, convinced
that someone had taken the control of their machines out of their hands.
    In an electronic mail message to his managers, one irate Digital em-
ployee wrote:

We seem to be totally defenseless against
these people. We have repeatedly rebuilt
system after system and finally management
has told the system support group to ignore
the problem. As a good network citizen, I
want to make sure someone at network
security knows that we are being raped in
broad daylight. These people freely walk
into our systems and are taking restricted,
confidential and proprietary information.
   The best that could be done was to take the VAX development cluster
in New Hampshire off the Easynet, which meant that engineers were
unable to work on those computers from home, or from remote sites. To
many of them, it was an irksome and unacceptable solution.
   In the meantime, those in the Workstation Systems Engineering
Group in Palo Alto who were working on the new PMAX workstation
had some real trade secrets to protect inside their computers. The Palo
Alto group was toiling secretly, trying to bring its workstation to market
ahead of Sun Microsystems, which was working on a similar computer.
The principal objective behind the development of the new workstation
was to take Digital's competitors completely by surprise. Losing that
element of surprise could significantly hurt the computer's chances for
success. Although the intruders logged on to machines belonging to the
workstation group, they didn't appear to be interested in taking anything
there. But as an extra precaution, the group kept the specifications to
the new machine off line, and gave them out in person to a select group
of people.

It might have had something to do with residual instincts left over from
his days in the intelligence community, but Chuck Bushey was con-
vinced that the break-ins were the work of an international espionage
ring, with malicious West German computer hackers feeding valuable
Digital programs to the East Bloc. Bushey was Digital's top cop and he
had a cop's way of thinking. Once he had gone to a morning appoint-
ment at a remote Digital office two hours early to test security by con-
fronting guards at all the doors.
   But Bushey was unschooled when it came to technical matters, and
the subtle points of computer penetration that a computer security expert
might find significant tended to mystify him. His suspicions arose in part
from the fact that he had recently received intimations of shady doings
on the part of the Chaos Computer Club, and partly because some of
the phone traces went back to network nodes close to East Bloc borders.
Another piece of evidence led to Karlsruhe, West Germany, which
Digital officials knew to be a hotbed of Chaos Computer Club activity,
and to a university student who had spent the previous summer working
at Digital in Palo Alto. While at Digital, the student had spent much of
his time in front of the photocopier. Moreover, the German student was
the only person who knew the location of Brian Reid's password-cracking
122   •   CYEERPUNK
software, which the intruders stole from a Palo Alto computer. By this
time, it seemed that the Digital security forces were building up a strong
case for the existence of an international conspiracy.
   In early November, Bushey met with Mark Rasch, a young attorney
from the Justice Department who had developed an expertise in com-
puter crime. Bushey told Rasch about the months of fruitless investiga-
tion and false starts with no suspect. When Bushey posited his
international conspiracy theory, Rasch found it very plausible. A tough-
minded thirty-year-old New Yorker who had gone straight to the Justice
Department from law school, Rasch had become proficient at investigat-
ing high-tech crimes. His first position was in the department's internal
security division and his early cases had included an investigation into
the unlawful export of VAX computers to Eastern Bloc countries. His
practice was to assume the worst. Presented with a set of brazen, system-
atic electronic break-ins aimed at the heart of Digital's software devel-
opment efforts, his inclination was to assume it was espionage until
proven otherwise.
   It was a busy week for Rasch. While meeting with Bushey he received
word that a computer virus had infected the Internet the previous eve-
ning, crippling thousands of computers around the nation. Mike Gib-
bons, an FBI special agent in the Washington field office, tracked Rasch
down at the Digital offices that morning to tell him he had already
opened an investigation into the Internet virus. As Rasch was hearing
the reports, there were still no leads on who had written the virus. Rasch
saw no apparent link between Digital's troubles and this new computer
virus, but he couldn't rule one out entirely. One of the virus's targets
was VAX computers.
   The Palo Alto scientists were beginning to disagree with Bushey. By
November, they had been monitoring the intruders' activities for
months. This was not international espionage at work, they decided.
First of all, the passwords that the intruders made up when they created
accounts for themselves-words like Spymaster and Spoofmaster-were
in consistently flawless American vernacular. Second, it was widely
known that the Soviets already had both VMS and UNIX and wouldn't
need to steal them. Moreover, if they did need to steal an updated
version, it was more than likely that among Digital's 150,000 employees,
or among those few customers with source code, there was at least one
Russian operative who had already taken home everything he needed. It
just wasn't necessary to go through all the effort that the intruders under-
took. Brian Reid, the Palo Alto office's resident computer security ex-
                                    ~~:      Tk 7)MJ;.-5~   H~   ...   123

pert, believed that in all likelihood this was a game being played by some
ne'er-do-well somewhere who slept into the afternoon, got up, ate some
breakfast and sat down at his home computer at three in the afternoon
for a long day of hacking. After a hard day at the office breaking into
computers, he (experience had taught Reid that it was inevitably a male)
would knock off around midnight, meet some fellow hackers for pizza
and go to bed around three in the morning.
   Not only did those hours correspond directly to the hours during
which the break-ins took place, but Reid had spent enough time around
the computer underground to know its mind-set. In the ten years Reid
had taught at Stanford, he had been involved in network security, had
met several such "cyberpunks" and had sparred with dozens of them
inside the Stanford computers. Both Reid and Vixie were impressed with
the technical skill displayed by this one, but they argued that anyone
with a modicum of talent and ten hours a day to devote to the problem
could probably do the same thing. The one piece of evidence in favor of
Bushey's international conspiracy theory was that the intrusions were so
energetic they had to be the work of at least two people, a tag team
perhaps. If these were truly bored teenagers, on the other hand, then
there was also in them something of the network cowboy, out on the
nets having a good time roping computers like so many steers, just out
to see how many network nodes they could lasso in a night. Witnessing
the frenetic activity every day was a little like watching a movie with
the sound turned off; Reid didn't get to hear the whooping with each
vanquished system, but he could imagine it well.
                                 .. ... ..
Retrieving the data from USC was a three-man operation, so Kevin and
Lenny pulled a reluctant but curious Roscoe in on the project. Fearing
that he would be recognized, Kevin stayed off campus; Lenny and Roscoe
went to USC together. Lenny stationed himself at a pay phone outside
the university's main computer center and called Kevin, who was sta-
tioned at a terminal at the offices of a friend. Kevin then instructed
Lenny to tell Roscoe to mount a magnetic tape. Roscoe walked in and,
dusting off his old social engineering skills, posed as a student working
on a project, handed a tape to a computer operator and asked him to
mount it. Once that was accomplished, Roscoe returned to the pay
phone where Lenny had Kevin on the line, and Kevin typed in the
appropriate commands to transfer the files from the USC hard disk to
the magnetic tape. Then Roscoe could retrieve the tape and hand it to
124    .&   CYE£RPLINK
 Lenny. This complicated procedure had to be done a few times in order
 to build the collection of tapes, which were gradually stacking up at All-
.Purpose Storage in Mission Hills, where Lenny had rented a locker.
 Roscoe occasionally provided a blank tape from his workplace, but most
 of the tapes came from VPA.
    Kevin and Lenny both considered what they had accomplished a
 tremendous feat. By having their own copy of the VMS source code,
 they could pick the operating system apart line by line, understand it,
 find its numerous flaws and even, if they wanted, create a modified
 version-say, with something like the Chaos patch included. If Kevin
 and Lenny really wanted to be malicious, they could send contaminated
 software back to the Star cluster, or even to the software distribution
 center in Massachusetts, which generates exact copies of what it has
 received. People who work at a software distribution center are typically
 not engineers and would have no way of checking to see if their copy of
 the information had been secretely altered. That meant that until the
 next release came from the developers, every copy of VMS that left the
 distribution center for a customer site would contain the corrupted soft-
    When the engineers back at the Star cluster discovered that someone
 had broken in, they had no way of knowing whether or not their master
 source code had been corrupted. So they were forced to go through a
 laborious week-long exercise of installing all the magnetic tape backups
 of their master files and comparing them with what was on the disks, to
 make sure that every change was authorized.
    But Kevin and Lenny had had no plans to visit havoc on the software.
 Their electronic joyride through Digital was harmless by their standards.
 They were copying the source code, but in their minds they weren't
 taking anything. As they saw it, they weren't even stealing, because
 they left the original software intact, just where it was and just how it
    By the same idiosyncratic view of the law, when Kevin and Lenny
 encountered a security program called XSAFE, which they learned about
 while reading the confidential electronic mail of an engineer named
 Henry T eng, they had no qualms about taking it. T eng appeared to be
 the program's chief architect and "Digital Confidential" was stamped all
 over any memos concerning XSAFE. The program wasn't due to be
 released for another year, and Kevin wanted the source code.
    It didn't take Teng long to notice they had been in, so he changed
 his password. Getting Teng's new password was a particular challenge,

as he was choosing from a list of randomly generated passwords. Al-
though the password T eng picked did not echo back on his screen, the
list from which he chose went to a log file he kept, which Lenny and
Kevin could see. They went to the log file, found the list of choices, and
eliminated their way to Teng's new password.
    When T eng discovered they were back in, he moved to a machine
that was on the Easynet but had disabled inbound connections. Through
reading mail, Kevin and Lenny found the name of the machine Teng
was now on. When they tried the "set host" command, a way to link
one computer to another, they got a response that all inbound connec-
tions would be refused. Figuring that T eng himself had to have a way of
connecting to the computer, they started looking at terminal servers
(computers that act as switchboards, making connections between com-
puters on a network) in the development center and finally found a
hidden avenue that gave them a way to reach to the machine. They
continued the task of transferring out the XSAFE source code.
    Perusing the personal mail of network security experts like Brian Reid,
and corporate security people like Chuck Bushey and Kent Anderson,
who was in charge of European security, was a good way for Kevin and
Lenny to track the progress of the hunt. They read mail in which net-
work security people were told to drop what they were doing to work on
the case. At one point, it appeared to Kevin and Lenny that at least
three people in corporate security were working on the problem full-
    Even if they were caught, Lenny and Kevin had read enough about
computer crime to know that Digital might be reluctant to press charges
against them. They were aware that few of the computer crimes detected
were ever reported to the police and still fewer were made public through
criminal charges. Lenny and Kevin knew that companies worried about
having their vulnerabilities publicized.

But in June 1988 there was a close call. Lenny got a telephone call one
day from a particularly overwrought Kevin.
   "Get a copy of today's L.A. Times!" he cried.
   When Lenny looked at the newspaper, he couldn't believe it. "JPL
Computer Penetrated by a 'Hacker' " was the front-page headline. The
duo's traversing of the Internet had taken them briefly into a computer
at NASA's Jet Propulsion Laboratory in Pasadena a few months earlier.
Kevin was interested in the JPL computer because it was one of the few
126   .A   CYf;ERPLlNK
VMS machines on the Internet and he had considered using it as a
storage locker for the VMS source code, but then decided against it. But
they didn't think they had been detected. Luckily, the laboratory ap-
peared to suspect the Chaos Club, because of evidence of the club's
trademark "insidious Trojan horse," the loginout patch. The newspaper
article said that although no classified material was compromised by the
hackers, one JPL engineer said officials there "worried that an intruder
could learn how to send bogus commands" to the eight unmanned inter-
planetary Explorer spacecraft the laboratory controlled. Kevin and
Lenny, of course, had no such intentions. They had just been poking
   By early fall of 1988, Kevin was acting particularly paranoid, eating
more than usual and making unreasonable demands on Lenny. He had
no regular job, leaving him with plenty of time to be obsessed with his
hacking projects. He insisted that Lenny be ready to drop what he was
doing in his legitimate job and start work for Kevin promptly at three
o'clock every afternoon. Or he would call Lenny at five in the morning
and instruct him to go to VPA to insert the Chaos patch just as everyone
in Massachusetts and New Hampshire was logging on. Kevin had
changed his telephone billing name to James Bond; the last three digits
of his new number were 007. He had changed Lenny's billing name, too
-to Oliver North.
   Kevin made it clear to Lenny that he wanted to stay several steps
ahead of the growing pack of hounds Digital had deployed for the chase.
By this time, Kevin had improved his phone work so much that he made
it almost impossible for USC to track him down. His most effective red
herring was his use of the call-forwarding feature on the telephone com-
pany's computers. This masking technique meant that when the phone
company initiated a trace, it would lead to a blind alley. Kevin was
ruthlessly arbitrary in forwarding the telephone numbers. And until he
and Lenny read Chuck Bushey's mail one day, they had no idea that
Digital had gone so far as to trace the calls back to the random numbers
they had chosen to forward to. One of these random numbers led back
to the apartment of a Middle Eastern immigrant in Santa Monica, a
middle-aged man who was watching television when federal agents
barged in looking for computer equipment. They found none.
   When in the throes of the hacking, Lenny was having a good time.
The idea of having control over many computers was incredibly seduc-
tive. The challenge of figuring out where to go to look for the informa-
tion they wanted was more stimulating than any college programming
                                    1Uv~:   Tltt-   7)M14-S~ H~   ...   127

course. And all the while they watched a twelve-billion-dollar company
flounder helplessly as it headed down one cul-de-sac after the other.
   Yet something in Kevin was changing. He had become so secretive
that when he told Lenny he had lost the Security Pacific job, the greater
surprise to Lenny was that he had gotten it in the first place. The same
thing had happened when Kevin got and quickly lost the job at GTE.
For Lenny, too, the thrill was beginning to disappear. He became suspi-
cious that Kevin was storing information to use against him someday,
while withholding anything about himself that Lenny could use. They
began to argue frequently over trifling things. And they were constantly
betting against each other-twice Lenny lost a $120 bet on whether
Kevin could break into the VPA computer from outside. Kevin did it
through sheer doggedness. When it came to perseverance, Lenny was a
distant second. He lost most of the bets.
   And since Kevin had come back from his months "underground" a
few years earlier, he seemed to take his electronic penetration more
seriously than ever. Perhaps he was thinking that if he was going to lie
to Bonnie every night about where he was, it should be for a reason he
considered defensible: he had goals to meet. On the other hand, some-
thing seemed to be beyond his control: he kept telling Lenny that he
wanted to stop the break-ins as soon as they finished the project they
were working on; but once they had finished one project, Kevin always
seemed to want to start another.
   By November, Kevin was insisting on coming to VPA every night to
break into Digital. Whenever Lenny tried to beg off so that he could get
some sleep, see other friends or spend time with his girlfriend, Kevin
would badger him until he agreed to spend the evening with him. Some-
times he even called Lenny at two in the morning and rousted him from
bed, insisting that he meet him at VP A. Kevin was obsessed and he had
drawn Lenny into his mania.
   Bonnie, who was paying most of the bills, had started losing patience
with her errant husband. She paged him constantly, and Lenny sat in
disgust as he listened to Kevin lie. "I'm at the UCLA law library working
on the Pierce case," he would tell her. Or: "I'm studying for my exten-
sion class." After he hung up, Kevin would grumble about Bonnie. She
was a nuisance. Getting in the way. He was thinking of divorcing her.
   Lenny could feel that things were heating up. He turned to Roscoe.
Kevin was getting out of hand, he told him. He felt as if he couldn't pee
without checking in with Kevin first. Roscoe, too, was beginning to get
fed up with this roller-coaster relationship between Lenny and Kevin. It
128   A   CygERPUNK
was true that Kevin was a nuisance. He was calling Roscoe all the time,
too. Mostly, Roscoe felt terrible for Bonnie, who knew of none of it. He
resisted a temptation to tell her to go to the UCLA law library one night
and look for her husband. Now Lenny was sounding panicked. Roscoe
didn't want to get involved, but he did offer him some advice: get a
   Even now that the VMS software was encrypted and safely locked
away in a storage locker, Kevin still wouldn't quit. His next project, he
decided, was to steal a game called Doom from Digital's Star cluster. He
was quite content with the setup at VPA. Judging from the unsuccessful
traces, there was no one even close to catching them. In the face of
Lenny's protests, Kevin was insistent. What had once been a friendship
between like-minded teenagers had become a series of demands and
contests and long nights in the stuffy offices at VPA. If this was big-time
electronic crime, then Lenny wanted nothing more to do with it.

It was the stunt with the paycheck that unhinged Lenny completely.
   "Hello, Ms. Sandivill," came the chatty, genial voice on the other
end of the line. "This is Carl Halliday with the Internal Revenue Ser-
vice. I understand you're in charge of payroll at VPA."
   "Who did you say you were?"
   "Well, ma'am, we're doing an investigation into an employee of
yours. A Mr. Leonard DiCicco. Seems he owes Uncle Sam some money.
So we need your cooperation. "
   "We need to ask you to withhold Mr. DiCicco's paycheck until we
clear this matter up."
   Livia Sandivill, VPA's bookkeeper, shook her head. There was some-
thing about the call that made her uneasy. She asked him to hold on
and went into Ralph Hurley's office.
   "The IRS is on the line and they're asking about Lenny," she told
him. "They want us to hold his paycheck."
   Hurley, VPA's vice-president, took the call. At first, the IRS agent
seemed credible enough, but it didn't take long for Hurley to become
skeptical. When he asked the agent to fax him a letter with the request
to hold the paycheck, the agent said his fax machine was broken. Then,
when Hurley asked for a number to call him back, the caller gave him
Sacramento's area code but a number with a prefix Hurley recognized as
                                    ~~:   Ttt. DMJ.-5.;.k H~     ...   129

one for California state agencies, not federal agencies. Moreover, there
was something very familiar about the voice. Of course! He sounded
exactly like Kevin Mitnick, that overweight, Eddie Haskellish friend of
Lenny's. Kevin was a frequent caller. He chatted up the secretaries in
the same friendly tone. "How was Palm Springs?" he would inquire of
one. "Sorry to hear about your appendicitis" were the kind words for
another. He called so often, in fact, that it had become something of a
joke among the women who answered the phone at VPA. Whenever
Lenny was on the phone to Kevin Mitnick, he sounded as if he were
fighting with a girlfriend. When Hurley tried calling the number back,
he got a recording that said there was no such number.
   So just after lunch, Hurley called Lenny into his office. "Lenny, are
you in trouble with the IRS?" he asked.
   Lenny shrugged it off. "Of course not."
   "Then are you in trouble with Kevin?"
   That was all it took. Before he even knew what he was doing, Lenny
found himself telling Hurley about Kevin. In a rather disjointed confes-
sion, he told Hurley that Kevin had been operating from VPA for the
past year, that he had been letting Kevin into the building after hours
to sit in front of the VPA computer until early morning. He told him
that he and Kevin had been secretly using a computer twenty miles away
on the USC campus as their personal data storage locker. And he told
him about all the break-ins at Digital. Hurley, who had suspected noth-
ing, was stunned to hear this.
   Lenny knew he was well regarded at VPA. He was the general troub-
leshooter who maintained the company's modest computer system. The
last thing he wanted to do was lose yet another job because of Kevin.
   Lenny told Hurley he wouldn't have been doing all this illicit com-
puting if it hadn't been for Kevin. In fact, he claimed, Kevin had forced
him into it. Kevin Mitnick was cannibalizing his life. So he kept letting
Kevin in to use the VPA computers, night after night. If he decided he
didn't want to work with Kevin one night, Kevin always managed to
track him down. He was a master at that. And Lenny wasn't the only
one to have been terrorized by Kevin Mitnick through the years, Lenny
said. He had seen Kevin cut off a person's phone service on a whim.
   Hurley leaned back. "Well, Lenny, what are we going to do? I'll tell
you right now that I want to get the authorities in here. Either you can
do it or I will."
   "Well, geez," Lenny said, looking alarmed. "I'll just get in trouble."
130   •   CYf;£RPlINK
   Hurley tried to be understanding. He told Lenny that if he was the
victim he claimed to be, he would be likely to get sympathetic treatment
from the authorities. "Why don't you think it over," he said.
   At 6:00 that evening, Lenny came back into Hurley's office.
   "Well, Kevin's really done it this time."
   "What do you mean?"
   "He just forwarded all our lines to my house and busied out the whole
switchboard." When Hurley picked up his phone and tried to call an-
other line in the office, he just got a busy signal.
   With this came another round of confessions. Now Lenny told Hurley
that Kevin wasn't just a computer expert, he was adept with telephones
as well. He told him about rerouting the wires in the telephone closet
and about the MCl access codes they used to call places as far away as
England. The phone forwarding stunt he had just pulled was something
Kevin had been threatening Lenny with for some time, Lenny said.
   "Something has to be done about this by tomorrow," Hurley said. "1
want to see you in here at eight-thirty tomorrow morning." Later that
evening, both telephone lines at the Hurley residence went dead.
   The next morning, Lenny told Hurley he knew just whom to call at
Digital, and he placed a call to the headquarters in Maynard, Massachu-
setts, from Hurley's office. He told the man who answered in Digital's
corporate security office that he needed to speak with someone about the
computer hacker who had been plaguing Digital for the past year. To
make his words sink in, he started listing things: the VMS source code,
the XSAFE program. Lenny said he would like to be put in touch with
the FBI. He was told to sit tight for a few minutes.
   Five minutes later, Chuck Bushey called him back. Lenny knew ex-
actly who that was because he and Kevin had been monitoring Bushey's
personal electronic mailbox for months. He knew Bushey was angry and
fed up. And he knew that the security department at Digital would give
anything at this point to catch those who had privately embarrassed the
company so thoroughly. "Uh, hello," said Lenny. "1 just wanted to let
you know that 1 know who's been breaking into Digital's computers for
the last year."
   There was silence on the other end. "Who is this?"
   Lenny stumbled. "My name, uh, my name is Leonard DiCicco and
I'm calling from Los Angeles and 1 want to tell you about Kevin Mitnick.
1 know that he's been breaking into Digital's computers all this time. 1
know you're looking for him." Bushey told him to go find a "secure"
phone and call him from there.
   Lenny and Hurley went together to a pay phone at a nearby super-
market, and Lenny called Bushey back using Hurley's credit card. "Well,
I'm, uh ... I'm a friend of his, and I'm calling you because I want to
turn him in," Lenny said, and he launched into a condensed version of
what he had told Hurley the previous day. He made certain to mention
all of the key incidents. He told Bushey that Kevin had taken a copy of
a highly secret security program that only a handful of people within
Digital were even aware of. To prove it, he named the program. He also
said that a copy of the latest version of Digital's operating system for the
VAX had been taken as well.
   There was silence on the other end. Lenny kept talking. He said he
wanted to be as much help as possible in apprehending Kevin Mitnick,
who was, Lenny intoned, a menace to society. Silence. Digital had
better move fast, Lenny added, because Kevin had been talking lately
about giving up his break-ins, and the company might not have many
more opportunities to catch him in the act.
    Bushey was speechless at the sudden lucky break. This breathless
young man on the telephone was describing the break-ins that had been
plaguing the company for months. That night, Bushey and Kent Ander-
son, the company's head of European security, were on a plane to Los
    While the two Digital people were on their way, an FBI special agent
named Chris Headrick summoned Lenny to a meeting at a hotel in the
San Fernando Valley. Lenny was impressed by the very ordinariness of
the agent. Slight and bespectacled, the low-key Headrick looked as if he
would be more at home in a classroom in front of a chalkboard, scrib-
bling out differential equations. In fact, some of Lenny's high school
teachers were flashier than this. When the Digital security men arrived
at the hotel, they turned one guest room into a makeshift command
center. Lenny drove out to All-Purpose Storage, retrieved the three
dozen tapes filled with the fruits of his and Kevin's electronic adventures
and surrendered them to his interrogators.
    Lenny stayed at the hotel, talking until 4:00 A. M. Feeling rather like
a hero, he was brazen enough to mention to Bushey that Digital could
have solved this frustrating case much sooner. If Bushey had pressed
people at USC, university records would have revealed a history of
problems with Kevin Mitnick. Also, Lenny told him, if Bushey had
consulted Digital's old records, he surely would have seen that Mitnick
had gotten into trouble as early as 1980 for breaking into a Digital
computer. Then there was the Pacific Bell security memo that followed
132   •   eygERPUNK
the Santa Cruz Operation incident and circulated around Digital several
months later, at the same time that Mitnick's activity inside Digital was
at its peak. The names had been deleted from the memo but a few phone
calls to Pacific Bell security would have made the Mitnick connection.
   The next morning, together with Lenny and Ralph Hurley, the au-
thorities drew up a simple plan to verify Lenny's claims. As usual, Kevin
was planning to show up at VPA at 7:00 that evening, after the regular
employees had all gone home. Headrick's plan was to use a computer
room on the first floor of the building to monitor everything Mitnick did
from the main computer room, which was one flight up on the opposite
side of the building. Hurley stayed around that night too. His presence
downstairs would prevent Kevin from using the first-floor computer
room. An FBI agent named Gerald Harman arrived at 5:00 P.M. and
began to set up a surveillance. He wired Lenny with a tape recorder,
fastening an elastic belt with two pockets-one for the tape recorder
and one for the microphone-to his chest. A Digital engineer installed
a monitoring device into the VPA computer that would let them watch
and record everything that happened from the system upstairs.
   When he arrived that evening, Kevin seemed to be in good cheer. In
fact, he was on a roll. He had recently gotten a more extensive list of
users on the computer at the university in Leeds, England, and wanted
to try to get their passwords. He was also hoping to break in to the
machine of Neill Clift, the Leeds researcher whose hobby was spotting
security holes in VMS.
   The agents planted themselves downstairs, and Hurley stayed in his
office. Lenny greeted Kevin outside and lowered his voice. "Kev, we
can't use both computer rooms tonight. Hurley's working late." The two
went straight upstairs and Kevin began to dial directly into Digital. His
first order of business was the Digital engineering cluster. Lenny was
wearing a loose-fitting Tvshirt, but he was still worried that Kevin would
pat him on the back, thwacking his hand against the recording equip-
ment. Kevin sensed that Lenny was jittery. "What's going on with you
tonight?" he asked. "You're acting strange."
    Downstairs, in the hastily appointed command center, the Digital
people must have been surprised to learn that all their theories had been
wrong. Not only was this not the work of an international spy ring, but
the notorious West German Chaos group didn't seem to be involved at
all. Instead, the culprit appeared to be an obese, nearsighted twenty-
five-year-old high school dropout from L.A. whose diet, according to his
friend, consisted of greasy cheeseburgers and Big Gulp colas from the
                                     Kw~:   Tk 1)~-5~     H~       ...   133

nearby 7-Eleven-to say nothing of Lenny, a whiner who had spent
months letting himself be bedeviled and blackmailed by the fat one.
This must have been a far less glamorous denouement than any of them
had imagined, inside a stuffy room the size of a walk-in closet, listening
to two grown men taunt one another like children.
   As the evening wore on, it became clear that this Kevin Mitnick was
indeed their man. Mitnick seemed to have a special fascination with the
Leeds computer that night. He talked someone at Leeds into giving him
a password, then logged on to the British system and perused lists con-
taining the names and access codes of everyone with legitimate access to
the Leeds computer. The FBI agent seemed befuddled by what he was
witnessing. Anderson explained to him that once Mitnick had obtained
the list of authorized users and their passwords, he could come back later
and log on as any of those people.
   This young man's facility with the telephone network in general, and
MCI codes in particular, was impressive enough. As soon as he finished
his session on the Leeds computer, Mitnick dialed his way into an MCI
Mail account, from which he was able to harvest MCI credit card num-
bers, so that the session would never show up on VPA's telephone bill.
He seemed to know exactly what commands to type, as if he had done
it hundreds of times before. But even more remarkable was Kevin's
ability to monitor Pacific Bell's telephone traces. And that was precisely
what had kept Digital at bay for so long. His knowledge of the phone
system was so extensive that he could tap straight into Pacific Bell's
monitoring equipment and keep a careful watch over anyone trying to
trace his call. For that purpose, he always had two active terminals in
front of him: one for his breaking and entering and a second to show
what was happening on Pacific Bell's computer. If a trace to his phone
line started, he could log off at once.
   Anderson had been in the computer business for nearly a decade and
he was aware that hackers were known for their patience and stamina.
This one seemed to be among the most indefatigable. At 1:00 in the
 morning, long after the FBI agent and the jet-lagged engineers had
grown bleary-eyed, Mitnick seemed to be just waking up. If it was true
 that he was recently married, didn't his wife wonder where he was?
    Lenny's role seemed to be a small one, though the surveillance team
 suspected he might be lying low for its benefit. Lenny's principal job that
 night was to keep an eye on the second computer terminal that the pair
 used to maintain their electronic lookout for any traces.
    Lenny had done his part. He had led them straight to the person who
134   ~    CYGERPUNK
had eluded them all these months. Kevin was doing so many illegal
things that Lenny was surprised the agents didn't storm upstairs and
arrest him on the spot. So he stole downstairs and rapped lightly on the
door of the small computer room. The door opened a crack and Harman
stuck his head out. "Wel1?" Lenny asked. "Aren't you going to arrest
him now?"
   Harman shot back a stern look. "Not just yet." Harman said they had
seen quite enough for one evening, and they were going to pack up for
the night. He ushered Lenny inside and removed the taping equipment.
   Lenny grew impatient. "What do you want? A signed confession?"
   "That would be nice." Harman smiled and gently pressed him out the
   Lenny climbed the stairs again, shaking his head.
   "What's going on? Where've you been?" asked Kevin, speaking, it
appeared, straight to the computer screen. He was back in Leeds.
   "1 was just checking to see if Hurley's still here. Listen, Kev, I'm tired
and I have to get up early in the morning. Can't we call it a night?"
   "No, I'm onto something great here."
   "Kev, come on. Let's go." Lenny was becoming exasperated.
   Kevin turned and glared at him.
   It wasn't until 3:00 A.M. that Kevin was finally ready to call it quits.
   When the two went outside, they saw that Kevin's car had a flat tire.
So Lenny offered to drive him home-after a pit stop at the Fatburger.
In a fit of generosity, Lenny told Kevin he would pay for Kevin's meal.
Surprised but not about to object, Kevin ordered two King Fatburgers
with everything on them, large fries and a large Diet Coke. Lenny
ordered fries and watched as Kevin, out of habit, systematically emptied
half a dozen ketchup packets onto his paper place mat then jabbed the
mound of sauce with his french fries. His meal dispensed with, Kevin
asked Lenny what time he should arrive at VPA the following night.
Lenny said he was busy. As usual, an argument ensued. But rather than
retread old ground, Lenny told Kevin to find his own way home and
stalked out the door.
                                   ~   T   ~

The FBI agents and Digital security experts gathered at VPA the follow-
ing morning to examine what they had collected. That morning, Har-
man and Headrick decided they had enough to make an arrest. In an
unusual move, Digital, normally intent on keeping the lowest profile
possible when it came to security matters, agreed to press charges and
                                    Kw~:   Tk 7)M/I.-~   H~     ...   135

risk public exposure. It was a daring move, for once Digital customers
learned that someone had been roaming freely throughout a network
they believed was secure, and that he not only had helped himself to
company software but, by reading people's electronic mail, had discov-
ered security weaknesses, Digital would have a lot of explaining to do.
But as far as some Digital officials were concerned, seeing this criminal
behind bars might be worth a little public disgrace. Once they had
Mitnick taken care of, they would handle the more nettlesome problem
of Lenny. Headrick went to find Lenny. "How do we find your friend to
pick him up?"
   Lenny was overjoyed. He told Headrick he knew exactly where to
find him. Kevin had just come by to have his car towed to a nearby
garage. Lenny called the garage.
   Lenny knew Kevin well enough to know that money ranked only
behind food and computers on Kevin's priority list. Lenny had lost a
$100 bet a few weeks earlier that Kevin wouldn't be able to crack one of
the electronic door codes at VPA. When Kevin punched the correct
code after just a couple of tries, Lenny accused him of cheating. He must
have seen the code scribbled on a scrap of paper inside sorneone's desk.
But Kevin not only demanded his winnings, he charged Lenny 10 per-
cent interest for each day that Lenny failed to pay him. So Lenny called
him at the garage to tell him he had just gotten some vacation pay and
he had the money for him. "I'll be there in twenty minutes," Kevin said,
and hung up.
   Headrick told Lenny he didn't want to arrest Kevin unless he had his
blue canvas duffel bag with him. Lenny had explained that the bag was
Kevin's tabernacle. It contained all his diskettes, papers and other com-
puting paraphernalia, as well as the computer-age equivalent of a lock
picker's kit-handwritten access codes and passwords Kevin used to steal
his way into heavily guarded computers from the safe remove of thou-
sands of miles. Lenny had told Headrick that the bag contained docu-
ments linking him to the stolen Digital security programs, as well as to
the abuse of the USC computer. In fact, Lenny had said, if they were
going to get any hard evidence, it would come from that bag. Kevin's
house had been searched so many times that he no longer kept anything
incriminating there.
   The material gathered from the eight hours of surveillance was help-
ful, but if the bag contained as much incriminating evidence as Lenny
claimed, it would be indispensable in building a case against Mitnick.
Lenny cheerfully suggested that he steer Kevin to the trunk of the car,
136   A   CYCERPlINK
where Kevin always kept the blue bag. As soon as Lenny saw it, he
would scratch his head.
   The signal was carefully arranged. When Lenny reached up and gave
the back of his head an absentminded scratch, they would make the
arrest. It was just after sunset, and in the gathering darkness half a dozen
FBI agents slouched under the steering wheels of their unmarked govern-
ment cars inside the small open parking garage beneath VPA.
   Just after 5:00 P.M., Kevin's black Nissan Pulsar rolled into the garage.
Luring Kevin there had been easy enough. His next challenge was to get
Kevin to open the trunk. Lenny told Kevin he needed an empty disk
from him to copy a piece of software from upstairs in VPA.
   But snaring the quarry wasn't so easy. When Kevin pulled into the
garage, he said nothing about the money. He wasn't even eager to get
out of his car. "Let's go eat," he said, apparently forgetting the argument
they had had the previous night. "I'm hungry."
   Lenny had to think fast. "But Kev, I really want to make a copy of
this great terminal emulator. Let me just run up and do it. Then we can
   "No, I wanna go eat now. Copy it later." Kevin's desire for nourish-
ment didn't surprise Lenny, but his insistence seemed unusual. This was
one test of wills Lenny planned to win. He stood his ground.
   Kevin finally agreed to get a disk for him. Grumbling, he went to the
trunk, opened it and lifted out the large blue canvas duffel bag. He put
one foot on the bumper and propped the bag on his knee. Lenny reached
up behind his head and gave his hair a tousle. Suddenly, the garage filled
with the sound of engines starting and tires screeching.
   Kevin looked around, surprised and confused. Lenny grinned broadly:
"Kev, you know that queasy feeling you get in your stomach when you
know you've been caught and you're about to get arrested?"
   Kevin looked at his friend. "Yeah?"
   "Well, get ready."
   Kevin was taken completely by surprise. The broad grin on Lenny's
face left him confounded. The FBI agents jumped out of their cars and
shouted to Kevin that he was under arrest. They demanded that Kevin
put his hands up and lean against the car. Kevin laughed a tight little
laugh. "You guys aren't from the FBI. Show me your folds." Six large
FBI identification folds emerged.
   Kevin looked at Lenny, who was dancing in little circles and laughing.
"Len, why'd you do this to me?"
   "Because you fucked me over" came Lenny's reply.
   The agents hustled Kevin into one of the cars.
   "Lenny!" Kevin cried out. "Could you call my mom and tell her I've
been arrested?"
   Ignoring the plea, Lenny turned to Chris Headrick and smiled. Head-
rick nodded approvingly. "You did so well you should be in my business."
I   .


        PART   TWO
     In   the late fall of 1986, two young men took the subway across
the border and got off at Friedrichsttasse in East Berlin. When they
arrived at passport control, Peter Carl, a dark and slightly gnomish
man in his early thirties, took over. With a businesslike flick of his
wrist, he slapped his passport down in front of the guard and said
he had an appointment. His companion, a tall and slender teen-
ager with a pale complexion who called himself Pengo, sat to one
side and waited to be cleared. As Pengo understood it, Carl had made
the initial contact a few weeks earlier by slipping a note containing
secret code inside his passport. From then on, he could enter
East Berlin any time he pleased, without exchanging the requisite
25 marks. A West Berliner usually had to apply a day in advance to
travel the few miles across the border. The guard waved the two men
   The corner where they emerged from the U-Bahn, Berlin's subway
network, was bustling by East German standards. But the elegance of
the cafes that once defined this part of Berlin had long since been
replaced by tall public buildings, their dull finish suggestive of a more
proletarian aesthetic. Carl and Pengo made their way to Alexanderplatz
to kill some time. If not exactly friends, the two were caught up in a

142   A   CYE6RPlINK
mutual adventure that bound them. They sat down on a bench. Peter
Carl produced a joint from his pocket, lit it and offered it to Pengo.
Eager as Pengo might otherwise be to do something so daring-smoke a
joint in the middle of Alexanderplatz, East Berlin, in broad daylight-
he declined. He was nervous.
   Shortly before noon, they walked the ten minutes to Leipzigerstrasse,
a wide boulevard that began at Checkpoint Charlie and flowed eastward.
The building they were headed toward-number 60 Leipzigerstrasse-
didn't differ in appearance from the dozens of other postwar apartment
buildings that lined the wide boulevard. While West Berlin had em-
braced a haphazard, extroverted approach to reconstruction, a Soviet-
inspired blueprint informed East Berlin. The dreary designs of the
fourteen-story buildings along Leipzigerstrasse were first cousins to the
high-rise apartments ringing Moscow's outskirts. But its proximity to
the West always gave East Berlin a bit of an edge over its East Bloc
brethren. Technology had been creeping into East Berlin, and the
change was especially evident on Leipzigerstrasse. There was a peculiarly
capitalist innovation-an automated teller machine-just outside num-
ber 60.
   From the names listed next to the buzzers in the entryway, the build-
ing appeared to be chiefly residential. They took an elevator the size of
a refrigerator, which jerked its way up to the fifth floor. They were
greeted at the door by a bearded, slightly stocky, well-tailored man in
his forties. He ana Carl shook hands like old friends. Then Carl gestured
to his associate. "Hier ist mein Hacker. Pengo," he announced with a
note of triumph in his voice. Sergei turned to Pengo-his real name was
Hans Hubner, but he had been better known as Pengo for years-and
extended his hand.
   The meeting was relaxed enough at the start. A secretary served
coffee. Sergei, cordial and businesslike, inquired about Pengo in accom-
plished German somewhat overwhelmed by a Russian accent. What was
the young West Berliner's background? What were his views, his poli-
tics? Pengo responded proudly that he was a product of West Berlin's
leftist scene-the sixties movement-and that he was sympathetic with
what Mikhail Gorbachev was trying to achieve in the Soviet Union.
Sergei seemed unfazed, perhaps even slightly amused. Then he got down
to business. "Do you have something for me?" he asked, turning to Peter
Carl. Carl produced a magnetic tape and diskettes from his briefcase,
handed them to Sergei, and the discussion turned serious. On the disks,
Pengo explained, was a security program for Digital Equipment Corpo-
ration computers, the computers of choice in the Soviet Union, and on
the tape Sergei would find some other interesting software.
    Barely acknowledging what he had just heard, Sergei told Pengo ex-
actly what he was interested in. Referring to what appeared to be an
order list, Sergei said he was anxious to obtain state-of-the-art engineer-
ing software-expensive and sophisticated programs that fell high on
the embargo lists intended to stop American and Western European
high-technology products from landing in Eastern Europe. He asked, for
example, if it would be possible for Pengo to obtain computer-aided-
design software for designing chips.
    Pengo deliberated for a few moments, then decided he should tell
Sergei what a hacker really did. He explained that he was capable of
traveling the world through a computer. He could hop from West Berlin
to Pasadena in a few seconds, dipping in and out of foreign computer
systems in the blink of an eye. But at the moment, Pengo complained,
his means of transmitting information were limited. He had a modem,
but it sent and received information at a paltry 120 characters per sec-
ond. The electronic theft of the kind of software that Sergei wanted
might take days at the rate Pengo would be able to gather it, making
such an exercise far too risky.
    Pengo began to make his pitch. "Of course, I could do it if I had the
right equipment"-perhaps a high-speed modem, one that would allow
him to manuever more swiftly, and a more powerful computer with
plenty of disk space. It was, after all, his dream to hack on powerful
computer systems, but he didn't mention that to Sergei.
    Pengo hesitated. Sergei was silent. The teenager continued: How
would the Russians like to set him up to do safe hacking from phone
lines that couldn't be tapped, perhaps even to do his hacking from East
Berlin? It would be to everyone's benefit.
    Pengo sensed no sympathy, nor could he see even a flicker of compre-
hension on Sergei's part. His only response was to suggest that Pengo try
to fulfill his requests, after which they could talk again. If Pengo needed
to talk in the interim, Sergei said, he should feel free to come back.
With that, the Russian invited his two visitors to take a meal with him.
If not an unqualified success for the two West Germans, the visit was in
its own way instructive: Pengo had established contact with a KGB agent
and had learned what it might take to engage in espionage for the
144   ~    CYG6RPUNK
Most eighteen-year-olds, no matter how interested in computers,
wouldn't have had the chance to talk business with a KGB agent. If
Pengo had been asked to describe how he felt about what he had just
done in political or even ethical terms, he would likely have shrugged
his bony shoulders, rolled another cigarette and passed the question off
not just as unimportant but as annoyingly beside the point. Politics and
ethics, he would say, had nothing to do with it. Hacking was the means
and the end, the information and its destination secondary. He wanted
a powerful machine to hack with. He couldn't afford one, the Russians
could, and it was the Russians who seemed interested in letting him
indulge his passion. He might even say that the purity of his purpose
struck him as somewhat heroic; his goal was to be the world's best
hacker. That he was living what he might have read in a spy novel made
it all the more exciting.
   Pengo was the son of middle-class Berliners whose lives were torn
apart in 1961 by the wall that divided their city. In 1945, five years after
his father, Gottfried Heinrich Hubner, was born, the first Western
troops marched in and Berlin became a city administered jointly by the
four Allied powers. It had been badly damaged, with only half its resi-
dents remaining and a fifth of its buildings destroyed; there was no
electricity or gas, and drinking water had to be hauled in from the
    In the late 1950s, when travel between East and West required only
the price of a subway ticket, Gottfried went to the technical university
in West Berlin to study engineering. Renate, small, blond and his high
school love, traveled frequently from East Berlin to visit him, against
strict orders from her father, an impassioned Communist Party function-
ary. Renate's trips to the west so enraged her father that he informed the
East German authorities of the nineteen-year-old girl's hostile act against
the Party. He banished her from his home. Renate continued her trips
to visit Gottfried, assuming that she could return at will despite her
father's opposition. Fate decided her future, however, during one late-
summer visit in 1961. On the day before she was to return, a line was
cut through the city's heart. The young couple went to the Brandenburg
Gate and watched East German soldiers unfurl a twenty-eight-mile-long
barbed-wire barrier. Within days, the Berlin Wall was under construc-
    Gottfried and Renate settled in Schoneberg, an older section of West
Berlin that became famous for its town hall, from the steps of which
John F. Kennedy delivered his famed "Ich bin ein Berliner" speech.
Renate was studying graphic design and Gottfried was just starting out
as a construction engineer. Both were pulled into the anti-authoritarian
movement, the European equivalent of the American Left's countercul-
ture that was sweeping through Europe. They married when Renate got
pregnant, and Renate dropped her studies to become a full-time mother.
   Hans Heinrich was born in July 1968 into a tumultuous time. Three
months earlier, Rudi Dutschke, a brilliant sociology student and pacifist
who had founded an opposition party in West Berlin, had been shot and
nearly killed outside his apartment on Kurfurstendamm, West Berlin's
main boulevard. Dutschke was shot by a housepainter who couldn't
stand what "Communist pigs" like Dutschke represented. The incident
triggered mass riots in the city, and the unrest in Berlin inspired student
uprisings throughout West Germany.
   Hans tended more than most children to give free rein to his fantasies.
As a young child, he used the rubble of a partially reconstructed Berlin
as an endlessly fascinating playground. Hitler's massive concrete fortifi-
cations, scattered throughout the city, were indestructible, so architects
learned to design their way around them, occasionally just building new
apartments on top of them. The abandoned bunkers beneath were ap-
propriated by the child explorers of Berlin, and Schoneberg had one of
the best. It stretched for half a block, stood several stories high and
reached four stories below-ground into a catacomb that was filled with
endless stretches of ramps and passages. Oblivious to the historical sig-
nificance of their romping place, neighborhood children scouted its ex-
panse for old helmets, uniforms and other remnants. The building was a
real-life fantasy world, a secret domain perfect for role-playing adven-
tures; perhaps it was even a first taste of the kinds of interlocking paths
and channels that computer networks would come to offer.
   Partly out of boredom at home, Renate joined a group of mothers who
were exploring alternative education for their children. Every day after
kindergarten the five-year-old Hans went to a Kinderladen, an experi-
mental preschool not unlike the Montessori schools in the United
States. Parents of Kinderladen children generally belonged to the anti-
authoritarian movement of the time and believed that their children
should decide for themselves what they would eat, when they would
learn to eat with utensils and even when their diapers needed changing.
Parents like Gottfried and Renate favored uninhibited self-expression in
their children, encouraging them to resolve conflicts by themselves and
to assert themselves in ways that had been forbidden to their parents.
   Yet the Hubner family was drifting apart. Hans was still in school
146   A   CygE~Pl1NI<

when his father and mother divorced. Hans stayed with Renate, and his
younger brother, Ferdinand, moved in with Gottfried. Renate went to
work as a dental technician building ceramic teeth, and Hans joined the
ranks of latchkey children.
   He wasn't much of a student. At ten, he was an awkward, slightly
plump, bespectacled kid who showed little aptitude for penmanship,
English or art, and none at all for sports. His performance wasn't en-
hanced by the breakup of his family. His teachers regularly sent home
reports stressing Hans's potential, while complaining that he was lazy
and disruptive. The only subjects in which he excelled were math and
physics. When report cards were issued and Hans called Renate with the
news, she would tell him to forget the grades and just read her the
written remarks.
   But Hans had a charm that tended to obscure his failings. From an
early age, he displayed a sophisticated, relaxed and dry sense of humor.
Whenever he got into trouble, he was able to talk his way out of punish-
ment, inspiring awe among his more academically diligent friends. His
teachers decided that what the young charmer needed was more intellec-
tual stimulation: they decided to advance him by having him skip the
eighth grade. But this backfired. The youngster did not throw himself
into schoolwork in a more mature surrounding. Instead, he withdrew
into his own world, alienated from the classmates for whom he felt little
   When he was twelve, Hans discovered squatting. Squatting in Berlin
started in the 1970s when artists, punks and runaway teenagers laid claim
to abandoned industrial buildings, apartment houses and stores. By
1980, there were more than one hundred buildings in West Berlin oc-
cupied by squatters. Hans joined a group that took over an abandoned
store in Schoneberg, where they set up a punk band. What the adoles-
cent lacked in musical talent he made up in aesthetic sensibility: his six-
inch spikes of hair dyed jet black soared above his head with the help of
copious applications of soap. He completed the costume with black mil-
itary boots and a heavy chain around his hips.
   He was now growing quickly, and sharp angles were replacing the soft
folds of childhood. The wild adolescent distressed his parents. First there
were incidents of shoplifting-a cola, for instance, or the electronic
parts to build an amplifier. When Renate went to the police precinct to
retrieve Hans after the cola episode, officers displayed pity mixed with
disgust at her for having borne such a son. She was so enraged that she
slapped Hans across the face, but she succeeded only in stinging her own
palm. And then there was the time Gottfried was called to the hospital
to fetch his barely conscious son, who had gone to a rock concert, drunk
too much and smoked his first appreciable quantity of marijuana.

Personal computers arrived in Germany in the early 1980s, mainly by
way of England. The first wave were toys that went into the bedrooms
of teenage experimenters, children of middle-class homes in trim and
tidy neighborhoods in antiseptic towns across West Germany. Parents
gladly paid the money for an Atari or a Commodore 64, toys that seemed
capable of serving an educational function, rather than see their adoles-
cent children experimenting with drugs or joining the punk scene.
    Hans was destined to become an electronics freak. In 1982, Sven, a
friend since first grade, borrowed a hand-held computer the size of a large
paperback from someone at his school and took it home. Hans began to
program it immediately, as if he had been programming all his life.
Though a miserable English student, he had learned enough to write his
first BASIC program, a loop that produces an infinite number of hellos:

                       10 print "hello"
                       20 9oto 10

 It was a rudimentary program, to be sure. Later Hans was to develop an
 elegant self-taught programming style. He had a logical mind, stimulated
 by the discovery of an ability to create something from nothing. He was
 transfixed. Squatting in abandoned houses had its pull, and spending
 time with girls was nice enough, but here was something that could
 command Hans's attention as nothing ever had.
- It wasn't long before the hand-held computer yielded to mail-order
 kits for bigger computers that had to be soldered together. As their first
 big project Hans and Sven put together an entire computer-a Sinclair.
 Until the advent of the Sinclair, personal computers, even those assem-
 bled from kits, were out of the reach of most young pockets. But now,
 for $250, Sven and Hans could buy a complete computer. Even measured
 against the standards of the day, the Sinclair was exceedingly primitive.
 At that time Apple Computer, Commodore and IBM were already on
 the market with machines that look similar to what people now use. But
 all that the two young Berliners could afford was a kit that, once assem-
 bled, resembled a small box of chocolates plugged into a television set.
 The software for the Sinclair came on tape cassettes, not on the floppy
148   A   CygfRPUNK
disks of today. The machine had very limited memory, but it was enough
to write a program or two. For Hans, the Sinclair provided a taste of the
hidden power of the machine, of binary worlds to be explored.
   Even as the computer lay in pieces in his room, waiting to be assem-
bled, Hans was getting ready to start programming it. He had taken the
programming manual with him on a vacation to southern Italy that
summer with his father and his father's girlfriend. Culture just annoyed
him, and if the two adults coaxed him into accompanying them to
museums and ancient churches, he sulked. Finally, they gave up and left
him in the rented villa, inseparable from the book.
   When Hans returned from Italy, the two budding computer experts
spent day after day in Hans's room, a dark little chamber covered with
graffiti, putting together the Sinclair. Oddly enough, although Hans had
never done much with his hands, and wouldn't have known what to do
with a hammer, his instincts for putting together the computer were
perfect. But after he had put it together, his desire for a better computer
was immediate. As soon as he could, Hans ordered a more powerful
model, the Spectrum. It had the same design as the Sinclair, but twice
the internal memory.
   Circumventing rules became an early obsession for Hans. His main
activity in programming was piracy-he attempted to bypass the copy-
right restrictions on the commercial software cassettes, mostly games.
Hans wrote little programs that allowed him to load a game into the
computer, save it and put it onto another cassette. It was a challenge to
the young hacker, not an ethical question. If the publishers tried to build
anticopying devices into their programs, they just had to expect hackers
to try to unravel them.
   Hans and Sven turned into computer-game freaks. Their favorites
were those with good graphics, arcade-style games that involved some
relatively simple task, such as shooting lasers at successive waves of bug-
eyed aliens. These electronic games really came to life when played on
the colorful, large screens at a video arcade. Hans was spending much of
his free time at the video arcade around the corner from his school. The
arcade was run by a shifty character who hired teenage boys willing to
work in exchange for the privilege of communing with the electronic
gadgetry. For doing minor repairs on the games, renting out videos
(mostly pornographic) and generally watching over things, Hans had the
run of the place. He also set up something of a pirate's copy center,
making illegal copies of Sinclair Spectrum software for all his friends.
   The arcade also provided the setting for his first true hack. Hans
discovered that the small strikers used to light gas stoves created the
same effect, when their caps were removed, as coins dropping into slots.
One electric spark from the striker next to the coin slot of a video game
and twenty free plays appeared out of nowhere.
   The lighter enabled Hans to spend hours exploring the various games
in the arcade. His favorite was Pengo. It was simple and addictive. The
player assumed the role of Pengo, a penguin who pushed ice blocks up
and down and back and forth on the screen, aiming them at malicious
monsters called suo-bees. The trick lay in destroying the sno-bees before
they destroyed the ice blocks and left the heroic little penguin defense-
less. Hans battled the sno-bee enemies for hours on end, oblivious to the
passage of time. Often he stayed at the arcade through the night and
stole into his bed at 6:00 A. M., half an hour before his mother got up to
go to work. At 7:00, she would tap lightly on his door and tell him to
get ready for school. He got up, got dressed and headed straight back
to the arcade. The ruse worked until the police visited Renate one day
to tell her that not only was her son a chronic truant but he spent all his
time at a video arcade, some of it lending out pornographic videos.
   Hans was soon to stumble into an even more compelling world. His
introduction to computer communications came in the person of Barnim
Dzvillo. A friend from school, Barnim had a Commodore 64 with an
acoustic coupler modem. Even though Hans disliked the Commodore 64
because it bespoke an affluence that Sinclair owners despised by reflex,
Barnim's modem was an irresistible enticement. Hans had first seen such
a device when he watched his physics teacher use one to call the school's
central computer to retrieve class schedules. But he had never used one.
   One evening Hans went to Barnim's house and Barnim showed him
how to use the modem to dial into an electronic bulletin board in West
Berlin. Before logging on, Hans had to choose a name for himself, a
handle. Pengo was the obvious choice. By the end of the evening, Hans
was Pengo, and Pengo was hooked.
   Barnim also gave Pengo his first taste of electronic adventure by show-
ing him how to sneak electronically into a mainframe computer by going
through T ymnet, a commercial data network owned by the McDonnell
Douglas Corporation with dial-up ports (computer access points) that
touched down in more than seventy countries. A computer user with a
modem can call the number of a local computer, then by typing the
correct address, connect to another computer anywhere in the world.
However, networks such as Tymnet have an unavoidable security flaw:
they publish their telephone numbers widely, making it possible for
150   A   CYE6RPLINK
unauthorized outsiders to get onto the network with one local telephone
call, then hop to computers attached to the network. To log on to the
network, all one needs is a user identification and a password. Soon
Pengo discovered 3M, the American manufacturer of Scotch tape. To
make communications easier for customers at 3M, the company had set
up a help system accessible from Tvmnet. Typing "3M" and "Welcome"
established Pengo's connection to the company.
   Attached to his Commodore 64, Bamim had a primitive modem that
was excruciatingly slow, retrieving just 300 bits of information-or
about 30 characters-every second. At this rate, it would take almost a
minute to read a single page of text being transmitted by a remote
computer. But it didn't matter to Pengo that it was slow. It seemed that
the whole world was opening up to him. Pengo and Bamim began to
wade through screen upon screen of information. Mostly what they saw
was a detailed description of other 3M computers they could reach-
computers in West Germany, France, Great Britain, Mexico and Chile.
They stored the information on a floppy disk and printed it all out.
   Being inside a global computer network gave Pengo the same adrena-
line rush that playing a video game did, magnified many times over.
Once he had entered a computer network, he was no longer playing a
game; he was master of real machines performing real tasks. He could be
nowhere and everywhere at the same time. From in front of a computer
screen, he could open doors and solve problems. He was able to get
things. The stuff on the screen wasn't anything more than electrons
hitting phosphor, but it was nice to imagine that there really was some-
thing there, around the back somewhere. The networks-unconstrained
by conventional geographic boundaries-had become a self-contained
universe known to a growing number of computer researchers as cyber-
space. It was inevitable that a young nonconformist like Pengo would
take up residence there. A world that his parents didn't even understand
well enough to forbid was an irresistible outlet.
   Pengo had to have a modem of his own, so he went straight out and
bought a kit. Unfortunately, his homemade device was primitive and
required intensive manual labor to operate. Whenever he reached a
remote computer and received a special connect tone, Pengo hooked his
acoustic coupler to the telephone and a set of loudspeakers and tuned
the coupler like a radio in order to match the connect tone. The modem
worked well for only a couple of hours before its printed circuit board
heated up, and Pengo had to unhook it and hang it out the window for
a few minutes to cool it down.
                                ... T ...

NUl, pronounced "NOO-ey," stands for "network user identification."
It was other people's NUls that sustained many a West German hacker's
career. A NUl opened the door to Datex-P, West Germany's govern-
ment-controlled computer data network. Once you were into Datex-P,
you were on a bridge into the equivalent T ymnet network, and from
there into thousands of computers throughout the United States. Unlike
private network services such as T ymnet and T elenet in the United
States, European data networks are usually run by the government. The
West German government, in particular, regulated all communications
very closely. The West German postal and telephone administration,
the Bundespost, owned and controlled the Datex-P network and issued
all NUls. Modems and answering machines had to be registered with
the Bundespost. Installing your own equipment without registering it
with the Bundespost meant inviting a hefty fine. None of these restric-
tions existed in the U.S. or even in most other European countries. The
Bundespost argued that the heavy regulation was necessary to protect
the integrity of the telephone network. But in a way, the conservative
Bundespost was asking for trouble. It was little wonder that smart young
rebels should want to skirt the system, to assert themselves by doing
something as relatively harmless as spurning the slow and outmoded
state-approved modems in favor of speedier brands, or even as brash as
wandering into computers where they had no right to be. And it wasn't
surprising that West German hackers made a sport out of stealing NUls.
   In the mid-1980s, most stolen NUls were obtained at the Hannover
Fair every April. Stealing them was simple. It was just a matter of
looking over a preoccupied exhibitor's shoulder as he typed in his NUl
and password while logging on to a remote computer. Stolen NUls
quickly made the rounds of the West German hacker community.
   In the mid-1980s, when Pengo began to break into systems in earnest,
many were invitingly easy to enter. Digital Equipment Corporation
equipped each VAX computer it shipped with three built-in accounts,
each with its own password. There was the "System" account, with
Manager as the password; the "Field" account with Service for a password;
and the "User" account, whose password was, conveniently enough,
User. It was up to customers to disable those accounts and passwords
once they had established their own. Often they didn't, leaving a back
door open to anyone who happened along. Once a single computer was
breached, if the computer was linked to others, network "cruises" were
152   ..   CYGERPl1NK
next. If the computer was on more than one network, it could be used
as a springboard into many other computers on other networks. Network
hopping gave Pengo a certain amount of protection from pursuers who
would have to trace him back through many connections and different
time zones.
    Breaching computers combined the challenge of solving a giant puzzle
with the thrill of breaking the law. Pengo would come home in the
afternoon, turn on his computer, find some others to "chat" with on
electronic bulletin boards and trade some information, perhaps a NUl
in exchange for a way to break into protected files. Then, his mind
racing around the world while his body remained in Berlin he would
start to probe the network, sometimes with a specific target in mind,
sometimes without. The farther away the computer, the better. And
seeing that fifty users were logged on to the machine he reached gave
rise to a special thrill of being inside something extremely large and
    There was a lot for Pengo to familiarize himself with, most of which
he taught himself. He figured out how to write scanning programs, au-
tomatic dialers that were programmed to run through the night, dialing
number after number in search of a high-pitched modem tone on the
other end. And he learned how to probe, ever so tentatively, inside a
system once he had broken in. Sometimes he couldn't resist trying to
engage a system manager in a conversation. Such was the case in early
1985 at SLAC, a high-energy-physics research center at Stanford, which
Pengo reached one day at 4:00 A.M. through a Tymnet connection. He
got into a chat with a system manager there, who seemed happy to
exchange pleasantries with a German hacker. But not long after that
first friendly chat, Pengo encountered a second manager on the same
system at SLAC who told him to scram. Pengo said he had no intention
of going away, and if they tried to get him off the system he would crash
it. When the manager replied that that was nonsense, Pengo carried out
his threat. He wrote a small recursive program and sent it to the com-
puter to execute. The program worked like a chain letter. It created two
copies of itself, each of which, when executed, created two copies, and
so on. This was one sure way to exhaust the computer's resources
quickly. In less than a minute, overwhelmed by the flood of work to be
done, the computer was gridlocked. In the vague ethical code Pengo had
fashioned for himself and his hacking, he knew he had just crossed the
line. On the other hand, he had felt compelled to be true to his word.
Moreover, crashing a computer at a big and important place like SLAC
was something to tell others about.
   This was electronic machismo for Pengo and the others he met on
line, hackers with handles like Frimp and Task and T reppex and Zom-
bie. On the one hand, hacking could be a solitary sport. But at the same
time, once a breakthrough was made, once a huge Digital VAX com-
puter at a research laboratory or company somewhere in the electronic
universe was penetrated for the first time by one of the West Germans,
it was an occasion to strut. Telling others of the hack would earn points,
but it would also oblige the successful hacker to share the password. And
that carried with it the risk that dozens would descend on the computer
in question. Soon after that, whoever was running the computer would
usually lock the intruders out by changing passwords. But it was difficult
to resist sharing a triumph with others who would appreciate the signifi-
cance of the deed.
   Pengo was one of the first to penetrate CERN, the high-energy-phys-
ics laboratory near Geneva. A consortium of researchers from fourteen
nations, CERN was locked in a race with such U.S. laboratories as
SLAC and Fermilab in Chicago to discover the top quark, one of the
fundamental entities of which nuclear particles are made. To that end,
CERN was planning the construction of a mammoth accelerator, sev-
enteen miles in circumference, a ring passing under villages, farms and
the nearby Jura Mountains. CERN was in a difficult position. Not only
did the hundreds of scientists at the laboratory have to grapple with a
confusing array of languages and cultural gaps, but their work had to be
accessible to thousands of researchers outside the facility. Sharing infor-
mation quickly and spontaneously lay at the heart of the consortium's
charter. Closing down the network connections to protect against some
electronic delinquents was out of the question.
   Soon after the first electronic intruders discovered CERN, its comput-
ers were inundated. The laboratory became the place to post messages
and to loiter for hours at a time, waiting for friends to log on. CERN
was also an effective springboard to other centers, since it had computer
connections all over the world.
   Different system managers had different reactions to break-ins. Some
were even friendly. One manager at a European research lab, who sub-
scribed to the notion that hackers might even serve a useful purpose
opened his system to a West German teenager who offered to identify
security weaknesses. Others didn't mind getting into the occasional chat
154   &    CYEERPllNK
with an uninvited visitor. The following conversation took place in 1985
between a hacker from Hamburg and an operations manager at CERN:

VXOMEG: :SYSTEM hallo HUHU. Do you have a
l i t t l e time???
VXCRNA: :OPS Yes, we're back from dinner
VXOMEG: :SYSTEM dinner? You think about
eating all the time, too?
VXCRNA: :OPS we're here from 1600 to 2300 and
we have to eat~ ~ ~
VXOMEG::SYSTEM Of course I'm always hungry
for more VAXEN ~ ~ I I HAHAHA
VXCRNA: :OPS Why the VAXEN at CERN and not a
VXOMEG: :SYSTEM No~ I'm a hacker and no
criminal or industrial spy~~ ~ ~
VXCRNA: :OPS Yes I understand.       but isn't
i t more fun with a bank?
VXOMEG: :SYSTEM A bank is ok but i t has to be
a databank~~~

   But mostly, system managers just got annoyed; they had better things
to do than chase juvenile trespassers out of their computers. In the
middle of 1985, Alan Silverman, a CERN system manager, posted a
message warning colleagues of the hackers inside the system. He urged
his colleagues to avoid obvious passwords and not to keep password lists
sitting in easily accessible files. Silverman pointed out that security bugs
in the system had enabled a lot of hackers to get in, and he urged his
people to eliminate the bugs.
   To explore the computers at CERN, it was necessary to have some
working knowledge of VMS, the operating system running on most of
CERN's computers, especially those connected to the network. The very
features that made VMS so attractive to the tens of thousands of Digital
Equipment Corporation users made it attractive to hackers too. Digital
promoted VMS vigorously as its operating system of choice for the VAX,
the company's most powerful computer. The smallest VAX could take
care of all the computing needs of a small business while the Sinclairs
and Commodores that Pengo and his friends had access to were good
only for games.
   Although the VMS operating system couldn't be used, for instance,
on an IBM computer, it was the most common operating system in the
Digital environment, running on any VAX, from the smallest to the
largest. VAX VMS could be very solicitous of unskilled commercial
users. For example, VMS had a help facility, obviating much of the need
for manuals. For intruders, the generous help feature was yet another
way of exploring a system's possibilities.
   By the end of 1984, sixteen-year-old Pengo was paying scant attention
to his schoolwork and even less attention to his school friends. He left
his job at the video arcade and spent more and more time by himself,
with his computer and his modem. Pengo's mother was puzzled by what
he was doing until dawn, and completely mystified by his constant use
of the telephone. But unlike many parents, Renate was determined not
to interfere with her child's interests. She didn't ask him for an expla-
nation and he didn't offer one. He didn't appear to be making long-
distance phone calls, so she couldn't complain about the bill. In West
Berlin, one call of unlimited duration to the local Datex-P access point
cost about fifteen cents. Pengo's computer friends were coming over
more frequently and working on the computer well into the night. Al-
though they tried to keep quiet, the close quarters made it all the more
difficult for Renate to ignore the mysterious goings-on in the next room.
She needed sleep. But like many parents, she was glad her child was
becoming computer-literate. Parents all over the world believed educa-
tors who argued that every child should learn to use computers at the
earliest possible age. Renate and Gottfried weren't the only parents who
didn't stop to think about some of the less attractive consequences
of computer literacy. For adventurous children with little respect for
rules, computer literacy was likely to lead to electronic juvenile delin-
quency-computer trespassing. So that Pengo could continue his hobby,
Renate and Gottfried exchanged sons and Hans moved into a private
room in the back of his father's large apartment across town.
   Gottfried's only ground rule was that the phone remain free until 8:00
P.M., so Pengo began his network excursions after 8:00 and kept them
up until his father appeared at his door at midnight to tell him to go to
bed. The rap on the door signaled his father's bedtime and another six
hours of uninterrupted work, punctuated by occasional trips to his fa-
ther's refrigerator. During intensive sessions, Pengo's preferred nourish-
ment was yogurt, not so much for its healthful aspects as for the fact that
it was a clean and efficient food with no crumbs that could fall into the
crevices of the keyboard. His headphones affixed to his ears, Pengo
tapped the keyboard to the strains of Kraftwerk, the West German syn-
156   A   CYE6RPlINK
thesizer band that sang paeans to pocket calculators and home computers
in the monotone of automatons. He would listen to one album for weeks,
until he couldn't stand it any longer.
   It was during one of Pengo's all-night sessions that he met up with
   Obelix was a round-faced nineteen-year-old from Hamburg with a
skittish edge. He had discovered computers when he was about fifteen.
He began programming on a Commodore VIC 20 in physics class and
before long he was better than his teacher. Games didn't interest him
much; it was the idea of exploring the forbidden territory of computer
networks that was really captivating.
   Pengo and Obelix met electronically in late 1985, when the two
happened to be logged on to a computer somewhere in Munich at the
same time. Obelix introduced himself as a hacker from the Chaos Com-
puter Club in Hamburg. Chaos members had become notorious as the
"VAXbusters. " Pengo had already heard of Chaos, and the Chaos hacker
and his new friend struck up an electronic conversation. Obelix invited
the seventeen-year-old stranger from Berlin to the upcoming second
annual Chaos meeting in Hamburg. After barely passing his final high
school exams-to the astonishment of Gottfried who had predicted utter
failure-Pengo headed for the Chaos meeting.

West Germans came late to computer hacking, in the early 1980s. To
the general public in Europe hackers were the perpetrators of computer
break-ins and they had little in common with the original hackers from
the Massachusetts Institute of Technology in the 1960s and 1970s, for
whom hacking meant extremely clever computer programming that ex-
tended the limits of technology.
   The Chaos Computer Club was formed in 1984 by Hewart Holland-
Moritz, an elfin, thirty-two-year-old computer programmer and fervent
radical with a boisterous sense of humor who shortened his name to Wau
(pronounced "vow") Holland for its simplicity. He chose the name
Chaos, on the other hand, for its pure shock value. In reality, his club
was the very picture of meticulous organization, with a hierarchy of
officers and subofficers, some of whom came to fall just short of celebrity
status. Wau was soon joined by Steffen Wernery, a jittery and intense
twenty-two-year-old high school dropout full of bravado and a natural
ability to organize. Like a seasoned PR man, Steffen could be articulate
and outspoken without saying anything of substance. When it came to
difficult questions, he deferred to Wau.
   Wau's approach to computers and to hacking was, he claimed, driven
by ideology. A crusader for privacy rights, he had written articles for the
left-wing Berlin newspaper Tageszeitung criticizing West Germany's con-
troversial 1983 census as an official conspiracy to pry into people's lives.
He also attacked as threats to privacy the computer-readable identifica-
tion cards that the West German government was developing, and com-
puter data-gat.herin g systems deSigned~to sniff o.u terrorists. On its face,
                                                   . /t
Wau's defense of privacy seems to h ve conflicted sharply with his ad-
herence to the Hacker Ethic, the wE1 l-established, if informal, code of
conduct observed by earlier generations of computer hackers: not only
should all information be free, but aCJ3essto computers and the infer-

mation they contain should be unlimited.
   As Wau saw it, however, there was no tension between his theory
and practice. Like a lot of computer people, Wau was a libertarian when
it came to these issues. That is, he believed government and other large
institutions sh~t,!ldtl'tbe meddling in people's personal lives, but individ-
uals should be able to get at anything. The systems Wau hacked into
were owned arid operated by authoritarian institutions, forces that hardly
championed an individual citizen's right to privacy. Hackers like Wau
thought of themselves as modern Robin Hoods. Through hacking, they
could expose gaping holes in computer security and heighten public
awareness of security loopholes. Paradoxically perhaps, the club was
trying to appeal to both the West German love of law and order and a
concern for civil liberties. If he broke into computers, Wau argued, it
was to make the point that West German institutions were wrong if they
thought their computers were safe from outside meddling. And it was
with the 1984 hack of the Bildschirmtext system that Wau sought to
prove his point.
    Bildschirmtext was an electronic information service that was sup-
posed to make the computer terminal an indispensable fixture in every
West German home, third only to the telephone and television. Simple
telephone lines would link home terminals with huge data banks offering
information and services for sale. In 1984, the Bundespost made bold
predictions that by 1985 one million West Germans would be using
Bildschirmtext, of Btx for short, for everything from calling up train
schedules to ordering opera tickets. But it was expensive for consumers,
and for those not familiar with computers it was intimidating. Getting
158    .&   CYE6RPlINK
 comfortable with the Btx electronic marketplace meant getting comfort-
 able with technology as a routine part of life. West Germans carried on
 an uneasy relationship with technology. It could have been the memory
 of the horrifying consequences of Germany's superior military technology
 in World War II, or it could just have been the same ambivalence that
 everyone in the industrialized nations felt toward technology. So a na-
 tional queasiness about the information age had a habit of emerging in
 ways both subtle and obvious. For some West Germans, computerization
 carried the specter of lost jobs; for others, it raised new and unsettling
  threats to Datenschutz, or data protection. West German citizens hardly
 welcomed new ways of gathering and interpreting ever more information
  about them. It carne as little surprise to most West Germans when, by
  1988, just 120,000 households in the Federal Republic had signed up for
 Btx. Eventually, the Federal Republic would concede that its $450 mil-
 lion investment in Bildschirmtext had been an extravagant failure.
"'"""..,Wau and Steffen were convinced that the Btx service was far from
 foolproof, and they used the Hamburger Sparkasse, Hamburg's largest
 bank, to prove it in 1984. The trick was simple: once the two had gotten
 their hands on the bank's Btx identification code and password, they set
 up an automatic dialer and programmed it to call the Chaos club's $6-
 per-call Btx information service continuously through the night-charg-
  ing the calls to the bank. By night's end the bank's bill had come to
 $81,000, but Wau and Steffen didn't try to collect the money; instead,
  they held a press conference to announce their coup. The bank stunt
  brought the fledgling club instant notoriety. Germans became convinced
  that their bank accounts were helpless victims in the hands of electronic
  hoodlums. Their children signed up with Chaos.

Pengo's first Chaos conference was an enlightening experience. Seeing
so many kindred spirits in one place was like being released from solitary
confinement. The most remarkable part of going to the Chaos congress
in the Hamburg suburb of Eidelstedt was staying up through the night
and hacking with others. Nests of cables connected computers to mo-
dems and modems to telephones. Mattresses were strewn in random
rooms throughout the building to accommodate waning stamina. Wau
took seriously the possibility that the authorities might disrupt the meet-
ing, so he stationed guards and metal detectors at the door. Outsiders
took an immediate interest in Chaos. They viewed the ~lub as a symbol
of harmless dissent in West Germany. Chaos seemed the very picture of
clean fun when compared to the dread Red Army Faction, terrorists who
advocated violence as the only effective instrument of change, or Ber-
lin's Autonomen, outlaw punks specializing in a less directed brand of
violence, or even the young Neo-Nazis, who signaled that Germany
could be in danger of losing its collective memory. Chaos welcomed the
attention, using any opportunity to hold a press conference, and the
1985 Hamburg congress was no exception. Newspaper reporters and
television crews from throughout Germany, filmmakers from England
and sociologists from Berlin's Free University nearly outnumbered the
computer enthusiasts. German television crews stalked the conference,
meters of cable snaking behind them. The nightly news carried reports
of the latest gathering of technological wunderkindef;',,;"
    Part of it might have been simple one-upmanship, part of it might
have had something to do with the dynamics of the group, but it was
during the Hamburg conferences that some of the most daring feats took
place. At one point, for instance, Pengo and a small group of others
slipped electronically into a development computer for the Ottawa Po-
lice Force; they stayed on the system for several hours, looking around
what seemed to be a data-base program for criminal searches, and
stopped only when a system manager discovered strange activity on one
of the accounts and shut it down.
    At age seventeen, Pengo cut a striking figure. The rangy young man
was dressed entirely in standard-issue Berlin black to match his jet-black
dyed hair, chic Girbaud pants billowing at his thighs. He quickly estab-
lished his credibility. As the others looked on, he broke into a Digital
Equipment computer somewhere in the U.S. and, by writing a small
program in DCL, a programming language used for automating common
command sequences, he built a bulletin board specifically for congress
attendees. People sitting two feet apart in the cramped room in Hamburg
could log on to the computer overseas and have a conversation. It was
an impressive feat, and it was clear to the onlookers that they had a true
talent in their midst.
    On a more personal level, Pengo found that despite their electronic
rapport, once they met face to face he and the butterball Obelix had
little in common beyond their computer fixation. Though both teenagers
came from middle-class families, Obelix cultivated many of the values
that Pengo had come to scorn. Obelix was a budding conservative with
strong anti-Communist sentiments and a desire for wealth that eclipsed
his other traits. He squired friends and visiting hackers around town in
his mother's Mercedes, claiming that it was his. His role models were
160 ...    CYE£RPLINK
Malcolm Forbes and Heinz Nixdorf, two avowed capitalists from rela-
tively humble beginnings. Obelix had every intention of becoming a
self-made millionaire and was forever coming up with electronic inven-
tions, each the basis for a new company that would flourish overnight.
Pengo didn't take many of Obelix's get-rich schemes seriously, but he
did take seriously Obelix's love of things technological. When he first
arrived at Obelix's house in Hamburg, Pengo saw that Obelix owned a
1200-baud modem, which was capable of retrieving more than five and
a half pages of typed text each minute. In Pengo's eyes such a device was
certainly state-of-the-. a~~nd it won his immediate respect. ~.

Pengo had wanted to meet Wau, and though he decided Chaos's founder
was interesting, his age and politicized discourse made it hard for the
teenager to relate to him. But then he met Karl Koch, a tall, wiry
twenty-year-old from Hannover with a gaunt and troubled face. When
Karl first introduced himself, he opened his laptop computer and pulled
from its battery compartment a brick-sized chunk of hash. The consump-
tion of copious drugs was just one aspect of Karl Koch's life. His preferred
name, he explained to Pengo, was Hagbard Celine.
   Hagbard told Pengo he had come to hacking by way of The Illuminatus!
Trilogy by Robert Shea and Robert Anton Wilson, an eight-hundred-
page conspiracy-theory novel whose hero, one Captain Hagbard Celine,
is out to fight the powerful secret society, the Illuminati. The Illuminati,
dating back to eleventh-century Islam, ruled the world. In the book,
Hagbard Celine infiltrates the Illuminati in order to defeat them. Not
only did the Hagbard living in Hannover in 1985 believe the Illuminati
were still around and responsible for plotting to rub out everyone from
John F. Kennedy to Ian Fleming, he believed that he was in fact Hagbard
Celine and that his mission was to stop the Illuminati from mastermind-
ing a nuclear holocaust. Hagbard of Hannover was convinced that the
world's computer networks were a conspiratorial ruse, and that the peo-
ple in charge of them were, accordingly, running the world. Hagbard's
idea was to insinuate his way into the computer matrix and carry out
electronic subversion.
   Hagbard didn't feel a need to learn how to program. He could work
just as well without doing so, as long as he had the right telephone
numbers and passwords. He would leave it to the others to write the
fancy little programs for collecting passwords and escaping detection.
    Pengo sensed a deep intuition in Hagbard. He admired an imagination
that could follow such a steady course on such an outlandish trajectory.
Pengo used computer equipment to patch into cyberspace but Hagbard
truly lived there. Pengo identified with his own father-cynical with a
logical, mechanistic approach to things-and he knew that explained
his fascination with electronics. He understood that Hagbard's Illumi-
natus theories were nothing more than the construct of a troubled mind.
Even so, their fanciful and poetic qualities were exactly what Pengo
liked and envied in the young man. He knew that the drugs played no
small role in warping Hagbard's perceptions, but at the same time they
enhanced Hagbard's already fertile imagination.
   Hagbard and Pengo took the train back to Hannover together and
during the ride Hagbard told Pengo that if he were ever to commit
suicide he would build an atomic bomb, climb to the top of the World
Trade Center in New York City and set it off. His tone was so serious
that Pengo felt he meant it. Entranced, he wanted to find out more
about this charismatic, deranged new friend.
   Karl Koch's family, it appeared, had come unglued early in his life.
His father had left Karl and his sister alone with their mother when Karl
was small. When his mother developed cancer, it was the young Karl
who watched her die. Karl's father, a high-profile journalist for a Han-
nover newspaper and a heavy drinker, also died of cancer when Karl was
sixteen. His inheritance of 100,000 marks, or about $50,000, was
enough for him to buy a revamped Porsche, rent a nice apartment in
Hannover and support an expensive drug habit. In high school, Karl had
been something of an activist. Like a lot of other West German students,
he had been an outspoken critic of nuclear power. But after school, and
after his father's death, hash and LSD shunted politics aside. After
reading Illuminatus!, he was only Hagbard Celine.
   Hagbard took a quick liking to the hacker from Berlin who was willing
to smoke joint after joint with him and show him new computer tricks.
Hagbard was bisexual, and he asked Pengo if he was as well or, if not, if
he was at least interested in experimenting. When Pengo declined,
Hagbard accepted his decision graciously. It flattered and intrigued
Pengo to have such an avid admirer. Pengo had believed himself to be
on the fringes of life in West Berlin, a technological rebel in a city
famous for its rebels. But in Hagbard, Pengo saw a character far closer to
the edge than any he had ever met and far closer than Pengo had ever
gone himself, even in the shoplifting days of his youth or the hours he
had spent hanging around the sleazy arcade. Pengo tagged along with
Hagbard, happy to meet the small but close circle of Hannover hackers
162   A   CygERPUNK
that called itself Leitstelle 511, after the area code for Hannover. They
met weekly at a bar called Casa Bistro for a hackers' Scammtisch-a group
gathering that takes place regularly at the same table at the same bar or
restaurant. These new acquaintances were a world away from what
Pengo had come to know of the hackers he had met at the Chaos
meeting in Hamburg, and light-years from the likes of his friends back
in Berlin.
   Among those young people who grew up surrounded by the reborn
opulence of Western cities such as Munich and Hamburg, a sizable
percentage demonstrated considerable ignorance about the "other" Ger-
many, East Germany. Many West Germans regarded East Germany as a
foreign country. Interest in the history that confronted West Berliners
at the end of every road was barely perceptible among young people from
the rest of West Germany. World War II had become grandfather's war,
not father's war. A sanitized version of history had been taught to the
generation born directly after the war. By the time a new generation of
teachers came along from the sixties movement, insisting that the his-
tory of the Third Reich be taught in West German schools, indifference
had taken hold among the students. They had heard too much, and they
wanted to close the book on it. The passage of time had left young
people, if not with a subconscious tendency to explain away Hitler's
crimes, then with a curious blank spot. While Pengo, like most Ber-
liners, had contempt for the self-satisfied myopia of West Germans, his
obliviousness to the political or legal consequences of what he did was a
curious contradiction in his character. Perhaps his being exempt from
the draft, as were all other West Berliners, had helped create his political
blind spot-politics had never intruded in his life in a real way.
   In the Hannover crowd Pengo finally found a group whose anti-
authoritarian yet apolitical frame of reference came surprisingly close to
his own. It was a scene where drugs and computers were so interlaced
that Pengo began to understand Hagbard's brand of reality. In Hagbard
and his friends, Pengo had finally discovered a world that seemed to
intersect with his vision of how he wanted to live and think; it was a
world where the only reality was a reality of one's own choosing. In
Hannover, he had happened upon a group of people who appeared to
share in his obsession. Parties meant all-night sessions at someone's
apartment in front of his computer, while hash provided inspiration, and
cocaine and amphetamines chased away fatigue.
   It was during his first trip to Hannover that Pengo met Hagbard's
cronies, including Dirk-Otto Brzezinski, nicknamed Dob, and Peter
 Carl. They and others would gather, break into systems and take drugs
 together on and off for the next few years. They were brought together
 and held together by a common passion, though they wouldn't necessar-
 ily have sought each other's friendship under other circumstances.
, Dob was another example of the link between computers and drugs
 that was new to Pengo. The twenty-six-year-old goateed computer pro-
 grammer had a taste for good hashish and expensive meals. By appear-
 ances, Dob was remarkably ordinary; his casually corporate garb and
 wire-rimmed glasses hardly indicated anything rebellious, or even free-
 thinking, though he had spent his youth as an outcast among school-
 mates in Kenya, where his parents worked in international development.
    Through geographical maneuvering, Dob had cleverly avoided his
 compulsory stint in the West German army, not so much out of political
 conviction as out of a simple desire to skirt the service. So he was forced
 to live the life of an itinerant, commuting back and forth between
 Hannover and West Berlin and taking on contract work. He was highly
 intelligent, with an astounding facility for absorbing vast amounts of
 technical information. When asked a technical question, no matter how
 obscure, he invariably knew the answer. He was an expert at program-
 ming large Siemens computers, the computers of choice for most West
 German government agencies and large corporations. Like IBM's main-
 frames in the United States, however, Siemens computers were not well
 liked by programmers, who found the machines' operating system cum-
 bersome and old-fashioned. Someone with Dob's talent, therefore, was
 in great demand. In a good month, Dob could make $12,000.
    When he was in Berlin, Dob took a room for long stretches at the
 Hotel Schweizerhof, a luxury hotel in the center of the city. While
 there, Dob didn't venture far from a three-block area, which had every-
 thing he needed-restaurants serving heavy German and Czech fare,
 and the Belmont bar, a refuge where he could drink himself into a
 stupor, then stumble back to his room. Berlin offered a counterpoint to
 everything that Dob found bland about Hannover: its soulless provin-
 cialism, its reconstructed orderliness, its implicit curfews. Berlin coin-
 cided more closely with Dob's mental geography. He was known to go
 through periods of dark depression, when he would do nothing but sit in
 the Belmont, drink Glenlivet whiskey and play backgammon. Later, he
 owned a handgun, which he kept loaded with one bullet.
    Then there was Peter Carl, a fast-talking croupier at a Hannover
 casino. When Dob lost his driver's license after a drunk-driving incident
 in 1988, Carl became Dob's chauffeur. Slight and youthful for his thirty-
164   .&   CYEERPLINK
one years, Carl fascinated Pengo. He was part of a world that Pengo, for
all his escapades, had seldom encountered in Berlin. Carl tried to live
the life of an urban sophisticate, but he was essentially small time. His
job in the casino paid about $2,000 per month. To augment his income,
he occasionally transported cars to Spain, a country for which he had a
special affinity, or engaged in petty drug trafficking. In the summer of
1985, Carl was arrested for smuggling hashish from Amsterdam to West
Germany. For the offense, he got nine months of probation.
   Peter Carl grew up dirt poor. He had been raised in an orphanage
near Frankfurt. Carl managed to get some training as an electrical in-
staller, and later he made an unsuccessful attempt to get through a
Siemens Fachoberschule-a vocational school-but dropped out after a
year. From that point on, he would accord undue respect to technically
proficient people like Dob, people who had managed to accomplish what
he couldn't.
    Before Hagbard introduced Pengo to Peter Carl, he told him of Carl's
unpredictable behavior, that in a fit inside his Hannover flat Carl had
shot up the rooms, and that on one occasion Carl had flown into a rage,
ripped a kitchen cupboard from its mooring on the wall and thrown it
out the window. Pengo was prepared to meet a lunatic; instead, he found
Carl interesting and friendly, if slightly off-balance.
   On a later visit to Hannover, Pengo met the fourth and final member
of Hagbard's Hannover band, twenty-four-year-old Markus Hess, a beefy
Hannover University physics student. He seemed friendly enough, if
slightly aloof, and more of a solid citizen than the other three. He didn't
have a parent's death in his past, and he hadn't flouted his military
obligation. Markus Hess was a product of suburban life, middle-class, a
conventional achiever. While Markus was at university, a friend got him
a job programming part-time at Focus, a small software company in
Hannover specializing in UNIX-based software.
    Each of the five new acquaintances had come to computers via a
slightly different route. Peter Carl knew next to nothing about comput-
ers, but he liked to keep company with those who were technically
savvy. For Dob, more programmer than true hacker, the computer was a
refuge not unlike the Cafe Belmont. Hagbard's Illuminati-inspired para-
noid fixation both engendered and sustained his obsession with hacking.
Pengo's was an obsession with the idea of living in cyberspace and with
a new ambition to be recognized as the world's greatest hacker. Markus
Hess, whose burgerliche childhood provided few outlets, had an element
of rebel in him. When he discovered computers, he found a fully absorb-
ing diversion from the straight and narrow path blazed for him by his
parents. And he could enjoy the tension that existed between his proper
job as a programmer at Focus and the secret life he was living with his
    Focus was in some ways a displaced American-style Silicon Valley
start-up. Its six founders tried to run the company as loosely as they
could in a largely bureaucracy-bound country. Udo Flohr, a young lin-
guist and programmer, was Focus's president. The company started spe-
cializing in UNIX as early as 1984, when the AT&T operating system
was still relatively obscure in settings outside of universities and research
labs. On the strength of its expertise, the company quickly grew to
fifteen people, winning many of its customers in the U.S. One of Flohr's
most talented programmers brought Markus Hess into the company. The
eager young Hess learned all that he could about UNIX.
    The first time Flohr saw Markus Hess he was dressed in a tuxedo, on
his way to a formal event. Unlike Flohr's other technically minded peers,
Markus appeared to pay attention to certain of life's enjoyments not
ordinarily associated with someone who spends the better part of a day
in front of a computer. Markus had a busy social life, and such highly
developed political opinions that at age nineteen he went so far as to
join the youth organization of the Christian Democrats, West Germany's
conservative party, which had been in power under Helmut Kohl since
    Markus was a hard worker. He spent so much of his time at Focus, in
fact, that his studies suffered. Twice he took the exam that would have
enabled him to continue in physics and both times he failed. Flohr was
just two years older than Markus, and the two were poles apart in their
political views, yet the Focus president adopted an avuncular attitude
toward the young programmer. He counseled Markus not to leave the
university, but Markus dropped out anyway and enrolled at a correspon-
dence school, majoring in computer science.
    Markus came from a close family. The eldest of four children, he was
a model son. His father was a career bureaucrat at a car parts manufac-
turer in the state of Hessen, his mother a medical secretary. Markus had
always done well in school, finishing Gymnasium, the most advanced
level of secondary school in Germany, with excellent marks on his final
exams. Markus's parents had high hopes for their son but understood
little of his passion for computers. The interruption in Markus's studies
apparently unnerved his father, who paid a visit to Flohr to tell him to
hire his son full-time or let him go. Offended by the elder Hess's de-
166   ..   CYEERPlJNK
mands, Flohr refused to be forced into a commitment, asked him to
leave and told his young employee he would rather not see the elder
Hess around Focus again.
   Hess was a fine programmer, but he was also hardheaded and occa-
sionally intractable. If Flohr came to the software team with news that
he had sold a concept to a new customer, if not necessarily a finished
product, Markus would throw up his hands and charge Flohr with selling
air. Flohr, Hess would complain, was demanding the impossible. Inevi-
tably, Markus's more flexible and imaginative partner would intervene
and a compromise would be reached. Finally, despite such fundamental
disagreements, in 1987 Flohr did hire Hess full-time on the small re-
search and development team.
   Hess discovered the Hannover hackers' group and Hagbard in 1985
by way of a friend who told him about a hacker who could do amazing
things. This young man, Hess was told, had logins and passwords to
dozens of military computers in the United States. If Hess wanted an
introduction to system cracking, his friend told him, Hagbard was the
one to meet. Hess was strictly a programmer; he had heard about the
hacker clubs in Germany, but hadn't been particularly interested in
finding out more about them. But when he met Hagbard, that changed.
The exceedingly thin young man with dark blond hair and a soft voice
seemed to cast a spell over others. He spoke as if he and only he were
keyed in to life's deepest secrets. Ordinarily skeptical, Hess was taken
with Hagbard, and soon he too became convinced of a certain Durchblick
-unusual insight-emanating from the young man. Hagbard shared
with Hess much of what he knew, and for the first time Hess realized
that computers were not just isolated tools for programming-together
with a modem, they were a flexible set of lock picks.
   When Markus met Hagbard, it was as if a seeker had suddenly hap-
pened upon a seer. From Hagbard, Markus learned about stolen NUIs,
chatting on line and coaxing computers open. Most important, he
learned the value of patience and persistence. Hess saw that Hagbard
was not idly boasting when he spoke of getting into the Arpanet.
   Hagbard certainly cultivated some bizarre ideas. He believed that the
power of what could be done with computers was limitless. Hackers, he
proclaimed, embodied the future; their ability to release "soft bombs"
into the world invested them with unspeakable power. If he, Hagbard
Celine, were ever caught, he told Markus, he would become a "martyr
for peace, disarmament and information freedom." Like Hagbard's other
friends, Hess dismissed the Illuminatus theories as so much nonsense,
but like everyone ~lse he was drawn in by Hagbard's soft-spoken zeal.
    Unlike the other three hackers, Hess had no use for the drugs that he
saw at once sustaining and destroying Hagbard. Hash was the mainstay,
but LSD appeared to be the drug of choice for Hagbard and the major
inspiration for his paranoid fantasies. Hess's only real vice was his chain-
smoking, accompanied by the occasional beer. And whereas he main-
tained a semblance of order in his two-room Hannover apartment, Hag-
bard inhabited chaos. He lived in a surround of overflowing ashtrays,
heaps of dirty laundry and other signs of the necessities of daily life not
just neglected but transcended.
    As one of the few on the West German scene specializing in UNIX,
Hess was something of an oddball, and he wanted nothing to do with
organized hacking, so he had no use for the "VAXbusters" from Ham-
burg. Further, their area of expertise, VMS, was of no interest to him.
Besides, Hess had a real job at Focus to attend to; he didn't have Hag-
bard's freedom to devote every waking moment to hacking. Even if he
had, he wasn't sure he would be able to sit as Hagbard did, with a
meditator's stillness, oblivious to outside distractions for hours on end.
    However, it was such an infectious passion that before long Hess was
using any spare moment he had exploring networks. In a way, it was
Hess's own break from the predictable way of life he had always known,
with its built-in constraints and expectations. Hess may not have been
interested in group hacking or VMS, but he was interested in hacking
into UNIX systems. Most of the time, he stayed late at Focus and worked
from there. Sometimes he used Focus's NUl, sometimes he used his own,
 legitimately acquired NUL Only occasionally did he resort to the stan-
dard practice of using stolen NUIs. That, as far as Hess was concerned,
was straying too far toward questionable ethics.
    One of the most useful lessons Hess learned from Hagbard was that
this pursuit required endless patience. It meant sitting for hours on end,
dialing and redialing. Once you had reached a system, and the friendly
 log-in prompt at computer's portal asked for an account and password,
 it meant more time expended in getting in. Once you were on the
 machine, exploring its crevices and testing the limits of your powers took
still more perseverance. Although Hagbard was no programmer, and was
fully dependent on others to write programs for him, his infinite patience
 and utter single-rnindedness made him a more effective cracker than
 many of the others.
168   .t.   CYF:ERPUNK
    Part of the tremendous attraction of computer networks for Hess,
Hagbard and Pengo came from knowing that beyond their underpowered
little personal computers, or even beyond Focus's comparatively more
powerful machines, there was always a larger, faster machine somewhere
to be explored. If a log-in prompt appeared, and a password worked, the
next task would be to try to guess the type of the machine and how
powerful it was. Privileges, sometimes referred to as superuser status,
were frequently a goal. Unlike Hagbard, Hess and Pengo could try some
creative programming once they were inside a computer. They could use
operating system commands and write programs to find and exploit weak-
nesses in operating system security.
    Once an intruder obtained superuser status, it meant that the machine
was entirely under his control. It was possible to read and change peo-
ple's files, peruse their electronic messages, destroy their work or even
do their work for them. The ultimate benign hack would be to send
someone a note announcing that the intruder had just visited and had
found-and fixed-a bug in his program. Nothing could be protected
from a superuser. And hacking was far more efficient than rifling through
a roomful of filing cabinets, scrutinizing every scrap of paper for some-
thing interesting. The power of the computer could easily be turned back
on itself. The machine could be instructed to show a list of all its
documents that mentioned a key word or phrase, such as classified or
nuclear, or whatever an intruder was seeking.
    Like the others, Hess was transfixed by the sheer number of possibili-
ties. The Arpanet alone linked thousands of computers. It in turn was
part of the Internet, a network that linked so many computers that no
one was really sure how many there were.
    Pengo's first visit to Hannover ended when, at a party at Hagbard's,
Dob said he was on his way back to Berlin and asked Pengo if he wanted
a lift. It was five in the morning. Pengo had been away from home for
two weeks. He had run out of money and clean clothes and he welcomed
the ride. When they got into Dob's sports car, Dob produced a nugget of
hash the size of a walnut. The two smoked it as they sped through East
Germany at over one hundred miles an hour. Dob told Pengo he made
the trip all the time, always at high speed, and always in an altered state
of mind. Dob liked to carry out logistical chores such as a trip from
Hannover to Berlin in the most efficient way possible, even if it meant
placing his life in danger.
    Pengo enrolled at the technical university in Berlin to study computer
science, but he couldn't stay away from Hannover too long. He started
to travel there frequently to spend a few days at a time with Hagbard. A
visit usually started off with a round or two of hash and progressed to all-
night hacking sessions. Once or twice, they went to Focus to visit Hess
and watch him hack from there.
                                  ... T ...

It would have been natural to select Cliff Stoll out of a lineup of sus-
pected computer fanatics. Skinny Cliff had a maniacal edge to his per-
sonality, cultivating what might even be a parody of the eccentric
scientist. Wild corkscrews of brown hair projected six inches from his
head; he had the gait of a pogo stick, a yo-yo always in hand and in
perpetual motion. Stoll's speech was punctuated by a parade of excla-
mations-"Hot ziggity!" and "Holy smokes!," "Jeez" and "Really, really
neat!" If his computer kept him waiting for more than half a second
before executing a command, he would yell "Commie!" at it. In fact,
"Commie!" was the mocking charge leveled against any recalcitrant
inanimate object.
    Stoll's political views leaned appropriately leftward, in part because it
never occurred to him to think otherwise. He went to college during the
1960s, and he opposed the Vietnam War, but he wasn't much of an
active crusader for the Left. In fact, he considered himself a nonideo-
logue who resisted left-wing dogma. He operated more on his principles.
It was on principle, for instance, that he made certain that all of his
employers were involved in nothing but the purest of research. He
couldn't bring himself to work for the Lawrence Livermore Laboratory,
a government research center whose bread-and-butter projects include
designing nuclear warheads and Star Wars weapons for the U.S. mili-
tary, to say nothing of the National Security Agency, whose computer
scientists were, by definition, spies.
    So in 1986, Stoll took a postdoctoral job at Lawrence Berkeley Labo-
ratory designing mirrors for the Keck Observatory in Hawaii. High atop
the Berkeley campus, LBL is Livermore's sister laboratory thirty miles to
the west. LBL is one of the United States' national research labs and the
site of broad-based unclassified scientific research. When Stoll's grant
money ran dry in late 1986, he was forced to look for something else to
    For an astronomer in the midst of the Reagan era it wasn't an unusual
fate. Federal money to support the basic sciences was shrinking. The
money was over the hill, literally, at the weapons laboratories. As it
happened, the computer knowledge Stoll had picked up in high school
170   •    CYE6RPUNK
and college gave him a certain advantage over other astronomers who
lost their funding. In August 1986 he became a system administrator, in
charge of maintaining the laboratory's dozen mainframe computers. This
meant being responsible for everything from backing up important data
and taking care of computer security to helping the scientists who were
his customers use the powerful machines more efficiently. Although it
wasn't exactly the work he was looking for, it allowed him to stay in
     One of his first assignments seemed simple enough: to reconcile a
small accounting error that had shown up. LBL used some home-brewed
accounting software, and the patchwork of programs, written by summer
students over the years, had come up with a seventy-five-cent discrep-
ancy between the normal system accounting and the lab's own charging
scheme. Cliff stayed at work until midnight puzzling over the mysterious
seventy-five-cent error, which he suspected might be a computational
rounding error.
    After careful examination, he discovered it wasn't a rounding error,
but the work of an unauthorized person from outside the lab using the
account of an LBL researcher who had left several months earlier. With
characteristic gusto, Cliff became a self-appointed one-man SWAT
team. He set up traps that captured the hacker's every keystroke on a
printer and alerted him every time the intruder was in the computer. He
kept a detailed logbook, and he wrote a software program that tripped
his pocket pager whenever the trespasser logged on. Before long, he was
doing little else but tracking the uninvited guest. Occasionally he even
slept in his sleeping bag on his office floor to keep a constant vigil over
the hacker.
     Finding an intruder on a computer system is often as serendipitous as
guessing the correct password to break into a system. Detecting a break-
 in can be a matter of timing, perseverance and ultimately luck, especially
 if the trespasser takes steps to cover his tracks. It requires the same skills
that the hacker himself employs, and it can often mean getting inside
 the hacker's head to anticipate his next move. A system manager like
Cliff Stoll hunts down an intruder not only because computer security
has been broken. It's also a matter of wounded pride, and a threat to the
 institution's ability to keep its doors to the outside world open and keep
hackers out at the same time.
     Stoll gradually began to understand the hacker's strategy. By exploit-
 ing a mistake in the way LBL's computer managers had installed certain
software, the intruder had managed to give himself privileges on the
system that are usually reserved for system managers. As a result, the
intruder was able to create accounts with names such as Hunter and
Jaeger, assigning them such passwords as Benson and Hedges. And the
hacker was careful: every few minutes, he typed the command "who,"
which listed everyone using the computer. Evidently, he knew how to
log out in a hurry. If the hacker thought he detected a legitimate system
operator on the computer, or someone else with full privileges, in a
single keystroke he could instantly disappear back into the electronic
   Stoll didn't know, of course, whether it was indeed just one hacker
plaguing his computers or a gang. Some empirical evidence seemed to
support the case for just one hacker, but he wanted to be certain. So he
set up a scientific probe, to see if he could discern typing patterns from
among thirty colleagues at the lab. Once he decided he could, he applied
the same test to the hacker's typing rhythms and discovered that most of
the time the typing came across the telephone lines in methodical,
evenly spaced strokes. Only occasionally was the typing more random,
as if someone might be hunting for the keys. He didn't pause to consider
that by the time the keystrokes had been transmitted through interme-
diate computers and data networks, all of the information that could
identify the typist had long since disappeared.
   The most obvious solution to the problem would have been to lock
the hacker out entirely. And that would have been simple. Stoll needed
only to change all the lab's passwords and tweak a piece of software
called GNU Emacs that ran on LBL computers. A powerful program-
mer's text-editing program, GNU Emacs was used by nearly everyone at
   Because LBL programmers had installed GNU Emacs on the lab's
computers in a way that gave a user special privileges, any user could
access any file on the system using a command called "movernail." In
effect, LBL created a hole in computer security-an effective window
into normally inaccessible areas within the computer, and the hacker
had discovered this. Stoll was beside himself. He decided that instead of
slamming the door and keeping the hacker out, he would let him roam
through the system with relative freedom and catalogue his every move
-and then trap him. He reasoned that by keeping the system open he
could get the hacker to stay on the line long enough to let the phone
company trace him back to his lair.
   What really was the danger of an interloper browsing around anyway?
The information in the system was often personal but, as far as the LBL
172   A   CYEERPUNK
system was concerned, not vital for national security. The hacker could
peruse grant proposals and information about the computer system, as
well as electronic mail carrying gossip, news and love letters. Stoll's
salary and resume were open for browsing, too.
   To Stoll the problem was not so much the hacker's rummaging around
the LBL computer as his use of that computer to ricochet his way to
other computers over the Arpanet-to virtually anywhere in the Inter-
net. With the simple command "telnet," the hacker could instruct the
LBL computer to connect to Internet computers anywhere in the world,
at military bases and Pentagon contractors and research laboratories. He
had only to come up with an account and password for those computers.
In fact, the hacker was losing interest in the Berkeley machines, and
was just using them as a launchpad to others, especially those on Milner,
the unclassified military network connecting Defense Department instal-
   Given Stoll's studious avoidance of military matters, he had no real
idea what went on in those computers, but their locations alone sounded
like serious business: Redstone Missile Command in Alabama, the Jet
Propulsion Laboratory in Pasadena, Anniston Army Depot, Navy instal-
lations in Virginia and Florida and the Air Force Systems Command
Space Division in EI Segundo, California. Not only was the hacker
logging on, but he was directing his searches in an ominous way. He
told the computer to seek out files containing such words as stealth,
nuclear and NORAD. He latched on to files about the space shuttle's
secret missions.
   Stoll had heard a lot of stories about hackers from colleagues at Stan-
ford, fifty miles south of the San Francisco peninsula, and Fermilab, near
Chicago, but they had struck him as harmless pranksters. This interloper
seemed different. Stoll was stunned by the hacker's ability to traverse
the nation, nudging computers open just by guessing obvious passwords.
The nation's computers appeared to be surprisingly unprotected. A guest
account, for instance, might have guest as its login and guest as its
password. Or the login might be a user's last name, the password his
first. So the hacker could ask the computer to show him a list of users
on the system, then try breaking in simply by typing in people's names.
On the handful of occasions that the hacker managed to give himself
superuser status, he would create his own accounts or change passwords
of existing accounts for his future use.
    When the hacker hopped over the network to a computer at Lawrence
Livermore, Stoll panicked. He called Lawrence Livermore and told the
system manager to shut down that machine.
                                  ...   ,. ...
In 1986, thoughts of breaking into computers for money were beginning
to waft through the air. One Hamburg district attorney had developed a
theory of how young hackers might unwittingly allow their computer
expertise to be exploited for purposes of industrial espionage, or even by
the East Bloc: Communist agents, he reasoned, already expert at manip-
ulation, would have no trouble engaging the minds of wayward teenag-
ers. Not only did nervous West German authorities and imaginative
computer crime experts have their own finely crafted theories about the
potential for exploiting impressionable young hackers, but the hackers
themselves were beginning to see what was possible.
   Sometime in early 1986 the West German group decided to try in
earnest to market its talents. Back in Dob's Hannover apartment, under
the influence of many pipes of hash passed back and forth, Carl, Dob
and Hagbard had a night of intense discussion. The first question, of
course, was how to execute the initial contact. One idea was to go up to
the door of the Soviet embassy in Bonn. Another idea, lifted perhaps
from the pages of an espionage potboiler, was for one of them to slip a
note written in code into his passport, a missive that would alert a border
guard to his mission. It was a fine idea, all agreed, but no one was quite
sure what the note should say. Peter Carl didn't have the technical talent
of his colleagues but he had made up for that with sheer bravado and a
sense of the possible, so the intrepid Carl readily agreed to travel to East
Berlin and do a little cold calling. The idea was simple enough: they
were hackers who could get into some of the world's most sensitive
computers. From those computers they could extract sensitive informa-
tion, information they knew would interest the Soviets. What was more,
they could provide the Soviets with some of the software they needed to
catch up with the technologically more advanced West. Why shouldn't
the Soviets want to do business with them? Of course it was illegal. They
all knew that. But in selling the Russians military and scientific: infor-
mation, they argued, they would be doing their part for world peace. A
name for the project? Equalizer.
   The idea was not to teach the Soviets how to hack themselves but to
keep them dependent on the group somehow, to tell them just as much
as they needed to know to stay interested, but not so much that they
174   •   CygERPUNK
would be able to hack on their own. If the hackers were going to sell the
access codes and hacking knowledge itself, then it would have to be a
onetime deal and their price would have to be steep. A million marks,
they decided. For that they would tell the Soviets about networks, and
throw in the lists of logins and passwords for computer systems world-
wide. The night wore on, and so did the hash. More hacking business
ventures came to mind. Why restrict themselves to doing business with
the Soviets? Why not include the Chinese? That idea was discarded
when the group agreed that China wasn't enough of a player in the
struggle between world powers. Project Equalizer had to stay focused.
   Hagbard and Dob took on the job of putting together a "demonstra-
tion package" to show to their future business partners. Hagbard assem-
bled a list of logins to computers, including those at SLAC in California,
a computer at the Department of Energy and the U. S. Defense Depart-
ment's Optimis computer. Once Hagbard had opened a computer, he
left it to Dob to do research inside the computer, looking for material
that might interest the Soviets. One document with the title "Radioac-
tive Fallouts in Areas 9a and 9b" sounded good. So did one called
"Propellants of ICBMs." Once he had downloaded the data, Dob trans-
ferred it onto diskettes and made printouts of each file, being careful to
delete any clues to passwords and break-in methods. All told, they gath-
ered material from thirty different computers.
   In early September 1986, Peter Carl drove across East Germany from
Hannover, left his car in West Berlin and took the subway to the Soviet
trade mission in East Berlin at Unter den Linden. He approached a guard
who was seated behind a glass partition, introduced himself as Peter Carl
from Hannover and asked to speak to someone from the mission, as he
had a business proposition to discuss. He assumed it was obvious that he
was there to speak to someone from the KGB. The guard told him to
take a seat. After a thirty-minute wait, a man appeared in the waiting
area and asked Carl what he wanted to propose. Carl explained that he
was a hacker from the West with access to interesting information, and
he wanted to propose a business deal. The man nodded and disappeared.
Ten minutes later, a tall, dark-haired man emerged from the building's
recesses. He introduced himself to Carl as Serge, the French pronuncia-
tion of the Russian name Sergei, and showed the visitor into a sparsely
furnished conference room.
   Sergei asked Carl what he had in mind. Carl explained again that he
was a member of a group that could get its hands on interesting infor-
mation. Sergei was only vaguely familiar with the term hacker, and asked
Carl to explain. Carl described most of what he knew. He said hackers
were people who could break into computers and retrieve information
and programs quickly and surreptitiously. Sergei still appeared to be
confused, but his interest was piqued and he asked for more specifics.
This particular band of hackers, Carl said, had the means to break into
dozens of Western computers and get everything from information on
high-energy-physics research to proprietary banking data. Carl said he
would like to offer the Soviets a package of West German hacking know-
how, including logins and passwords to dozens of military computers in
the United States. His price: one million marks-more than half a
million dollars. Sergei raised his eyebrows but remained silent.
   Carl kept talking. He wanted very badly to appear to be someone
worth one million marks. He had a salesman's delivery and a salesman's
confidence even if the terms he tossed around were not quite accurate.
He said he didn't have the demonstration package with him, that it was
in West Berlin, ready for him to pick up and deliver to Sergei. He hadn't
wanted to bring it with him on this first trip, as there was no telling who
might search his bags.
   Sergei must have been amused and curious. If nothing else, the ap-
pearance of a computer hacker at the trade mission in East Berlin had
no precedent. As a rule, when it came to gathering technology the
Soviets had a long tradition of doing their own fieldwork. Since most of
what they were interested in, especially technology for advanced com-
puting, was on a list of highly restricted technologies maintained by a
consortium of Western nations known as COCOM, the Soviets had
long since resorted to extralegal means of procuring hardware and soft-
ware. The FBI liked to maintain that Northern California's Silicon Val-
ley, where much of American computer innovation resided, was
crawling with KGB agents. The FBI claimed that one of the primary
missions of the Soviet consulate in San Francisco was to funnel U.S.
technology into the Soviet Union. The consulate building was suspected
of having a hidden forest of antennae and other surveillance equipment
on its roof-all targeted at capturing sensitive telephone calls in Silicon
   Through the years, a smattering of Soviet espionage cases had become
public, but only because the spies were caught. The Soviets acquired
advanced computers by hiring agents to set up dummy companies, order
whatever was needed and then fold quietly. For the most part, the
Soviets had a long-standing practice of reverse-engineering technology
based on what they were able to gather. Of course, there was no way the
176   ~    CYEERPUNK
Soviet Union was going to build a technological infrastructure by that
method. Nevertheless, it was a system the Soviets continued to use
partly because of trade restrictions, and partly from a simple predilection
for doing things that way. All told, their software was a motley collection
of retooled operating systems and compilers, roughly equivalent to the
American originals but transposed into Cyrillic. Their hardware was
based principally on VAX and older IBM 360 designs and they were
always on the lookout for good VAX software, especially VMS source
    Given the ease with which American computer networks can be
entered from a safe redoubt outside of the United States, it isn't hard to
consider the possibility of a sophisticated Soviet intelligence-gathering
operation targeting the vulnerable computers of commercial American
high-technology ventures and nonclassified U.S. military systems. In the
early 1980s, officials in the Reagan administration expressed alarm at the
existence of a circuitous computer link that would have allowed Soviets
in Moscow to log directly in to American computers: an out-of-way
international research center located outside of Vienna and known as
the International Institute for Advanced Statistical Analysis was con-
nected by a commercial computer network to the United States and had
a direct computer tie to a research center in Moscow. The institute lost
its U.S. funding as a result of the computer link.
    Some American officials argued that even though nothing secret was
accessible from the center, it was conceivable that the Soviets could use
the power inherent in computers to scan quickly through vast amounts
of information and piece together a clearer picture of classified data. But
there was no evidence to support this scenario, and several years later
U. S. funding for the Vienna institute was quietly restored.
    Whether the Soviets truly had designs on young computer outlaws is
unclear. That the Soviets would have called upon a select group of young
experts to rummage around inside U.S. computers for them, or would
have dispatched someone like Sergei on a recruiting spree to the Chaos
meetings in Hamburg, was probably a notion that existed only in the
minds of nervous Western officials. But with this self-described hacker
on his doorstep promising tapes filled with digital delicacies from the
West, it wasn't surprising that Sergei considered the idea worth pursu-
    The Soviet official made it clear to his visitor that while he was by no
means uninterested in what Carl was suggesting, he could hardly agree
to hand over a million West German marks for something he not only
hadn't seen but didn't quite understand. And Carl himself, still strug-
gling with some of the technical concepts with which his cohorts seemed
so comfortable, was in no position to deliver an extemporaneous lecture
on the navigation of data communications networks, the computers that
resided on those networks or the specific information they contained.
Like that of any marketing man, Carl's job wasn't so much to understand
what he was trying to sell but, by virtue of his unerring enthusiasm, to
convey the value of his product. But even Sergei's most basic questions
were too tough for him. Sergei asked Carl to return in the next few days
with his demonstration package. It would be sent to Moscow for a thor-
ough analysis and, if it were deemed to be worth a million marks, then
a million marks would be forthcoming. Sergei then asked to see Carl's
passport. He took some notes and left the room briefly. When he re-
turned, he told Carl that the next time he came, as long as he used the
border crossings at Friedrichstrasse and Bornholmer Strasse, the guards
would let him pass freely.
   Two days later, Carl took Dob's car and drove to the border at Born-
holmer Strasse. After a brief glance at his passport, the guard waved him
through. At the trade mission, he asked for Sergei. This time, Carl had
with him the demonstration package: an index to computers around the
U.S., with Pentagon computers at the top of the list. Under each head-
ing was an index of what was contained in the individual computers.
Account names and passwords had been carefully deleted. Sergei re-
mained polite but skeptical. This time, Sergei gave Carl 300 marks for
his expenses and made out a receipt. He also gave Carl a telephone
number in East Berlin where Carl could reach him. He told Carl to learn
the number by heart and to call only in an emergency. Carl used part of
the money to fly back to Hannover. With the rest, he went straight out
and bought a small Casio electronic notebook, where he entered Sergei's
telephone number. When he spoke on the telephone to Hagbard or Dob
about his trip, he imparted his information in an appropriately cryptic
manner. Paris meant East Berlin. Teddy Bear stood for Sergei, Russia and
the East Bloc. Equalizer, of course, was understood by all as the code
name of the operation.
    A week later, on instructions from Sergei, Carl appeared at a building
on Leipzigerstrasse, a main thoroughfare for traffic in and out of East
Berlin. He took a rickety elevator to the fifth floor. From what Carl
could observe, the office served as a business dealing in heavy machinery
and railroads. Sergei greeted him, and this time they had a general
conversation. Sergei wanted to know more about Carl's background.
178   A   eygERPUNK
Carl could only assume that Sergei hadn't yet received word back from
Moscow on the demonstration material. From then on, Sergei said, their
meetings would take place at the Leipzigerstrasse location.
   At the next meeting the following week, Sergei told Carl that he had
gotten a response from Moscow. While the package contained some
interesting information and bore out Carl's claim that the group could
get into certain interesting computers, it wasn't exactly what they were
looking for. More to the point, a million marks for the demonstration
package was out of the question. However, Sergei was interested in
certain things that could bring the hackers some money. While he
wasn't interested in general hacker know-how per se, he wanted to know
if Carl's group could produce information about radar techniques, nu-
clear weapons and the Strategic Defense Initiative. Moreover, he said,
the source code for VMS and UNIX, compiler programs and programs
for computer-aided design and computer-aided manufacturing could
bring the West Germans a tidy sum. Sergei said his customers back in
Moscow also wanted software from the American firms Ashton-Tate and
Borland, two highly successful personal computer software companies.
   Such an exchange wasn't precisely what Carl had in mind. He had
imagined that he would present Sergei with a menu of sorts, courtesy of
Hagbard and Dob, and that Sergei would tick off the various computers
he was interested in. Then Carl would dispatch Hagbard to root out
whatever he could. But Sergei appeared to have a different notion of
what a hacker could provide. Not only was Carl not altogether certain
he could get what Sergei needed, but he wasn't altogether certain what
Sergei was talking about. Source code and compilers? These were things
to consult with Dob about. It was beginning to look as if the things Carl
had to offer Sergei didn'twant, and the things he wanted Carl couldn't
offer. Nonetheless, Sergei remained sanguine about the prospects of
what Carl and his friends might be ableto supply. This time, he gave
Carl 600 marks, or $300, for his expenses and took him out to eat. Over
lunch, they made small talk. Carl learned that Sergei was married with
children, and that he liked to go fishing. But when Carl asked him to
explain precisely what his job was, Sergei declined to answer.
   Sergei gave Carl photographs of a young woman and a small child,
along with the woman's name, address and telephone number. In the
event that his frequent trips to East Berlin should prompt Western au-
thorities to question him, Sergei said, Carl should tell them that he went
to visit his girlfriend, with whom he had a daughter.
   Sergei seemed to keep at least two lists containing names of data bases
and software. It appeared to be a priority system of sorts, as the lists had
a numbering scheme and items that were crossed out. By the fifth meet-
ing or so, Sergei asked Carl to tell him something about the others in
the group. Carl told him about Hagbard and Dob while Sergei made
notes in a black binder.
   And so the meetings went, once every week or so, through late 1986,
always starting at noon, always consisting of a meal in a restaurant at
which Sergei smoked Marlboros, drank glass after glass of orange juice
and at the end handed Carl 600 marks for his expenses. Even when Carl
had nothing to deliver, he went anyway, just for the 600 marks. Occa-
sionally Sergei gave Carl small presents: a nice cigarette lighter, a bottle
of spirits, some Russian caviar.
   In spite of Sergei's measures for protecting Carl and his generally
solicitous manner, the Soviet didn't appear to be particularly satisfied
with what he was getting. Most of the material, he told Carl, consisted
of indexes to information rather than the information itself. And that
was often available only on microfiche. So Sergei brought his own com-
puter expert to one of the meetings, but the expert could speak only
English and Russian. He and Carl could barely communicate. Carl was
frustrated. He didn't want the entire business going sour just because of
a communications problem. So he asked Dob to accompany him on one
of the trips to East Berlin. Carl told Dob that he would be able to clear
things up because he would be able to understand exactly what Sergei
was looking for. It took some persuading, but Dob finally agreed to go.
During the meeting Sergei explained once again that he was not going
to buy the hacker know-how for a million marks, and that he wasn't
satisfied with the material that had been delivered so far. His interest,
he emphasized, was in information from U.S. military computers, source
code and compilers. Dob knew exactly what Sergei was saying, but the
meeting was a disappointment. He didn't see that he would be able to
get source code, nor could he get very excited about the idea of making
a lot of small, low-paying deliveries over an extended period.
   Then Carl told Sergei about Pengo. He told him that Pengo was
a particularly capable hacker specializing in VAX computers, and that
he could get good material. The Soviet expressed a great deal of inter-
est in what Pengo might have to offer. He said he wanted to meet
him in person and make his own assessment. Carl said he would bring
him by.
 180   A   CYE£RPUNK
Meanwhile, Cliff Stoll's vigil wore on. He was obsessed with his hacker;
all but the most basic housekeeping for the lab's computers was shunted
aside. And the hacker seemed to get even more single-minded. It be-
came clear to Stoll that this was no mere computer science student
romping in an electronic playground-he had developed a keen interest
in things military. Now he seemed to want to see files pertaining to
intercontinental ballistic missiles and the Strategic Defense Initiative.
Stoll watched him try with a persistent series of educated guesses to get
into a computer at the White Sands Missile Range:

login: guest
Password: guest
Invalid password,              try again
login: visitor
Password: visitor
Invalid password,              try again
login: root
Password: root
Invalid password,              try again
login: system
Password: manager
Invalid password,             disconnecting after 4

   If it seemed that the hacker could cause some harm in the system he
was poking around in, or if Stoll thought that the people in charge of
the computer should know there was a hacker poking around in their
data, he would call them. Perturbed and incredulous at first, they would
close the hole that the hacker had used to climb in. So far the hacker
hadn't found any sensitive national security data, or at least Stoll didn't
think he had. But it was almost certainly on his agenda.
   The hacker plaguing Cliff Stoll didn't seem to be much in the way of
a computer genius. He was seldom inventive. In fact, his most remark-
able trait was his plodding persistence: he created connection after con-
nection, then, like a dog trained to sniff out drugs, systematically
searched each system for military information. Long after Stoll had be-
come overcome with fatigue, the hacker would continue twisting door-
knobs. Stoll began to think the intruder might not be human at all.
Could it be a robot, a computer programmed to look for military infor-
mation? Stoll decided that it wasn't, simply because whatever it was
made spelling errors.
   At first, Stoll believed the intruder to be somewhere on the Berkeley
campus. But there was evidence against that theory. The hacker was
very familiar with UNIX, but his behavior showed that he didn't know
anything about Berkeley UNIX, a variation on UNIX that was de rigueur
in Berkeley. Instead, he was using the traditional UNIX commands first
developed at AT&T's Bell Laboratories. He was speaking UNIX with a
strong AT&T accent.
   From the high keep of various universities, Stoll had never had much
cause to interact with the outside world. His view of things left little
room for shades of gray. As Stoll saw it, scientists who engaged in pure
research and divorced themselves from the military were on the right
side. The CIA, NSA, FBI and military establishment were, by contrast,
sinister and shifty, not to be trusted. And so was this hacker. Not only
was he inside computer systems that he had no right to be using, but he
was robbing Stoll of his time, time to spend on the work he had been
hired to do: to help the lab's astronomers use computers to design their
telescope. If his more forgiving colleagues were entertained rather than
upset by teenagers who broke into computers, Stoll saw nothing to
excuse. In something of a contradiction to his self-consciously liberal
way, Cliff Stoll was, in the end, a bit of a crank. He saw the hacker as a
dark foe and he wanted to see him behind bars.
   If he was going to catch the hacker in the act, Stoll knew he would
have to trace the telephone calls. And in order to trace them, he would
need a search warrant. So Stoll did something that was at odds with his
political sensibilities; he called the local FBI office. He explained that
there was a hacker in his computer who seemed to have a taste for
military information. He was surprised and upset by the FBI's response:
the agency had far bigger things to worry about than a loss of seventy-
five cents.
   The second call across the great academic divide to the nasty world of
bureaucratic authority proved more successful. The Oakland district at-
torney's office was immediately interested. Stoll explained that the
hacker was coming in through one of LBL's Tymnet links. Tymnet's
network ran all over the United States; the hacker could be calling LBL
from almost anywhere. In order to trace the call farther than the Oak-
land Tymnet connection, Stoll needed help from the phone company,
and the phone company required a search warrant. The Oakland DA
took care of that.
   Pacific Bell, the local telephone company, traced the call from the
T ymnet node to McLean, Virginia, and from there to Mitre Corpora-
182   A   CYEERPUNK
tion. An MIT spin-off, Mitre is a Pentagon-funded research center.
When Stoll confronted computer security managers at Mitre with the
news that a mysterious hacker was using it on his way to supposedly
secure military and university computers all over the United States, the
Mitre officials swore it couldn't be true; their computers were absolutely
impenetrable, secured from the outside world, they claimed. But it
turned out there was an enormous hole. The hacker was using a local
area network (these networks tie computers within a building, permit-
ting them to communicate locally at high speed) at Mitre to slip around
the side of the company's computer security protections. He would dial
into a pool of modems at Mitre that were shared by the local area
network and then use the same modems to dial back out again.
   The hacker used the loophole as his personal bridge' to reach other
computers. It was costing Mitre thousands of dollars in long-distance
telephone calls because when the hacker called from Mitre he stopped
network hopping and started dialing directly into other computers. He
had also planted a Trojan horse program in the Mitre system that cap-
tured passwords as others typed them in and copied them to a hidden file
he could retrieve later. Mitre's computer people were shocked by this
discovery. They pleaded with Stoll to keep the whole matter secret. For
the public to discover that a defense contractor engaged not just in
classified work but with contracts to build secure computer systems had
fallen prey to a hacker would have been a disastrous turn. In exchange
for that promise, Stoll coaxed Mitre out of its past months' telephone
bills. From scrutinizing the telephone bills, he discovered that the hacker
had been active for several months before Stoll had even detected him
-for far longer than Stoll had thought. Stoll had already counted close
to thirty computer systems the hacker had broken into. The number of
attempts was at least ten times that.
    Soon after the phone company had completed the Mitre trace, Stoll
watched the hacker jump from LBL into the Milnet Network Informa-
tion Center, a computer network directory information service, where
he discovered four network addresses and telephone numbers for CIA
personnel. The hacker wasn't actually inside a CIA computer. That
would have been far more difficult, because those machines are not
directly connected to public computer networks. But he seemed to be
getting closer. Stoll wrestled for a moment with his conscience. Why
consort with the Establishment and alert the CIA to the electronic spy
in its midst? Why not just drop the matter right there and let the hacker
romp unchecked? But before he could do much more ruminating, Stoll
was reaching for the telephone and calling the phone numbers the
hacker had discovered. In contrast to the tepid response Stoll had gotten
from the FBI, the CIA immediately dispatched four people to Berkeley
to discuss the matter.

It hadn't taken much to convince Pengo to join Project Equalizer. At
first, Dob flattered him by letting him know how much they needed him
and by telling him that Sergei had expressed a particular interest in
meeting him. Hagbard had too many limitations. Given enough phone
numbers and passwords, Hagbard could browse for hours inside VMS
systems. He knew how to log in and find a directory. He knew how to
search for keywords and files. He knew what a disk was and where
electronic mail was stored. But beyond that, he was lost. Ask him, for
instance, to determine which machines were connected to the system
he was on and he was utterly confused. Often, he was too stoned to work
very efficiently. He needed too much time to think of the next step. A
gifted hacker worked as much on spontaneous leaps of intuition as on
anything else. And for all his grandiose theories, Hagbard couldn't pro-
gram. The group needed a VMS expert who could program. Pengo was
the natural choice. Carl made it clear that it could be a lucrative ven-
    Pengo had theories of his own on how to get money for hacking. He
developed a three-point strategy. One idea was to sell a certain amount
of know-how and charge a more reasonable sum, perhaps 150,000 marks,
or $75,000. Another suggestion was that he organize some seminars for
the East Bloc, to teach the Soviets about technology and hacking. The
third idea, which Pengo seemed to latch onto with the most enthusiasm,
was to convince the Soviets to set him up to do "safe" network prowling
from East Berlin-that is, to supply him with a top-of-the-line VAX
computer with plenty of storage capacity, a high-speed modem for fast
data transfer and secure telephone lines that couldn't be traced. Of the
three ideas, the last one seemed to have the most potential: it would
give the Soviets the software they needed while providing Pengo with
some easy money. That, Pengo decided, was the idea he would empha-
size with Sergei.
    But he could hardly meet Sergei empty-handed. At Carl's urging,
Pengo looked for something to take over as a first enticement. Any
software he could get his hands on, as long as it seemed impressive,
would probably do. First he managed to procure a magnetic tape contain-
184   l.   CYEERPl1NK
ing some software from previous forays-a chip-design program called a
PAL assembler from a hack into Thomson-Brandt, the state-owned
French electronics manufacturer, and smaller programs for the VAX.
But to get something worthwhile from the networks, he would have to
stay logged on to a computer system for a long time. That meant choos-
ing a computer on which he knew the security was lax. He had been
inside Digital Equipment Corporation's Singapore computer center be-
fore and he knew it fairly well. As far as he could tell, there was little
security at the Singapore facility. The system manager seemed to be
asleep at the switch, seldom checking to see who was logged on to the
computer. It was easy to get on the Singapore VAX with full privileges,
and Pengo knew it would be easy to stay on for a long time. He logged
on late one night and found exactly what he needed. It was a security
program for VMS called Securepack, developed at Digital in 1983 for
internal use on Digital computers. The program, which allowed system
managers to alter levels of privilege on a computer, would make an
interesting nugget for the Soviets. Pengo downloaded the program and
put it onto several diskettes. He also made about thirty pages of print-
outs, taking care to delete any information as to how the computers were
actually penetrated.
   On the night before the trip to East Berlin, Pengo, Dob and Peter
Carl spent long hours hacking from Dob's room at the Hotel Schweizer-
hof. Hash was burning in abundance, and although it helped to keep
spirits high, it also produced one unfortunate side effect: when he was
high, Pengo had the habit of dwelling for hours at a time on one prob-
lem, which undermined his overall productivity considerably and aug-
mented the phone bill. Every few hours, Peter Carl would poke his head
into the room to check on Pengo's progress, and each time he threw up
his hands in disgust. Pengo was costing them hundreds of marks in
connect time, and from what Carl could tell, he wasn't producing any-
   Pengo decided to put a digital record of the session from the Schweiz-
erhof onto several diskettes and take those over too. He and Carl
emerged bleary-eyed from the hotel the following morning and started
out for Sergei's office while Dob slept off the excesses of the night. Their
loot stored in Carl's briefcase, they boarded the subway at Wittenberg-
platz, a short walk from the hotel. They changed to a different line and
boarded the train that traversed the eastern portion of the city, speeding
past abandoned stations. It was a familiar trip to Pengo. He had been to
East Berlin many times when he was younger. On Christmas and Easter,
when his parents were still together, the Hubners from West Berlin went
to visit the Hubners in East Berlin. And even if Pengo hadn't stopped
to consider the quirk of fate that brought him into the world in the
West, it's likely that his indifference to borders was a sign that he
understood, if only subconsciously, Berlin's provisional nature, a place
where, in the end, allegiance was arbitrary.
   Just fifteen minutes from the time they set out, they had reached the
Friedrichstrasse station in East Berlin. With forty-five minutes to spare,
they took a slight detour to Alexanderplatz. Carl produced a fat joint
and lit it, muttering that he needed it. Pengo could only laugh and wave
his hand at the offering. It was too early in the day to get stoned. Besides,
he was nervous.
   When Sergei greeted them in his office, Pengo already had an inkling
that this was not going to be an easy sell. He had to convince the Russian
that he was a valuable asset, someone worth investing in. Dob had
portrayed Sergei as someone who knew nothing about computers and
could only recite from a list of things the Soviets might be interested in:
compilers, source code and information from military computers. Pengo
could see that he might not get much further that day. But where Dob
tended to be passive, letting things happen to him, Pengo was more
aggressive and outspoken. He told Sergei that he was well versed in
VMS, and that he could bore his way into many different computers. By
way of example, he named Mostek, a U.S. semiconductor maker; Tera-
dyne, a Boston high-technology company; Thomson-Brandt; Philips in
France; and Genrad in Dallas. He went on to list more conquests:
SLAC, Fermilab, MIT, Union Carbide. If Sergei were interested in
individual accounts and logins, Pengo said, he would sell them. For an
account at the Jet Propulsion Laboratory, for instance, he suggested a
price of 150,000 marks.
   Sergei didn't seem to be responding, so Pengo came out with the rest
of his plan. First he suggested that he conduct hacking seminars for the
Soviets. An alternative would be that the Soviets set him up to conduct
safe hacking from East Berlin. Sergei said that didn't interest him. He
took out his order list and read aloud from it. His "customers" back in
Moscow, he said, needed UNIX and VMS source code and compilers.
VMS version 4.5 alone could bring the West Germans 250,000 marks
 ($125,000), compilers another 30,000 marks each. As for the thirty or
so pages of computer printouts the two young visitors had brought along,
186   A   CYKERPUNK
Sergei said he had no idea what to do with them. Nonetheless, in
addition to the customary 600 marks, Sergei handed Carl an envelope
packed with 100-mark notes and invited his two visitors out to a meal.
   As he and Carl left East Berlin and headed back to Dob's room at the
Schweizerhof, they counted the money. It was 3,000 marks, of which
Carl gave Pengo 1,000. Pengo felt less discouraged than self-important.
He had made progress. A Russian KGB agent had listened to him talk
for an hour. The Russian hadn't said yes, but then again he hadn't said
no, and when he examined more closely the information Pengo had
already provided, Sergei would be ready to deliver the VAX. Project
Equalizer was becoming Pengo's own first step toward becoming a paid
hacker-not just any hacker, but the best hacker in the world. He knew
that he could get what the Russians wanted, if only he had the right

It was Markus Hess, the one who liked to hack alone, who turned out
to be Cliff Stoll's intruder. When Markus and Hagbard had first met,
Hagbard had told of being in Fermilab and CERN. With a little guid-
ance, Markus learned to explore around West Germany and Switzerland
and the United States. Soon, he found a gateway into the Internet
through University College, London. He called Hagbard immediately to
have him come over and watch. The London computer was just the kind
of springboard into the Internet they were looking for. From there they
found a T ymnet node, and from there Hagbard discovered a way into a
bank of modems at Mitre Corporation. Neither knew precisely what
Mitre was, or even where it was, but it was a rich find. The modems at
Mitre, it seemed, saved the last number called, and Hagbard and Hess
could easily redial those numbers. That was how they first happened
upon a computer at the Anniston Army Depot and Optimis, a U.S.
Defense Department computer data base with information about military
studies. Optimis gave access to anyone who typed anonymous as a login
and guest as the password. Guest and guest did the trick at Anniston.
   In the middle of 1986, Hess and Hagbard discovered SLAC in Cali-
fornia. Hagbard was content to stay there and poke around a bit, engag-
ing the system managers in occasional on-line conversation and chatting
with others who had also found a way into the SLAC computer system.
But Markus wanted to see where he could go from there. From SLAC
he soon found a path that led to the University of California at Berkeley,
and Lawrence Berkeley Laboratory. The Berkeley university computers
foiled Markus's attempts to get in, but LBL was wide open. The labora-
tory liked to encourage outside researchers to use the LBL computers,
and passwords at LBL were often the same as user names. The situation
in Berkeley, in fact, seemed too good to be true. Security seemed to be
a joke to these laid-back Californians.
   Hagbard, with his pedestrian approach and his lack of programming
skill, didn't fully appreciate what the LBL computers had to offer. Mar-
kus, however, could spot bugs in programs and exploit holes in the
system. He had been playing around with the GNU Emacs "movemail"
program on the LBL computers and it dawned on him that the program
had been installed to run with superuser privileges. It was a major dis-
covery. It freed him to wander through the LBL system as much as he
liked. He began poking around in people's directories and looked for an
account to appropriate that had lain unused for some time. It was always
a better idea to use an existing account than raise eyebrows by creating
a new one. He found that someone named Joe Sventek hadn't logged on
for months and decided to become Sventek for a while. With superuser
privileges, he could change his password to anything he pleased. Benson
was one choice. Hedges was another. When he noticed that one of the
people whose directories he was peeking into was getting suspicious, he
logged out at once. From then on, he made certain to check to see who
might be on the system whenever he logged on, just in case a real system
manager was surveying the system's activities.
    Sometimes Hess connected straight to America; other times he went
via the University of Bremen, another security sieve, where he added a
new account for himself called Langmann and used the university's mo-
dems to dial out to the Datex-P international data network. From there
he got to LBL. Others discovered the fuzzy security in Berkeley. In the
early summer of 1986, a lot of different hackers broke into LBL, partly
because it was easy to get into, and partly because it was so easy to go
from LBL into other computers. And from LBL Hess could explore a
rich array of other computers. Military sites offered a special thrill. He
never logged on without checking a computer for its military informa-
tion. Before long, Markus had developed an expertise of sorts in U.S.
military acronyms.

By late 1986, Stoll had settled into a routine based on the hacker's
movements. Every time the hacker appeared, Stoll's pocket pager
sounded. Whatever the time of day and wherever he was, in the shower,
188   &   CY££RPUNK
on his bicycle or sitting down to breakfast, Stoll would drop everything,
call Tymnet to start a trace, then run to a computer, log on to the LBL
system and watch the intrusion. But the hacker wouldn't stay on the line
long enough for a trace to be completed. One Saturday in early Decem-
ber, Stoll came a step closer. A T ymnet trace showed one of the calls
coming from a transatlantic satellite, and to the satellite from the Datex-
P network in West Germany. From somewhere in Germany, it seemed,
the hacker called into Datex-P, asked for Tymnet and then connected
to U.S. computers.
   By now, Stoll was locked in a Faustian embrace with the CIA, the
National Security Agency, the FBI, the Air Force Office of Special
Investigations and the Defense Intelligence Agency. But even as he
probed his conscience, he was caught up in the excitement. As far as
Stoll was concerned, any international links to the hacker smacked of
intrigue. He envisioned spies muttering to one another in darkened
alleys. Yet back in Berkeley, there wasn't much he could do. For all his
vigilance, he was powerless. His success in tracing the hacking depended
entirely on the cooperation he got from those authorized to do the
tracing. He kept watching the hacker's every move. The wheels of the
Establishment forces turned slowly; all Stoll could do was conjure up his
own visions of "the other side" -and try to minimize the damage.
    If the hacker began to delete files or tamper with a system, Stoll could
use the UNIX "kill" command to disconnect him immediately. And
when the hacker seemed to be getting into computers containing sensi-
tive information, or tried to download sensitive files, or what Stoll could
only surmise were sensitive files, Stoll employed a low-tech but effective
solution: he took his keys from his pocket and dangled them next to the
wires connected to the hacker's line, shorting out the circuit for just an
instant. To the hacker, it looked as if his connection was being inter-
rupted by simple line noise and he would try again. Again came the
keys. Eventually the hacker would give up.
    The next step was to trace the call within Germany. Stoll stood by as
the network experts at T ymnet negotiated with the Bundespost authori-
ties to put on a trace. Finally, he got word that the call had been made
from the University of Bremen. The German authorities informed the
university that its computer systems were being infiltrated by an outsider.
 Flustered by the news, the university shut down all outside connections
for three weeks. But that didn't seem to stop the hacker. The next trace
showed he was coming from Hannover. But even that wasn't conclusive
proof that he was a German. Who was to say that Hannover wasn't just
a computer stopover from a flight that started in Botswana, Islamabad or
anywhere at all? The only way to continue the trace was to get a search
warrant, and that meant negotiating an agreement between American
and West German authorities. For Stoll, it meant more waiting.
   The New Year, 1987, came and went. Stoll's frustration was mount-
ing. As a research scientist, he knew all too well the virtues of patience,
but this research project had yet to yield any results. His main constraint
was his utter dependence on outside authorities and their cooperation.
And the fact that foreign authorities were now involved could only
complicate matters.

Hess began to realize that something was going on when he met Hagbard
and Dob for a beer one night in October 1986. "What have you been
doing with the LBL account?" Hess asked his friends.
   "Not much," Hagbard insisted.
   But they did appear to have a new scheme. They spoke of a Sergei
and a "Teddy Bear," and allowed that they had ways of making money
with computer accounts now. Hess was afraid that those two might abuse
the LBL account and cause it to go the way of so many other computers
locked tight when intruders were detected. He decided that even if Carl
and Hagbard hadn't done much with LBL so far, this was one hacking
sandbox he wanted for himself. A week or so later, he changed his LBL
passwords to just one: LBLHACK; he banished Benson and Hedges from
the system and didn't tell Dob or Hagbard. He wasn't concerned about
Pengo. If Pengo wanted to get into LBL he could probably do it on his
   Two weeks later, Dob proposed a business deal to Hess. He needed
him to make a copy of the Berkeley UNIX source code that Focus had.
There would be money in it for Hess. Hess agreed to do it. It seemed
like an insignificant favor. Berkeley UNIX, the variation of UNIX used
at most universities and research labs, was widely distributed and easy to
license. Making one copy didn't seem like a major breach. It took Hess
about a week to pull what he could of the UNIX source code together.
Carl picked up the software and Hess thought nothing more of it until
nearly a month later, in November, when Hagbard and Dob asked him
to take a walk with them one night outside the Casa Bistro. In a matter-
of-fact tone, Dob came out with what was going on. "The UNIX source
code was sold to the East," he said. "And that means you're in it with
us now." Hagbard stood a few feet back and said nothing.
190   A   CygERPUNK
   It confirmed what Hess had suspected: Dob wasn't jesting. In fact, if
anything, there was more than Dob was letting on. It seemed that this
was part of an established operation. Apparently Peter Carl had been
making regular trips to East Berlin to deliver the fruits of his friends'
hacking and general software piracy. Fancying himself the group's Wind-
macher, the one who makes things happen, Carl, it seemed, had spent
several weeks conducting a circumspect survey of local hackers on their
willingness to supply him with material for the East. Hess realized that
he had joined a core group composed of Carl, Dob, Hagbard and Pengo.
Hess gathered that Dob, while not doing much of the hacking himself,
had been over once or twice to meet Sergei as a technical expert. More-
over, Dob was the group's hub, the only one with a link to each member.
   Hess went straight home and the scene outside the bar played back in
his head. Great, he thought to himself, what a fine business you've
gotten yourself into now. Carl had already paid him 500 marks for the
software. Hess had consumed his share of spy thrillers, including gener-
ous helpings of Robert Ludlum novels, with their requisite abundance of
derring-do, double agents and double-crossing. He suspected at once that
he was suddenly subject to blackmail. And he didn't need to consult a
book of West German law to know that he was now part of something
not only illegal but extremely serious. Nonetheless, Hess had no inten-
tion of interrupting his hacking routine. He had just seen the movie War
Games for the first time, catching the film on German television. Thor-
oughly inspired, he had made it his goal to do what the movie's young
protagonist had managed to do: to get into NORAD, the North Ameri-
can Air Defense Command in Colorado. He still had the LBL computers
to himself, and they stood so wide open and so widely connected to
other computers that he was certain he would be able to find a way into
NORAD from LBL, given enough persistence.
   Hess would maintain to the end that hacking for him was merely a
thrill-seeking game. His interest in military sites had only to do with
their being the most forbidden fruit on a network, and nothing to do
with his being an errand boy for the Soviets. Hess continued to hack.

It was Cliff Stoll's girlfriend, Martha Matthews, who came up with a
brilliant ruse to catch the intruders. Martha was a twenty-four-year-old
Berkeley law student headed for a Supreme Court clerkship, her calm
bearing an ideal counterweight to Stoll's manic edge. If this rogue was
so persistent in his pursuit of military data, she argued, then they should
use his insatiable appetite to trap him. The idea was to round up volumes
of government data, disguise it as secret military information, plant it in
the LBL computer as bait, then entice the hacker by naming the false
files something irresistible like "SDInet."
    The two of them set about collecting hundred of pages of excruciat-
ingly dull and technical government documents they found lying around
on the computer system and put them into a single account on the LBL
system. They created new titles for each file, spicing up the language so
that it looked as if the documents described a new computer network
that would coordinate research on the Strategic Defense Initiative. In
one of the files they even included a letter inviting people to send for
further information on SDInet. For the price of a first-class stamp, inter-
ested parties could get documents on "SDInet connectivity require-
ments," "SDInet management plan for 1986 to 1988," or even a
"classified SDInet membership list." Inquiries were to be sent to Mrs.
Barbara Sherwin at LBL, a name of Stoll and Martha's conjuring. Be-
cause the information was so voluminous, they explained, it would have
to be sent via the U.S. mail rather than over computers. To speed things
along, they added that requests for documents must be received by the
lab no later than January 30, 1987. If the hacker jumped at that bait,
they might even get a return address.
    Stoll set up the SDInet file so that only he and anyone posing as a
system manager would have access to it. The next step was to sit back
and wait for the intruder to log on.
    A few days later, the hacker was back for a routine cruise of the LBL
system. Within minutes, he noticed the SDInet file. And sure enough,
he stayed interested for more than an hour. Soon thereafter, Stoll got
word that the trace had been completed to a certain residence in Han-
nover. But he wasn't given more details, and certainly not the hacker's
    Then, as if to provide positive proof that espionage was involved in
this hacker's activities, a few months later, well after the January 30
cutoff date, the lab received a letter addressed to Barbara Sherwin. The
stationery letterhead said Triam International in Pittsburgh, Pennsylva-
nia. The author of the letter was one Laszlo Balogh, and he asked for
specific classified information that had been listed in the bogus SDInet
file. Stoll decided that Laszlo Balogh must have had some connection
with the hacker, since Stoll and the hacker were the only two people in
192   •   CygE~PUNK

the world who could get at the SDInet file. Stoll's first call was to the
FBI. He was told to find a glassine envelope, presumably to preserve
fingerprints, and mail the letter at once to FBI headquarters.

At 6:00 in the evening on June 27, 1987, it was business as usual at
Focus Computer in Hannover. In the company's second-floor offices,
most of the employees were still engrossed in their work. Udo Flohr,
Focus's president, was preparing to leave when the office bell rang.
When Flohr opened the door, he was greeted by seven people-two
police investigators from the Bundeskriminalamt, West Germany's FBI
equivalent, four local police investigators from Bremen and a Bremen
district attorney, who presented Flohr with a search warrant. Computer
fraud was the charge, and the officials wanted to be shown to Markus
Hess's office. Flohr was too taken aback to be cordial; his first reaction
was to be defensive, even rude. He knew Markus to be an adventurous
spirit, and he wouldn't have been in the least surprised to hear that
Markus had done some hacking. But a lot of programmers hacked around
sometimes. It was to be expected. It was even part of Hess's job to test
security systems. Flohr couldn't believe that Hess's offense had been so
odious that it required the presence of seven people. He showed the
group up to the fifth floor, where Hess was working in his office. As he
left flanked by three of the officers, Hess looked more surprised than
panicked. Flohr immediately called the firm's attorney and asked him to
go over to Hess's apartment.
   The remaining officers spent the next several hours at Focus, search-
ing through everything in sight. When they began poring through stacks
of old newspapers, Flohr lost his composure. "What are you looking
for?"he demanded.
   "You tell us," one of the officers replied, as if Flohr were somehow
involved in a well-orchestrated cover-up scheme, and kept rummaging
through the newspapers.
    "Well, whatever it is, it can't be in there," Flohr snapped back.
   After nearly four hours, finally convinced that they had searched
everything possible, the police left Focus.
   Hess had never been one to panic, and the sudden appearance of a
clutch of police officers at his office door hadn't given him any reason to
break his calm. He said nothing during the fifteen-minute ride to his
apartment, and once they arrived, he insisted on waiting for his lawyer
before a search could commence. Within a few minutes, the Focus
attorney was on the scene, and the police investigators grew discernibly
more cordial. As the police searched through every nook in sight, Hess
made a pot of coffee and he and his lawyer waited for them to leave.
After two hours the search ended, and, since they weren't quite sure
what they ought to be looking for, the officers left with Hess's two home
computers and dozens of papers tucked under their arms. Markus re-
turned to Focus, where the search was still in progress, picked up a few
of his things and went straight to the Casa Bistro, where the weekly
hacker meeting was just getting started. He said nothing about what had
just happened. He was back at work the next day.
   Hess's attorney promptly filed a complaint on behalf of both Focus
and Hess, charging that a search had been conducted based on insuffi-
cient evidence.
   The surprise visit had been enough to scare Hess away from hacking,
but not enough to stop him from supplying Peter Carl with software for
Sergei. The money was so easy to get that it was difficult to turn his back
on it. Since late 1986, he had received reasonable pocket money-
always in cash and always in lOO-mark bills-in exchange for the various
programs he had handed over to Carl, many of which he had simply
copied from Focus computers. That, Hess reasoned, was a completely
separate matter from the hacking. Hess was less taken with the conspir-
atorial, adventurous side that the others seemed to enjoy so much. He
knew that Pengo and Dob had accompanied Carl to East Berlin, but he
had no desire to go. Aside from a family vacation to Yugoslavia he had
made as a child, Hess had never been to an East Bloc country, and he
had no intention to go now. He was content in his role as the software

It wasn't until Peter Carl produced the UNIX source codes from Hess
that Sergei made a first substantial payment of 25,000 marks-about
$12,500-in 100-mark notes: Carl noticed that on the lists he kept,
Sergei had crossed out "UNIX" and written "25,000" next to it. Carl
also noticed that a notation for IBM's VM operating system had been
crossed out and "50,000" had been scribbled next to it. Curiously,
"VMS" was also scratched out, even though Carl hadn't yet been able
to deliver it. Acknowledging Carl's confusion, Sergei remarked, "You
have competition."
   After that, the payments amounted to 3,000 marks here and 5,000
marks there for bits of software that Sergei thought were worth some-
194   ... CYEERPLINK

thing. Carl's practice was to keep at least half of the money for himself
and dispense the balance to the provider of the goods. After all, he
figured, why not keep at least half the money when he was doing the
dirty work? Carl had lost his job at the Hannover casino in the midst of
a highly publicized table-fixing scandal, so the income from the Soviet
business was becoming important to him. Still, no one was getting rich
from the scheme, and by the time the whole business was over, Sergei
would dispense roughly 90,000 marks to the group.
   After his meeting with Sergei, Pengo was hacking as often as he
could. He broke into the Japanese branch of Lotus Development Cor-
poration, but found nothing. Every time Carl dropped by on his way to
see Sergei, Pengo had to tell him he had nothing to offer. His ability to
hack uninterrupted was a function of his access to passwords and NUls,
and by the end of 1986, stolen NUls had a much shorter useful life. The
Bundespost had developed far more sophisticated means of tracking
down abused NUls. In fact, once a NUl was discovered, if it was shared,
it disappeared with a few days. So hackers had stopped sharing NUIs.
Pengo, too, guarded the ones he used closely. With a NUl in hand, he
could stay on line for at least two weeks, doing nothing but hacking.
Without one, he stayed off line, stoned most of the time, programming
here and there, occasionally hitting up Dob for a good meal and some
   Pengo was beginning to be extremely careful with his work. He tried
to maintain as low a profile as possible. When he was breaking into
computers on the Easynet, Digital Equipment's internal electronic mail
network and the place where he thought he might find VMS source
code, he probed only as much as he knew was safe without being de-
tected. But it was a frustrating exercise. Aside from the source code and
compilers, Pengo suspected Sergei wanted not just any military material,
but information on sensitive military networks, along with detailed in-
formation about the logical setup of the U.S. military machine, in order
to infiltrate it. But to find that kind of information, he wasn't sure where
to begin. He became less and less willing to share information with
others, such as those from the Hamburg/Chaos scene. He had already
made the mistake of telling Obelix and the others about the Digital
computer in Singapore, and within a few days the machine was locked
down. Now that he was to make a livelihood from hacking, the stakes
seemed higher.
    It wasn't ideology that made Pengo want to be a paid hacker for the
Soviets. Despite what he had said to Sergei, Pengo didn't feel particu-
larly strongly about politics. He hadn't exactly lied; he did come from
Berlin's leftist scene. And he was sympathetic, at least in vague terms,
with what Gorbachev was trying to achieve in the Soviet Union. He
had developed a strong sympathy for West Germany's eight-year-old
Green Party, which stood for the preservation of a threatened environ-
ment and opposition to nuclear weapons and nuclear energy, and was,
for a time at least, gaining a stronghold in Berlin. And Pengo was an
avid reader of the left-wing newspaper T ageszeitung. Occasionally, he
even fancied himself an anarchist, a fitting image for a young man from
Berlin. But he was skeptical of the others' talk of world peace through
hacking, their justification for Project Equalizer.
   Pengo suspected that Carl and the others were motivated mainly by
money. But Pengo had other reasons for what he was doing. Being a paid
hacker, traveling over to East Berlin and dealing with the Russians were
simply extensions of his life in front of the computer, the same life that
alienated and mystified his family. Pengo believed that he was upholding
a commitment to hacking as a thing in itself. Part of this idea had come
from reading Neuromancer, an intense and chilling science fiction novel
populated with high-tech lowlifes. Written in 1984 by William Gibson,
the book came to define what later was called cyberpunk. The novel's
protagonist, a drug-addicted computer cowboy named Case, is presented
with the opportunity to save his rapidly decaying life by breaking into
computer networks and stealing data. Neuromancer became Pengo's per-
sonal cyberpunk primer. After reading it, Pengo decided that if he hadn't
already established himself with a nickname, he would have chosen
Case. In Pengo's reality, working for the Russians held its own justifica-
tion. It was something Case would have done. Pengo knew vaguely what
was written in the German law books-that selling so much as a page
tom from a West German phone book to the KGB was espionage-but
he didn't really connect that to what he was doing. He was doing what
he had always done: hacking. And now someone was acknowledging his
   Late in 1986, a few weeks after Pengo's trip to East Berlin, Dob had
given Pengo another sizable chunk of money, thousands of marks to pay
his phone bill. The money was meant to defray the costs of hacking with
a legitimate NUl. But Pengo was still using stolen NUIs. He used the
money to buy a Sony Walkman, a telephone answering machine and
some records. In fact, he set up an impressive computer center in his
bedroom at his father's apartment. He was teaching himself UNIX and
had put together a computer, a decent modem and a printer.
196 A     CYEERPUNK
   It was a purloined NUl that got Pengo into trouble a few months after
his meeting with Sergei. When the Society of German Engineers noticed
that its Bundespost bill for Datex-P use was a hundred times the monthly
average, the association complained to the Bundespost, which traced
the unauthorized use of the NUl to one Gottfried Hubner in West
Berlin. One night shortly after the trace had been completed, Pengo's
phone suddenly went dead. He called the Bundespost to complain. The
line was connected again. On December 1 at 9:00 A. M., Pengo was
roused from his sleep by three officers from the West Berlin police. He
remained perfectly calm and asked to see their search warrants. After
several hours of searching every possible corner of Gottfried's apartment,
the police confiscated all the computer-related evidence they could find.
But they forgot one vital piece of equipment: the hard disk where all of
the computer's information was stored. A few weeks later they returned
for it, but Pengo just shrugged his shoulders and said he didn't know
what they were talking about. He was eventually accused of concealing
evidence and was fined for his unregistered modem. It had been a close
call, but Pengo's father was more upset than Pengo-not at his son's
transgression, but at the invasion of police into the apartment.
   As 1987 went by, Pengo kept working for Sergei, but he wasn't mak-
ing much progress in finding things that the Soviets might want. Once
a month he got a call from Peter Carl, who was on his way to his regular
meeting with Sergei whether he had something to take with him or not.
Showing up empty-handed was embarrassing. "Have you got anything?"
Carl asked Pengo each time. "No" was the response more often than
not. Pengo knew he wasn't living up to his part of whatever haphazard
arrangement he was involved in. But breaking into systems was getting
more difficult, mostly because NUls were so much harder to get. So he
asked Peter Carl to give him Carl's own NUl, explaining that without a
NUl he could always rely on he was at a distinct disadvantage. Carl
agreed. What Carl didn't expect, however, was that Pengo would use
the NUl for weeks on end and run up a bill of 4,000 marks-or $2,000
-in one month.
   Carl's patience with both Pengo and Hagbard was wearing thin. Carl
hadn't fully forgiven Pengo for spending an entire night at the Hotel
Schweizerhof the previous September, talking a big line, smoking lots of
Dob's hash and coming up with nothing particularly worthwhile. Sergei,
moreover, was pressing Carl to cut Hagbard off from the group because
of his heavy drug addiction and his habit of talking too much. Dob, too,
had proved a disappointment to Sergei. Dob's motivation was flagging
and he hadn't produced any source code for the Siemens computers he
was so expert in. Carl had even taken Dob to East Berlin a second time
so that Sergei could try to get him motivated, but he was increasingly
    Markus Hess was the only one who seemed to have Sergei's respect.
Sergei told Carl he was glad that Hess didn't take drugs and pleased that
Hess had come up with the source code to Berkeley UNIX. Most im-
pressive to Sergei was Hess's expertise in U.S. military computers.
    Indeed, Hess was the only one who came through with any regularity.
Gradually, it seemed, the scheme had deteriorated from the great hope
of the million-mark coup to just seeing what they could get away with.
Some of Sergei's more gullible purchases, in fact, originated with Hess.
As soon as Minix, an operating system with the look and feel of UNIX,
came into the Focus office in late 1987, Hess copied it and handed the
copy to Carl. Sergei handed Carl 4,000 marks ($2,000) for the software.
When Sergei later discovered, apparently from reading an American
computer magazine, that Minix source code normally sold for 120 marks,
he was furious. But that didn't stop Hess. A few months later, Hess filled
a disk with UNIX software, freely distributed at the last European UNIX
User Group meeting, and Carl sold it to Sergei for 2,000 marks. When
Sergei found out that it was all public-domain software, he warned Carl
never again to sell him something like that. Carl's reaction was indiffer-
ent. At least Hess was producing. Pengo couldn't even come up with a
list of passwords.
    Dob, too, was losing his patience with Pengo. More than anything,
Pengo just seemed irresponsible and self-absorbed. For months, Pengo
had been using Dob's Rainbow computer, a Digital computer similar to
an IBM Personal Computer. When Dob left for a vacation in Nairobi in
the summer of 1987, he asked Pengo to retrieve from the computer a bill
for work Dob had done and send it on to his employer. Despite profuse
promises, the matter slipped Pengo's mind and the bill remained un-
touched in the computer. Dob flew to Kenya via Amsterdam, but on his
way back from Kenya the flight made a stop in Hannover. He was
arrested almost immediately at the airport on charges of evading his
military obligation. Dob sat for weeks in prison in Hannover and it
wasn't until he saw his funds dwindle to nothing that he discovered
Pengo hadn't done the small favor he had asked of him. It translated to
a loss of nearly 10,000 marks for Dob. To further erode their friendship,
when the Rainbow broke, Pengo neglected to have it fixed.
    Then, in the late summer of 1987, the hack that was to focus world
198 •      CYEfRPUNK
attention on the Chaos Club happened. The NASA hack, as it came to
be called, had been started innocently enough by a group of Chaos
hackers as a simple, if all-consuming game of seeing how many comput-
ers could be reached on the space agency's vast SPAN computer net-
work. The NASA hackers did it by inserting an ingenious piece of
software of their own invention into VMS version 4.5. VMS 4.5 was
approximately the twenty-eighth iteration of Digital's operating system.
Each time an operating system is updated and new features are added,
new security flaws almost inevitably arise. Often, they go unnoticed.
When a small group of Chaos hackers discovered a flaw in VMS 4.5,
they seized on it and wrote what they called "the Loginout patch." It was
a clever little Trojan horse program embedded in the VMS source code
designed to collect passwords on any computer they entered. And it was
the very piece of code that Kevin Mitnick and Lenny DiCicco would
independently discover a year later and use to infiltrate scores of Digital's
own computers.
   In two computers named Castor and Pollux at NASA headquarters in
Washington, the group had found shuttle proposals and reports on
booster rockets. This was not particularly sensitive information, but
NASA didn't necessarily want its files opened up to a pack of curious
West German computer hackers.
   Obelix, who was part of the group, had been keeping a list of all the
computers they penetrated. But there came a time when the hackers lost
count of the machines they had infected with their loginout patch-it
could have been 150, or it could have been 500. They began to fear that
the Trojan horse could run out of control. Someone more malicious than
curious could find it, copy it, install it in another computer and cause
some real damage. It was time to notify someone official. Their first idea
was to inform Digital itself and leave the problem in the company's
hands. But after thinking it through, they decided that it might not be
the best idea: Digital was likely to respond with a call to the police.
Their next thought was to alert the American ambassador, but they
rejected that idea because, they decided, he probably wouldn't under-
stand what they were talking about.
   The NASA hackers decided to tell Wau Holland and Steffen Wer-
nerv. The two Chaos leaders weren't sure how to react, so they called
two television journalists, Thomas Ammann and Matthias Lehnhardt.
   At thirty-one, Thomas Ammann was well established as afree-lance
television journalist who did most of his work for "Panorama," a West
German television magazine. Ammann specialized in technology and,
of course, the German hacking scene. The thirty-seven-year-old Lehn-
hardt was Ammann's terse and cynical partner. The journalists sat down
with the Chaos leaders, who told them the story and said they were
planning to inform the authorities. Eager to save the story for them-
selves, the journalists tried to discourage them. They argued that it
would just bring an unwelcome investigation and put Chaos in a bad
light. But Wau and Steffen thought the material they were sitting on
was too sensitive to withhold. Eventually, they reasoned, it would come
out anyway. A few days later, they called a contact who, they believed,
might know whom to call at the Bundesamt fur Verfassungsschutz, a
West German agency roughly equivalent to a domestic CIA, and they
asked their contact to inform the agency. As it turned out, the agency
itself wasn't sure what to do with this odd, indirect confession. Without
informing DEC or NASA of the problem, the Verfassungsschutz simply
sat on the matter. Ammann and Lehnhardt decided to force the matter
into the open and rushed to put together a program for "Panorama"
about the hack.
    NASA refused to comment. Digital public relations, holding firmly
to a policy of refusing comment on sensitive matters, didn't return the
journalists' telephone calls. After a dozen attempts to rouse someone at
the Digital office in Munich, Lehnhardt finally spoke with a public
relations woman and challenged her directly. "This doesn't look very
good for DEC," he said, hoping to chisel loose a response.
    "I know, but we can't say anything," came the answer.
    On September 15 the show aired, declaring that NASA's top-secret
network had been violated. File footage of the shuttle hurtling into space
filled television screens throughout the Federal Republic, as if to imply
that as a result of their wanderings inside the NASA computers, the
hackers could manipulate a shuttle launch. The journalists even inter-
viewed Obelix, citing him as an outside computer security expert. Obelix
commented on the hackers as if he had no idea who they were. The
program threw Digital's Munich office into some disarray. West German
hackers discovering a gaping hole in Digital's operating system was hu-
miliating enough. The fact that they were romping undetected for
months was even worse. Finally, NASA confirmed that hackers had
been in the SPAN network but insisted that they had only browsed
through insignificant information.
    For his part, Pengo had kept his distance from the NASA incident.
A year had passed since his trip to East Berlin and he had stopped trying
to hack for Sergei. He had gotten no response whatever from his first
200   •   CygERPUNK
delivery, none of the feedback he was hoping to see. It was a big disap-
pointment, especially for someone whose ambition had been to become
the world's greatest hacker. Moreover, Carl's visits and phone calls had
tapered off, a signal that he wasn't placing much stock in Pengo's con-
tributions any longer. There had been no formal ouster, but Pengo
figured he was no longer part of the group.
   It appeared that the whole espionage affair was dying down. It even
looked as if the hackers weren't going to get caught. Pengo's attention
had turned to other things. He was looking for a way to go straight. He
had just started his own small company with a friend in Berlin named
Clemens. Called NetMBX, the small firm would write networking soft-
ware, do some consulting and run an electronic bulletin board. Starting
out with Clemens's equipment and the merest whiff of start-up capital,
the two budding entrepreneurs set up shop in a cramped little office on
the outskirts of West Berlin. Pengo was still enrolled at the technical
university, but he neglected his studies to flirt with the possibilities of
capitalism. He joked that he was becoming a "yoopie."

Stoll's frustration was mounting. Five months had gone by since the first
successful trace and the authorities were as reluctant as ever to let him
in on their side of the investigation.
   Finally, in late June, Stoll got a call from the FBI telling him that the
hacker had been apprehended, and his home and office searched. But
Stoll still couldn't pry loose a name. With the hacker caught, LBL no
longer had to lure him in with its flimsy security. That day, Stoll's lab
changed its passwords and tightened security. The attempts at the LBL
system stopped. Stoll was able to pat himself on the back and at the
same time expound to his colleagues on what he considered to be the
moral questions raised by this act of illegal computer joyriding. These
computer networks, Stoll argued, which scientific researchers, computer
scientists and students depend on for sharing information and for coop-
erative work and even for sending electronic love letters, assume a cer-
tain level of trust. And as the reach of the networks grows, so too must
the trust. How, then, can computer scientists expect to build and main-
tain open networks such as the Internet if someone like the Hannover
hacker is going to abuse the networks? For Stoll, the incident high-
lighted an insoluble dilemma: security versus information exchange.
   Stoll began to think about publishing the story of how he chased the
West German hacker. In early 1988 he sent out a book proposal and
prepared a technical paper for a computer journal on the various entrap-
ment methods he had used. But in April, a few weeks before his techni-
cal article was to appear, he was preempted when a story about the
Hannover hacker came out in Quick magazine, a West German hybrid
of People and Vanity Fair. The magazine identified the hacker as Matthias
Speer, a play on Markus Hess's name based on the fact that both names
-Speer and Hess-were also those of prominent Nazis. Mysteriously
enough, the story relied heavily on Stoll's logbook, the very logbook
that he had turned over to the FBI and the CIA months before. Some-
one, Cliff decided, most likely the authorities in Bremen, must have
leaked the contents of his logbook to Quick. Shortly after the Quick story
appeared, The New York Times put the story of the West German hacker
on its front page, mentioning espionage and the Pittsburgh letter for the
first time. The laboratory then held a hurried press conference.
    Within a few days, the story of the hacker who had roamed freely
inside sensitive U. S. computers was everywhere. One diligent reporter
even managed to root out Hess himself and journalists began to call his
    Hess took cover from the press. After his bust the previous June, he
had stopped hacking entirely. But he hadn't stopped supplying Peter
Carl with software to take to Sergei. In early 1988, Hess sent over a
magnetic tape containing a copy of X Windows, an advanced software
program, with the GNU Emacs program-the very software he had used
to gain superuser status at LBL-as an extra bonus. He was rewarded
with 2,000 marks. During his June interrogations, he figured out that his
LBL excursions had been traced. But he was taken completely by surprise
to learn that he had been monitored so closely. And he was shocked to
see his photograph in Quick, a picture that had been taken through the
window of his apartment while he sat at the computer. The Balogh
letter, in particular, created quite a stir in Germany.

As it turned out, Laszlo Balogh was another strange character.
    Howard Hartmann's wife had long since decided that Laszlo Balogh
was a "charming snake." But Hartmann was more understanding. After
all, he had known Balogh for nearly twenty years, and had even hired
the Hungarian emigre at his small geological surveying firm in Pitts-
burgh. The thirty-seven-year-old Balogh had immigrated to the United
States in 1959, settled with his family in Pittsburgh, attended local
technical schools and, through the sixties and seventies, worked at local
202   £   CYEERPUNK
firms as an engineering technician. The attributes Mrs. Hartmann
thought unsavory Hartmann decided had more to do with Laszlo's
"other" life.
   Laszlo Janos Balogh had always cultivated an aura of mystery. When
Hartmann hired Balogh as an $8-an-hour technician in 1984, the tall
and dark-haired Hungarian, who bore a slight resemblance to Omar
Sharif, told Hartmann that there would be periods when, because of his
work for the U.S. government, he would have to leave town for two or
three weeks at a time. Laszlo's resume stated that from 1966 to 1985, he
worked for the FBI and the CIA "as a consultant on counter intelli-
gence. "
   Hartmann already knew a little bit about Laszlo's other life. A year
earlier, Laszlo had been in the Pittsburgh newspapers when he blew the
whistle on former associates who had been planning to sell millions of
dollars' worth of stolen computer parts to Soviet agents in Mexico. The
suspects' defense attorneys had disputed Balogh's credibility and accused
him in turn of stealing diamonds. The three defendants in the illegal
exports case said Balogh had boasted to them of being a CIA hit man, a
black belt in karate and a bodyguard for two Kuwaiti princesses.
   Laszlo also had his own small firm, called T riam International, which
provided security and surveillance services. Triam had European connec-
tions, Laszlo told Hartmann, so he would be making occasional calls
overseas from Hartmann's business telephone, but he would promptly
pay the firm's bookkeeper for the calls. Laszlo also hired Hartmann's
daughter, Linda, to type occasional letters for Triam. Laszlo promised
that he would conduct all of his own work outside of normal office hours.
   The arrangement seemed perfectly acceptable to Hartmann. Balogh
was a hard worker and a competent technician. As long as Laszlo was
working in the interests of the U.S. government, Hartmann saw no
reason to object to the unconventional arrangement. Hartmann had
only one stipulation for Balogh.
   "I don't want anything that would bring any discredit to my firm,"
Hartmann warned.
   "Oh, that's no problem at all," Balogh responded.
   Hartmann had trouble figuring this complex character out. Why
would he work for so little pay when his other life offered so much
excitement and material reward?
   "There's only so much excitement I can take," Laszlo responded. He
told Hartmann that he didn't care much about the money, and he just
wanted to keep a low profile.
   Vagueness was the essence of Balogh's life. He never invited col-
leagues to his house in a middle-class neighborhood of Pittsburgh, where
he lived with his wife and three daughters.
   As far as anyone could tell, Laszlo always carried a gun, either in a
shoulder holster under his suit jacket or strapped to his ankle. But when
Hartmann asked him about it, his response was terse.
    "Packin' a rod, Laz?" Hartmann once teased.
    "Yup. Always" was the only response.
   And whatever Balogh was doing for the government had its material
rewards: he owned two Mercedes and a Jaguar, and he dressed in tasteful,
expensive clothing. When asked a question about anything personal, he
responded in the most circumspect of terms. His refusal to answer di-
rectly the most straightforward questions even started carrying over into
his work at the firm. If Hartmann asked Balogh whether he had per-
formed a certain calculation, hoping for a simple yes or no, he got a
convoluted and confusing response instead.
    There was also something slightly cartoonish about Balogh, an awk-
wardness that, together with his secret life, put Hartmann in mind of
actor Peter Sellers's Inspector Clouseau: smooth and charming in some
ways, clumsy and foolish in others. Once, for example, when Hartmann
was suing a former client who refused to pay his bill, he sent Laszlo out
to have a talk with the delinquent customer. A tape recorder was hidden
beneath Laszlo's coat. In the middle of the conversation, the microcas-
sette came to the end and the recorder shut off with a piercing electronic
    There were some people who mistrusted Laszlo, and they told Hart-
mann to watch himself. They were sure he was involved in shady activ-
ities, but no one ever had a scrap of evidence. When Hartmann got
curious enough to ask a friend who was a police sergeant to run a search
on Laszlo, it came up completely clean-perhaps too clean, the sergeant
told Hartmann. Someone could be protecting Laszlo.
    When Hartmann got a subpoena for his telephone bills in February of
1987, Laszlo was the first reason to come to Hartmann's mind. And the
federal prosecutor in charge of the investigation confirmed his suspi-
    "It's not you we're looking at. It's an employee of yours, Laszlo Bal-
ogh," the prosecutor said. He added that he would probably call him
again for Laszlo's personnel file. But he never did.
    In early 1988, Laszlo went to work for another engineering firm in the
area. In April, when news about West Germans and a mysterious letter
204   A   Cyg6~PUNK

to a Berkeley research laboratory surfaced, Hartmann maintained his
loyalty to Laszlo. Balogh told one newspaper reporter who tracked him
down in Pittsburgh that he had been answering an advertisement in a
trade magazine when he wrote the letter to LBL. He told the newspaper
that he had been a consultant to the FBI since 1966 and that he worked
as a weapons broker to such countries as South Africa and Saudi Arabia.
   Despite overwhelming evidence to the contrary, Hartmann wanted to
continue to believe that the matter had something to do with Laszlo's
work for the U.S. government, and that as long as he was working for
the right side he deserved some support. When reporters found Hart-
mann and started asking him questions, he would say only that Laszlo
had been an excellent employee. Laszlo called him to thank him for his
   "You're a true friend," Laszlo said.
   "Well, Laszlo," Hartmann replied, "I couldn't discuss things I had no
absolute knowledge about. The only thing I know for certain is that
you're definitely unique."
   Nonetheless, the incident with the letter had piqued Hartmann's
curiosity sufficiently that he asked his daughter about it. She had typed
about ten of Laszlo's business letters on Triam letterhead, and she re-
membered having typed the letter to Barbara Sherwin at LBL, but she
couldn't recall much beyond that.
   "Do you remember what material you were referring to when you were
typing out the request for specific information?" Hartmann asked his
   She thought about it. "I think it was a magazine," she responded. But
she couldn't be sure. "Laszlo took lots of material from magazines."
   It wasn't until Hartmann stumbled upon what he considered to be a
strong link between Laszlo and a theft of equipment from Hartmann's
firm two years earlier that Hartmann realized just how elastic Laszlo's
loyalties may have been. Now he wanted to find out all that he could
about Laszlo's side work. But when he had his lawyer call the assistant
U. S. attorney who had once asked for the phone bills, the attorney
would only say that the case was still under investigation.
   The finer points of Laszlo's life didn't matter much to Cliff Stoll. The
very appearance of a letter from someone with what appeared to be
shifting allegiances was enough to convince Stoll that there was some-
thing strange going on. If what Laszlo had told reporters was true-that
he was an FBI operative-then what was Laszlo doing responding to a
notice that only the hacker could have seen and presumably delivered to
the Soviets? After that one brief interview Laszlo disappeared, and the
FBI maintained a curious silence on the subject of Laszlo Balogh. Every-
one had a theory, but no one would ever know for certain how much of
what was said about Laszlo was true and how he found out about SDInet.
His links to the hacker who had been in the LBL computers were never
fully explained. And the only clue to what he was doing was on Hart-
mann's telephone bills. On April 21, 1987, the same day Laszlo wrote
the SDInet letter, he made several calls to Bonn, West Germany, to a
residence near the American embassy.

If it hadn't been for Nixdorf, the entire telex incident might not have
happened. And to Pengo it seemed that if the telex incident hadn't
happened, the entire spying business might have faded away.
    In early 1988, the local police in Munich needed to set up a high-
speed data link between Munich and federal police headquarters in
Wiesbaden for all their telex traffic. It was an unrewarding, time-con-
suming programming job because of the ad hoc nature of telex commu-
nications. The police asked Nixdorf, a large West German computer
company, to take it on, but Nixdorf was too busy, so Nixdorf farmed it
out to a small software company in Berlin. That company had qualms
about working for the police, so it subcontracted the work to NetMBX,
Pengo's start-up company, for 40 marks an hour. Suffering from lack of
business, little NetMBX was in danger of folding before it had a chance
to get off the ground. Clemens and Pengo were only too happy to take
on the work. Pengo enjoyed the irony of it all. Not only was he a
prominent West German hacker whose name was known to the author-
ities, but he was one with leftist leanings who had been doing business
with the Soviets. And now his firm was going to be responsible for setting
up communications between two state police headquarters.
    Mindful of the problems Pengo might encounter if he were so brash
as to go to police headquarters in Munich himself, in late March he
dispatched Clemens. Clemens spent two days in Munich working on the
link. When he had finished, he asked the police in Munich to send a
series of sample telex transmissions, which he then copied onto tape and
took with him back to Berlin for use in testing. When he got home,
Clemens printed out the telexes and showed them to Pengo. Pengo was
highly amused. Among the missives were a death threat allegedly sent
by the Red Army Faction terrorists to West Germany's research minister
and the travel schedules of two top police officials, complete with the
206   •   CYE£RPUNK
security measures planned for the trips. When Hagbard came to Berlin
to visit a few weeks later, Pengo couldn't resist showing him the telexes.
    Hagbard wasn't doing quite as well. Since early 1987, he had been in
and out of psychiatric hospitals and detoxification centers. He was re-
covering from a difficult love affair with an American diplomat who had
returned home. And he was more convinced than ever of the Illuminati's
grip. But now he was turning the world's problems into his own scourge.
He believed that the AIDS virus had originated with him as a means for
annihilating the Illuminati, that the letters were in fact short for "Anti-
Illuminati Destruction System." To make things worse, Hagbard's in-
heritance had long since run dry, and his drug dependency was acute.
So when a new set of journalists from Hamburg approached him while
working on a story about computer break-ins, he told them that he was
one of the most powerful and talented hackers they would ever encoun-
ter, but that his story came at a price.
    He told them he could steal his way into most anything. They wanted
to see proof, so they paid him 500 marks to come to Hamburg and stay
for a week in a good hotel. They even paid for Pengo to come, too.
Germany's computer hackers were accustomed to seeing some reward for
revealing their secrets. Usually, these were informal arrangements-a
filling meal in a nice restaurant, round-trip train fare in exchange for a
window into hacks at CERN or Thomson- Brandt. The two journalists
told Hagbard and Pengo that if they could publish the story in a German
magazine such as Quick, substantially more money would be forthcom-,
    While they had Hagbard there, the two journalists sat him down with
his back to the camera and interviewed him about what it was like to be
a hacker. They also had him compose a hacking manifesto of sorts,
complete with his theories on Hagbard Celine and the Illuminati. In his
seven pages of rambling prose, Hagbard disclosed that the National
Security Agency was operating a secret department called OSAD, the
Offensive Software Applications Department. There, Hagbard said,
NSA was preparing for a war of the future, a computer war fought with
"soft bombs," computer viruses. Hackers, therefore, represented the
pivot of the world's fate. "Yes," Hagbard concluded, "the computer war,
our soft war, has begun."
    But when the journalists put Hagbard in front of a computer, he had
trouble backing up his extravagant claims. He tried and failed to get into
Bolt, Beranek & Newman, a research center and think tank near Bos-
ton, and again into the Jet Propulsion Laboratory. Hagbard was simply
going to have to produce something more convincing.
   So the telexes came in handy. When Pengo showed him the telexes
that Clemens had collected, Hagbard seized the opportunity. He per-
suaded Pengo to let him have some of them and presented them to the
two journalists as a product of hacking, claiming that he had gotten
them from Pengo, who had broken into the police computers in Munich.
   By now it was July 1988, three months since the "Matthias Speer"
story. But rather than diminishing with time, the story of Markus Hess
and the system manager in Berkeley was still making headlines in West
Germany. The publicity was beginning to frighten Pengo. It wasn't so
much the Quick story itself that worried him as the mention of the Laszlo
Balogh letter. Pengo knew nothing of SDInet or Laszlo Balogh, but it
made Pengo queasy. Everyone in the country was talking about espio-
nage. Pengo had a brief conversation with Hagbard about it, and he
asked Hagbard how much he thought the authorities knew about their
role in the entire affair. But it was useless. Hagbard was crazier than
ever. He was beginning to talk to too many journalists about too many
things. Pengo had a brief conversation with Dob to discuss Hagbard's
loosening lips. Dob assured him that no one believed anything Hagbard
said anymore.
   But Ammann and Lehnhardt, the two television reporters who had
broken the Chaos NASA-hack story, had heard about Hagbard's telex
story and were on their way to Berlin to question Pengo about it. Am-
mann was also curious about Balogh. He was convinced that something
was going on involving the East. And when the head of "Panorama"
suggested to Ammann that he look into this story about police telexes,
Ammann agreed immediately. The perplexing telex story was reason
enough to go to Berlin for the day and talk to Pengo.
   Pengo didn't show up for the interview. When the journalists called
him, he apologized and said he had forgotten about it, so they made a
date for later that night. Ammann, Lehnhardt and a third reporter,
Gerd Meissner, a gifted young free-lancer, took Pengo to a crowded and
smoky cafe, a student hangout near West Berlin's technical university.
Ammann was struck by the streetwise ways of Pengo, who did not con-
form at all to the picture of hackers Ammann had developed back in
Hamburg. Instead of a social maladroit, Ammann was greeted by an
amiable and appealing nineteen-year-old. After ordering one of the most
expensive dishes on the menu and falling on his food like a starving
208 •      Cyg£RPUNK
man, Pengo set the record straight about the telex incident. Fine, the
journalists said, but now they wanted to know about Markus Hess, a
mystery in that his name had never before surfaced in connection with
West Germany's hacking scene.
    "What about the Hess story?" Ammann asked. "How well do you
know Hess?"
    "I don't know him at all," Pengo responded, fairly shouting to raise
his voice above the surrounding din.
    "Well, what about hacking and espionage?" Ammann asked.
    "I don't know anything about hacking and espionage," Pengo in-
   Ammann wouldn't give up. "What about those keywords, NORAO
and SOI?"
    Pengo shrugged.
   Ammann changed his tack. "What about the letter from Balogh?"
    By this time, Pengo had consumed three tall beers and he was begin-
ning to loosen up. The question about the Balogh letter sent him over
restraint's edge.
    "Okay," he said calmly, as if he had hit upon an afterthought, "it's
true that some software has been passed to the Soviet Union. But it
didn't amount to much." He told them that a contact had been made in
East Berlin with a Soviet agent named Sergei. Lehnhardt said nothing
but Meissner nearly slipped out of his chair. Ammann's first reaction was
not to believe it, and he asked Pengo to back up his claim. So, main-
taining his cool, speaking as if he might be noting a change in the
weather, Pengo told them the story of Helmstedt.
    Helmstedt, a quaint border town in West Germany, was the police
checkpoint to the East German highway leading to Berlin. It was where
Westerners traveling to Berlin by car were stopped and their papers
scrutinized by a lineup of stiff young border guards. As Pengo had heard
the story from Peter Carl, when Carl wanted to make his initial contact
to agents in the summer of 1986 he slipped a small piece of paper inside
his passport at Helmstedt on which he had written something in code
expressing his desire to talk to someone about technology transfer. When
he got his passport back, more than an hour later, he found a second
note inside containing a telephone number, the number of a contact at
the KGB.
    The Helmstedt story, of course, had been another of Carl's grand
illusions, but Pengo believed it. So did the journalists. It was classic, the
journalists decided, and it must be true. What, Ammann asked the
surprisingly calm young hacker, did he plan to do now? Pengo looked as
if that were the first time he had considered the question. Well, he said,
he had been out of the action for more than a .year now, and he was
hoping the whole matter would simply die on its own.
    This was one story Ammann didn't want to lose. But it also presented
a tricky situation: he could hardly base a potentially explosive story on
the word of one nineteen-year-old, but if he started sniffing around, he
could endanger Pengo. Moreover, it hadn't taken any grand effort to pry
the story loose from Pengo. Given the fact that he had forgotten about
the interview at first, Pengo probably hadn't intended to bare his soul.
There was no telling, given a few beers, whom else he might entrust
with this information. And as a free-lancer, Ammann didn't have the
resources to post Pengo to Elba or another suitably isolated island until
he could confirm the story.
    On the one hand, Ammann was surprised that something like this
hadn't happened earlier. He and other journalists who followed the
hacker scene had often discussed the likelihood that hackers would be-
come spies. But Ammann saw that the confluence of friends, circum-
stances and drugs had created a perfect setting for what Pengo was
describing. Hagbard seemed to be dependent enough on drugs to look
for any route to easy money, and Pengo struck Ammann as a quirky mix
of naivete and shrewdness. Aside from seeing Pengo each year at the
Chaos congress, and observing his steady rise to the top of the hacking
scene, Ammann didn't know much about the smooth and accomplished
hacker from West Berlin. Mostly he knew him as a bright, opinionated
young man, something of a would-be anarchist, somewhat distanced
from the rest of the Chaos group. Now he seemed at once oblivious to
the consequences of what he had done and hardened in a way that
Ammann hardly expected from a teenager. It didn't surprise Ammann
to learn that this youth had been traveling to East Berlin to deliver the
fruits of his electronic marauding in America and Western Europe. If he
was telling the journalists this much, it was highly likely that there was
a lot more to the story.
    The three journalists told Pengo he should think carefully about what
to do, but that just letting the situation sit was probably not the best
idea. By the time they left the cafe, Pengo was trembling visibly.
    A few days later, Ammann and Lehnhardt returned to Berlin, and by
then Pengo was truly anxious. Pengo told the journalists that he wanted
their help in getting out of the entire mess. One way Ammann knew to
assure an exclusive story was to gain Pengo's trust, even to go so far as
210   •   CygERPUNK
to put him in touch with the right people. Ammann called Ulrich
Sieber, a law professor, attorney and computer crime authority in Bay-
reuth, a small university town in southern Germany best known for its
annual Wagner festival. The thirty-seven-year-old Sieber hardly seemed
the most natural choice to defend a hacker spy from Berlin. He was well
respected in the Federal Republic as an attorney who represented cor-
porations deprived of their rightful profits at the hands of software pi-
rates. But Ammann also knew him to be fair and, in his own way, open-
minded. When Ammann called Sieber, the journalist was circumspect
in his description of the case. He explained simply that there was a
hacker who had a problem and needed advice. With no specifics in
hand, Sieber agreed to devote a Saturday to discussing the case in person
with Ammann and the unnamed hacker.
   Gerd Meissner was put in charge of chauffeuring Pengo from Berlin to
Bayreuth, a six-hour drive. He had to pick Pengo up at his father's house
at 5:00 A.M. in order to get to Bayreuth in time. When they arrived at
Sieber's university office, Ammann was already there. A relentlessly
cheerful and proper man, Sieber was friendly and gracious to the three-
some, and he made clear his desire to help with whatever the problem
could be.
   Pengo appeared calm, perhaps even a bit uninterested in the whole
affair. "Yeah, there are some problems," he told Sieber as he sat down,
and he gestured to his two companions. "These guys say I should talk to
you about it." Sieber asked Pengo to tell him exactly what had hap-
pened. Pengo was reluctant at first, even slightly distrustful of Sieber,
whom he knew to be no proponent of computer hacking. But Sieber
assured him that his story would be safe with him. So, rolling cigarette
after cigarette, Pengo came forth with the story. He told of Carl's con-
tacts with Sergei, of his own visit there and of the various pieces of
software that Carl had delivered to East Berlin. He described his own
deliveries: the security program, the assembler from Thomson-Brandt
and the session logs. And he described what he thought the others had
sent over. He told of American military sites, of source code and of log-
ins to sensitive computers throughout the world.
   When Pengo was finished, Sieber sat back in his chair. The two
journalists spoke first. They wanted to know what the options were.
They asked what Pengo's rights were and what loopholes, if any, Sieber
could find for him in Germany's computer crime statute. Pengo inquired
about the possibility of trying to forget about the whole thing, and
disposing of evidence.
   Expert though he was, Sieber wasn't quite sure where to start. He was
certainly riveted by the tale. As a youth, he had taken his English studies
partly by way of Ian Fleming's James Bond stories, and in at least one
small aspect of his work he had managed to find a connection between
the world of international intrigue and the relatively obscure realm of
computer security. In his otherwise dry writings on the topic of computer
crime, Sieber had warned a decade earlier that hackers could easily fall
into the hands of the KGB. Few people believed Sieber at the time, and
some of his most ardent detractors had dismissed him as a publicity
seeker. So for Sieber it was a clear vindication of his theories to learn of
such a case.
   First, he issued a polite but firm warning that if Pengo had any inten-
tion of destroying evidence, he had come to the wrong lawyer. Then,
thinking out loud, he expressed his first reaction to Pengo's predicament.
It was clearly not so much a hacking case as one of espionage. It was
patently criminal, and a serious offense at that. As for Pengo's idea to
forget about it and let the whole thing blow over, Sieber said that was
one option to consider. But as a pragmatic lawyer he advised against
it, explaining that it could hang over Pengo's head for years and
could someday be used by the Soviets to blackmail him, even ten or
twenty years hence, when Pengo had a family and a job to protect.
Another problem was that some of Pengo's friends might already have
informed others of what they had done, thus placing Pengo in a
great deal of danger. Pengo said he thought it was entirely possible
that Hagbard had already said more than he should, and that worried
   Sieber's second suggestion to Pengo was that he inform the authorities
himself. As it turned out, West German law contained an amnesty
provision for espionage. If a West German citizen-turned-spy gave him-
self up to the state before the crime was discovered, and if turning
himself in prevented further damage to the Federal Republic, he could
receive full immunity from prosecution. Sieber was familiar with the
provision because he had tried and failed to convince the German par-
liament to adopt a similar provision for hackers when it adopted its
hacking law in 1986. No sooner had Sieber mentioned it than Pengo
latched onto the idea. It seemed like a perfect escape hatch. It had been
a long time since he had called Dob and Carl his friends. Hess he hardly
knew, and Hagbard seemed too confused and unpredictable to protect in
any case. So he had no qualms about betraying them. Sieber warned
that if Pengo did turn himself in, amnesty wasn't necessarily guaranteed.
212   •    CygfRPUNK
There would always be the risk that the authorities would choose to
prosecute him anyway.
   Sieber told the group he knew of someone from the Bundesamt fur
Verfassungsschutz, the German domestic secret service. As it happened,
Sieber had been at a conference some time earlier where he had met a
Verfassungsschutz officer and sometime security consultant who had
given the lawyer his card. Sieber asked Pengo if he would like him to try
to reach the man at home. Pengo nodded, and Sieber disappeared into
the next room to make the call. He reemerged ten minutes later to
report that he had reached the officer and, just as Ammann had, ex-
plained the problem without mentioning a name. The secret service
contact needed some time to think things over and would call back
within the hour. So the group settled down to wait for the return call.
Ammann found the incongruity of the scene entertaining. Here was one
of West Germany's leading experts on computer crime, in his conserva-
tive blue suit, seated across a desk from one of West Germany's most
prominent hackers, clad in his Berlin black sweatshirt and black jeans,
nervously rolling cigarettes. And Sieber seemed more than willing to
take on this unpredictable young man's case. It was in some ways, he
decided, a perfect hack.
   Only once in Sieber's questioning of Pengo did incredulity register on
the lawyer's face. "Did you have any scruples about what you were doing
and did you consider whether it was unethical?" Sieber asked Pengo.
   "I don't care about ethics," came Pengo's flat response. "If it's Russian
interests or Western interests, I don't care about that stuff."
   After a pause, Sieber suggested that such sentiments would hardly
constitute a solid defense, and he recommended that Pengo not stress
that line of thinking in his talks with the authorities.
   An hour later, Sieber's telephone rang and the secret service contact
said he was willing, just as Sieber had been, to devote a day of his
weekend to hearing about the case without knowing any names. He said
he and two colleagues would drive to Bayreuth from Cologne the follow-
ing day, Sunday. Ammann and Meissner drove home, and Sieber sug-
gested he get Pengo a room in a nearby hotel.
   Hartmut Pohl, one of the Verfassungsschutz's computer security spe-
cialists, showed up in Bayreuth early the next day with two colleagues.
One of Pohl's companions was a senior officer and more of a generalist.
The other,· an old hand at intelligence work, was to become Pengo's
regular contact. Sieber met with them privately first and, still withhold-
ing his client's name, attempted to strike a deal with them. Finally, it

    was agreed that if Sieber's client agreed not to dissemble in the slightest,
    then his chances for amnesty would be excellent.
       And so the interrogations began. The first session that Sunday lasted
    four hours. Sieber wasn't one to underestimate the reach of the KGB
    and the East German secret service when they wanted someone out of
    their way. Concerned for his client's safety, when the interrogation was
    over Sieber insisted that Pengo not travel by train back to West Berlin,
    a trip that would take him across East Germany, and that he fly instead.
                                      .& ... .&

    Pengo sensed he was in for a difficult time. Now that he had exposed
    the story, he knew things were somehow beyond his control. At the
    same time, in finally coming out with it, he had unburdened himself
    and, in a way, he felt better. But this meant that a new burden had
    landed on his shoulders: the weight of having set things in motion, of
    having betrayed the others in order to save himself. Yet he wasn't spend-
    ing much time worrying about that. He was somehow sure that the
    others had their own plans for getting out of the predicament.
       Pengo entered into a form of receivership. The matter was an official
    case, with a specific, no-nonsense agent from the secret service assigned
    as Pengo's official contact. And Pengo was expected to comply with any
    of the contact's demands. If the contact called and told Pengo to get on
    a plane the next morning for two days of interrogations in Cologne,
    Pengo was expected to put aside all other plans and do so. And the three
    television journalists who had pried his confession loose in the first place
    had taken a proprietary attitude toward the situation. Pengo got the
    impression that the journalists wanted to have at least one of the trio at
    his side during his every waking moment, lest another journalist should
    start sniffing around the story.
       But as a practical matter, nothing really changed. Pengo's little com-
    pany continued to take on sporadic work, and, much to his amusement,
    even did some consulting for the German commands of the armies of the
    United States and France in Berlin. He moved out of his father's place
    and into a small studio apartment of his own in Kreuzberg, a neighbor-
    hood known not only for its dense population of Turkish immigrants but
    also as the focal point of Berlin's counterculture. It was a scene with
    strong appeal for Pengo; he had always felt that Berlin was the only place
    to live, and Kreuzberg had become the only place to live within Berlin.
       In fact, it seemed that those around Pengo-the German officials,
    the three journalists and Sieber the lawyer-were exhibiting far more
214   A   CygERPUNK
concern over the matter than Pengo himself. As if to thumb his nose at
the entire Western Establishment, just two weeks after his confession,
to the great consternation of those in the government working on the
case, Pengo accepted an official invitation to consult for the Bulgarian
Cybernetic Institute in Sofia. Before the trip, Pengo had to visit the
institute's branch in East Berlin for a meeting. Remembering Sergei's
invitation to come by whenever he liked, Pengo decided to try Carl's
passport trick with his own. Hoping to bypass the process of getting the
one-day advance permission that was required of every West Berliner,
Pengo went straight to passport control at Friedrichstrasse, handed the
guard his passport and said he had an appointment. Unimpressed, the
guard pushed the passport back at him and told the young West German
to get the prior clearance. Either the trick worked only for Peter Carl,
Pengo thought, or the Soviets had become alarmed by the attention
Hess had attracted and now considered the espionage operation closed.
   The trip to Sofia was uneventual enough, at least as far as the work
itself was concerned. Pengo donned a "yoopie" outfit of jacket and tie,
and spent the week making formal presentations about computer net-
works to the Bulgarians, who seemed hungry for knowledge of any West-
ern technology.
   Several months after Pengo's unexpected trip to the prosecutor's office
and his confession, he discovered just how fortunate his timing had
been. It turned out that Hagbard had done exactly the same thing at the
advice of his own lawyer a few weeks before Pengo did. The handling of
the two cases was strikingly parallel. Hagbard's lawyer, too, had invoked
a section of the amnesty provision in the espionage laws and had his
client turn himself in. It was a close call, to be sure. If Pengo had waited
even another week, the authorities might have been ready to open a
formal case against him, and might not have offered him the promise of
lenient treatment.
   A few weeks after Pengo's initial confession, the interrogations in
Cologne became more intensive. His trips there usually consisted of an
evening flight from Berlin to Cologne, where his contact would pick him
up and deposit him at a hotel outside of the city. He was under strict
instructions to make no telephone calls from his room. Promptly at 9:00
the next morning, the questioning would begin.
   It wasn't so much the idea of being beholden to the authorities that
began to fray Pengo's nerves, but rather the psychological pressure his
interrogators exerted. They played out a typical good-cop, bad-cop rou-
tine with him: one would cajole while another threatened. They refused
to believe that he hadn't done it for the money. Pengo could see they
didn't understand that he had simply wanted to become the world's
greatest hacker. He tried to explain what hacking was all about. He
suggested they read Neuromancer. Bearing in mind that his chances of
getting off were vastly improved if he told the complete truth, Pengo
tried to follow the Bayreuth lawyer's suggestion and be as forthcoming as
possible, but there were times when his recall was less than perfect. Not
only had two years elapsed since Pengo's one and only delivery to Sergei,
but apparently the years of hash smoking had taken something of a toll
on his memory.
    One pivotal subject on which Pengo was less than direct was that of
the magnetic tape he had given to Sergei. Since he hadn't had the proper
equipment for making such a tape, he had had to ask someone else to
make it for him. That would have meant calling a friend and asking him
to copy certain files from a computer onto the tape. Pengo had faced the
basic technical problem of transferring information out of a computer in
a portable form. And for that, he had needed the cooperation of some-
one with access to the right equipment.
    At first, Pengo told his interrogators that he thought he had obtained
the tape from Obelix in Hamburg. But as soon as he had suggested
Obelix as a possible source of the tape, Pengo reversed himself and
denied it. Perhaps he had second thoughts about implicating his friend,
who could face harsh penalties if he were found to have collaborated,
even unwittingly, with Pengo's spying adventure. Not only could Pengo
 not recall who had made the tape for him, but, he claimed, he wasn't
sure what was on it. Pengo tried to explain that just as he had no tape
 mounting equipment at his disposal, he also had no tape reading equip-
 ment, so once he got the tape, he had no way of seeing what was on it.
 But the men questioning him were skeptical. Perhaps to see if Pengo's
 sudden bouts with amnesia might be cured with a little gruff treatment,
 an agent from the army's Secret Service branch attended one of the
 interrogations and scowled through the session. When Pengo's memory
 faltered, the military officer threatened that if Pengo continued with his
 evasive answers, he would flush the young hacker down the nearest
 toilet. Pengo was unperturbed, whereupon the Verfassungsschutz offi-
 cers, more embarrassed than anything else, apologized to Pengo for their
 crusty military colleague.
    Still, the young Berliner was hardly relaxed about his fate, which now
 lay in the hands of the very authorities he had been trained to mistrust.
 But he had to laugh at the conspiratorial manner in which his interro-
216   ~   CYG6RPUNK
gat ions were occasionally conducted. During one trip to Cologne, he
was driven from the airport to a secluded patch of woods where a second
vehicle awaited him. Upon arrival, he was shown several photos of
strangers and asked to identify them.
   Often, after a day of exhaustive questioning the interrogators and
their subject repaired to the hotel bar for a more relaxed probe into
Pengo's political leanings. He was as frank with them as with anyone
else. As Pengo saw it, these men were "black-and-white people." They
viewed Communism simply as bad, while Pengo liked to believe he was
more open-minded. Perhaps because of his upbringing, or perhaps be-
cause of his firsthand experience with the border life in Berlin, he rec-
ognized shades of gray in politics. If anything, he told them, he was a
leftist. The authorities seemed always ready to listen to what Pengo had
to say, if not quite ready to be converted to what they considered to be
the young man's wayward political ideals.
   The interrogations continued through the summer of 1988, after
which the Bundeskriminalamt-the investigative arm of the federal po-
lice and the rough equivalent of the American FBI-was on the case.
For each new set of questioners, Pengo repeated his story. Time and
again, he was asked about how the whole affair was conceived and who
made the initial contact with the Soviets. He was asked repeatedly to
describe the core group of five that eventually emerged. Pengo said that
he had come into the group later than the others, and that after his own
delivery to Sergei, the Soviets' lack of feedback had disappointed and
discouraged him. His involvement had diminished gradually, and by the
time he engaged Sieber, he had been out of the business for more than a
year. But he couldn't say what the others were doing. And he refused to
comply with what he saw as unreasonable and irrelevant demands.
When his Secret Service contact asked him to get them some informa-
tion about Dob's trips to a Berlin brothel, Pengo refused in one blunt
phrase: "I'm not your spy."
   In response to repeated demands for more details about the precise
workings of the group, Pengo insisted that he knew little about the
others, with the exception of Hagbard and Dob. And since their falling
out over the Rainbow and Dob's forgotten invoice, he had had very little
contact with Dob. He said he knew that Dob owned a gun and could be
dangerous. He believed that Carl had been over to East Berlin at least
two dozen times, and he knew that Dob had gone at least once as well.
He said that he knew Markus Hess only vaguely, and that he had scant
knowledge of Hess's breaking into Lawrence Berkeley Laboratory. He
said he might have been inside LBL computers himself once or twice,
but he insisted that he knew nothing whatever of SDInet or the myste-
rious Laszlo Balogh from Pittsburgh.
    To an outsider Pengo might have maintained that he was perfectly at
ease during the interrogation process, and that he was comfortable with
being in league with the authorities. In fact, if he were asked to come
up with a moral principle that guided his actions, he would have re-
sponded that he had been working for one side and now he was working
for the other. It was as simple as that; morality was not really the issue.
    He continued to work at his little consulting company, which was in
the process of merging with another small Berlin start-up. But the effect
of the interrogations was beginning to show. Even without Dob as sup-
plier, Pengo's dependence on hash increased. Throughout the autumn
he was high more often than.not, and he drank liters of coffee every day.
He was a classic "stress smoker," as the Germans say, lighting a cigarette,
stubbing it out after two or three puffs, then lighting another one. He
made dates with people and forgot about them, or simply sloughed
friends off. When he bumped into his ex-girlfriend at a concert one
night he was skittish and jumpy and muttered something about being in
trouble with the secret service.
    If Pengo was beginning to have a hard time with life as a snitch, the
interrogations with the authorities were taking a deeper toll on Hagbard.
Each of his attempts to overcome his dependence on drugs had failed.
He was living in a halfway house at Hannover and had no money. His
ever-narrowing circle of loyal friends was at a loss for how to help him.
When they were with him, he spoke of nothing but the great conspiracy
that reigned over their lives. And there was little family support for
Hagbard. His stepmother, his sole remaining parent, wanted to have
little to do with him. Not only had he squandered his generous inheri-
tance, but she was certain that he was selling some of the valuables he
had inherited as well. More than once he had visited her and intimated
that he would like to have certain paintings that had belonged to his
father. She suspected that he planned to sell those as well, and told him
she would no longer allow him inside the apartment if she was alone
when he came.
    Hagbard had explained his predicament to Johann Schwenn, the
Hamburg lawyer to whom he had turned for help. To Schwenn, the
calm man with the soft voice didn't seem any more disturbed than other
young people in trouble with the law. Schwenn was a prominent civil
 libertarian with a busy caseload; he had little time to make deep inquiries
218   ~   CYG£RPUNK
outside of what his client told him. Though very thin, with skin so pale
it was nearly translucent, Hagbard was polite, self-consciously neat and
well spoken. And in a sign to Schwenn that this young client had some
sense of right and wrong, Hagbard hadn't been eager to betray his
friends. Schwenn didn't inquire much into Hagbard's personal situation,
and the young man offered no details.

In late 1988, the unusual case of the hacker spies landed squarely, if also
fairly arbitrarily, in the hands of Ekkehard Kohlhaas, one of a handful of
federal prosecutors at the attorney general's office in Karlsruhe who
worked espionage cases. The diffident forty-three-year-old prosecutor
had recently arrived at his post after several years in the public prosecu-
tor's office in Karlsruhe, followed by a stint at the Ministry of Justice in
Bonn, where he worked on hostage cases. He didn't know whether or
not to consider himself lucky to have this high-profile case. It was a first
for his small department; not only was there no precedent in the Federal
Republic for such activity, but there was no precedent anywhere. More-
over, West Germany's attorney general was placing an unusual amount
of significance on the case. It smacked of an entirely new brand of
espionage. If indeed military secrets had been stolen from U.S. comput-
ers and delivered to the Soviets, as well as hundreds of logins and pass-
words, the case could prove to be explosive.
   Kohlhaas, whose spare office betrayed his avoidance of anything so
technical as even a typewriter, was rather intimidated by the technology
and technical matters surrounding the case. He joked with others that
spelling computer was challenging enough for him. Suddenly he was faced
with the task of prosecuting a ring of young men whose esoteric en-
deavors-breaking into computers thousands of miles away and stealing
information-he had trouble comprehending. But he decided to ignore,
for the most part, his technical limitations, and focus on the espionage
involved. After all, whether they had downloaded information destined
for the KGB from computers in the Far East or photocopied it was
immaterial. Espionage was espionage.
    The Bundeskriminalamt and the secret service had distinct interests
in the case. The last time the secret service and the police had been one
and the same was during the Third Reich. To prevent even the possibil-
ity of another Gestapo, the two agencies had been completely separated.
Not only is communication between the two entities minimal, but they
tend to compete with one another. To the secret service, this was an
interesting and novel case that it would have liked to keep for itself. The
attorney general, who worked with the Bundeskriminalamt, simply
wanted to see justice prevail. Unlike the secret service, the prosecutors
were not about to promise anything like amnesty to an informant.
   It was prosecutor Kohlhaas's assignment to build a case against the
three suspects based primarily upon the claims of two somewhat unreli-
able informants. In fact, the credibility of the two young men was a
matter of some concern. The one with the nickname of Hagbard was a
drug addict from Hannover who had been in and out of psychiatric
hospitals and drug rehabilitation centers. During his full-day interroga-
tions, Hagbard could remain alert for half an hour at most before his
head began to fall toward his chest. His hands trembled and occasionally
he excused himself to take pills of some kind. Only the medication,
coupled with copious quantities of sweets and coffee, helped keep him
fully conscious. Moreover, he seemed to have trouble remembering dates
and details of any kind. Eventually, if he was given long enough to
concentrate, he could reconstruct an important event, or remember the
name of a computer he had broken into. At the bottom of the transcript
of one interrogation, the young man had signed a disclaimer stating that
the medication he was on made it difficult for him to articulate his
thoughts clearly.
   Hagbard's friend Pengo, who had followed closely on Hagbard's heels
to the West German authorities, had his own idiosyncracies. He was
clearly very bright, more direct and lucid in his police interrogations,
and certainly more in command of his faculties, but his manner was
unspeakably arrogant. Here was a young computer hacker, under inves-
tigation for a crime against the state, who displayed no regret whatever
for his actions. When Pengo was summoned to Karlsruhe to restate his
police testimony before a court magistrate, the judge asked the young
hacker who would pay for the thousands of marks' worth of connection
time that had accrued from all the stolen NUIs. "I don't know," came
Pengo's response. "It's not my problem." Kohlhaas suspected that it
wasn't so much a moral imperative, or even a vague sense that he had
done something wrong, that prompted Pengo to unburden himself to the
West German authorities, but a fear that his friend Hagbard had already
done so. More irksome still than Pengo's hubris was his curious lapse of
memory when it came to the magnetic tape. The prosecutor simply
couldn't believe that someone who fancied himself a superior hacker and
who wanted to make a favorable first impression on a KGB agent would
hand over a magnetic tape filled with data without knowing what the
220   A   QygERPUNK
data were. Kohlhaas had little doubt that Pengo was protecting someone,
most likely the Hamburg hacker who called himself Obelix.
   And there were places where Hagbard's version of events differed
markedly from Pengo's. Hagbard insisted that it was not just lists of
computers that went to Sergei, but hundreds of computer logins and
passwords to highly sensitive military computers as well. Pengo admitted
that he had often given logins to Hagbard in the past, but he doubted
that the information he gave to Hagbard eventually made its way to
Sergei. And while Pengo insisted that his share of the money from Sergei
had come to no more than 5,000 marks-or $2,500-at the very most,
Hagbard claimed that Pengo had got at least three times that.
   In order to bring some outside proof to bear on the claims of the two
young informants, Kohlhaas decided to initiate some surveillance on
Markus Hess and Peter Carl in Hannover, and Dob Brzezinski in Berlin.
He also ordered telephone taps, but very little came from them.
   From the start, perhaps to keep Pengo guessing, the West German
officials wouldn't let him know what lay in store for him. Pengo's lawyer
kept up a steady correspondence with the authorities, urging them to
view his client's confession as a demonstration that cooperation between
hackers and the state could yield positive results, in this case the con-
tainment of a new form of espionage.
   The three journalists, in the meantime, were preparing their televi-
sion broadcast on the hacker-spy case. They crisscrossed West Germany,
interviewing everyone from the head of the secret service to members of
the Chaos Computer Club. They dispatched a correspondent in the
United States to interview Stoll at his Berkeley laboratory. They took
their cameras to East Berlin and produced footage of the building that
housed Sergei's office. And they even asked Pengo to sit for an interview.
The authorities had warned Pengo time and again not to talk with
television journalists. For once, common sense prevailed. Delighted
though he usually was to talk, this time he turned them down.
   By the end of 1988, Peter Carl sensed that he was being watched.
During a visit to Sergei's just after the first of the year, Carl passed on
detailed reports of the latest security incident that had swept through
the U.S. It was the case of a computer virus released by Robert Tappan
Morris, a graduate student at Cornell University. Morris had unleashed
his program one evening in November of 1988. Within hours it had
crashed computers at universities and research institutions throughout
the U.S., alerting the nation to the vulnerability of its computer net-
works. The West German hackers assumed that anything that had been
so quick and so devastating would be of considerable interest to the
Soviets. In addition to their virus report, Carl gave Sergei a virus report
written by Cliff Stoll.
   At the end of their usual repast, the KGB officer told Carl that a
pause might be in order. For one thing, Sergei said, Mikhail Gorbachev
was corning soon to visit East Berlin. With so many players involved,
the security of the operation was already low. The last thing the Russians
wanted was to have an espionage operation exposed while the Soviet
leader was in town. Sergei's second reason for halting their meetings was
that he was fairly certain that Western intelligence had its eyes on Carl's
movements. But he assured Carl that the operation was just being placed
in temporary abeyance.
   When Carl reported back to Hess that Sergei had called for a hold on
the action, Hess was a little bit relieved. The whole affair had long since
lost its attraction for him and the best he could hope for now was to see
the entire thing disappear. There had already been a couple of close
calls. First there was the initial scare in the summer of 1987, when
Bremen police had traced calls to his telephone. Then there was the
rush of publicity in April 1988, when the Quick story appeared, describ-
ing Cliff Stoll's hunt for the hacker inside the Berkeley laboratory. But
since then, everything had quieted down. Hess had a new job as a
programmer at a publishing firm in Hannover and it was going well. It
just seemed that the money he had made from this entire KGB business
-about $9,OOO-was too little to risk continuing.
                                  ... l' ...

December carne, and still no arrests had been made. For nearly six
months, Pengo had been living in a state of official limbo. He had few
people to talk with about his predicament, so for the most part he
learned to live alone with the secret. When it carne time to go to the
annual Chaos meeting in Hamburg just after Christmas, Pengo wel-
comed the chance to get out of West Berlin.
   The turnout at the 1988 conference was smaller than it had been in
previous years. Nervous parents, by now well aware of the hacker cul-
ture, kept their teenagers away because of the club's increasingly sullied
reputation. Breaking into computers was now against the law, and that
was deterrent enough for some would-be hackers. What was more, the
Bundespost's new security measures on the Datex-P network had made
casual hacking much more difficult. Many of those who still hacked,
working on the principle that solitary work was safer, kept their distance
222   ~   CYGERPUNK
from the Chaos crowd. And for those who did attend the meeting, much
of the time was taken up with ruminating over politics: a network for
exchanging information on environmental protection was planned and
the groundwork was laid for a push for a law that would mirror the U. S.
Freedom of Information Act. To the hard-core hackers in the group,
such talk just seemed to prove that Chaos was losing its way.
   But for Pengo, going to the conference was a way to pretend that
nothing had happened. In some ways, it seemed like old times. Obelix
picked him up at the airport in his mother's Mercedes, and once he got
to the conference, Pengo immediately felt better. In spite of the laws
against hacking, nearly two years old now, those who counted them-
selves among the hard core couldn't help but stay up through the night,
trying to get into what computers they could.
    Pengo was now well known to the crowd. Those who didn't know
him certainly knew of him. At twenty, he was already a veteran. Some
self-styled hackers had little skill beyond collecting passwords from oth-
ers. Pengo was famous as an elite hacker, one who could also program.
He was famous for being from Berlin. He was famous for having stayed
out of trouble. And the fact that when he did get into trouble he
outsmarted the police by hiding his hard disk was the stuff of Chaos Club
lore. The younger hackers were in awe of him-after all, he had the
reputation as one of the first to crack Philips and Thomson-Brandt com-
puters in France.
   Throughout the three-day meeting, Pengo was followed around by
reporters and television cameras. In fluent English, a language whose
American vernacular he picked up on the computer networks and
electronic bulletin boards he frequented, Pengo chatted amiably and
self-confidently into the viewfinder of a BBC TV journalist doing a
documentary on European hackers. When he talked about the Chaos
VAXbusters, he said, "We thought we were the best. I don't know if we
were." Pengo was a particular favorite among the West German journal-
ists; he was always willing to sit down for a long interview with a local
newspaper reporter, chain-smoking his hand-rolled cigarettes as he de-
scribed some of his best hacks.
    Pengo dipped in and out of the individual conference sessions, leaning
casually against doorframes, his stance that of someone accustomed to
situating himself next to the closest exit. At a session on the future of
Chaos, with a note of irritation in his voice, Pengo spoke up. He told
those in the room that he was unhappy with where he saw Chaos going.
To a strictly technical hacker like Pengo, Chaos's political direction was
    unacceptable. Concentrating on things like environmental protection,
    he told them, was diverting the group from its technical origins. It was
    little wonder, he said, that the truly talented hackers were beginning to
    abandon the club.
        It had nothing to do with the espionage investigations that were under
    way, but rumors were beginning to circulate among the Hamburg crowd
    that someone was betraying them, working for the West German gov-
    ernment. Wau Holland suspected that there was a plant somewhere
    inside the Chaos Club's ranks. But no one had a specific reason to
    suspect anyone of espionage. And Pengo was relieved to see that Wau's
    suspicion was directed at others. Hagbard wasn't there this year and no
    one appeared to miss him.

    Peter Carl had been thinking for some time of moving to Spain and
    starting a computer company. There was nothing tying him to Han-
    nover; he had been divorced for more than a year and his wife had
    custody of their ten-year-old daughter. Since leaving his croupier's job
    in 1986, he had been collecting unemployment. To augment the 880
    marks he got each month from the state, Carl occasionally ferried cars
    to Spain for a local used-car dealer. And then there was the money from
    Sergei. But that business relationship had gradually diminished, and
    there was no telling if, now that Sergei was calling for a break, it would
    ever start again. Certainly it was nothing to count on for a steady in-
       So in early 1989 Peter Carl began to give serious thought to moving
    to Madrid. He already spoke reasonably fluent Spanish, and Spain had
    always had a certain pull for him. So he traveled to Madrid, opened a
    bank account and looked into renting an apartment. In February 1989,
    Carl went to Berlin to discuss his prospective new business with Dob and
    Hess. He told Hess that he had money to invest in the new company
    and he wanted to discuss technical details with his two friends. Hess had
    come along to Berlin and was happy to offer advice, but he had no
    intention of uprooting himself from Hannover. He was happily en-
    sconced in his new job, making a comfortable salary of 2,800 marks-or
    $1,400-a month. He was glad to serve as adviser to the other two and
    nothing more. Hess's days of risk-taking were behind him. For Peter
    Carl, one of the major attractions of a fresh start in another country was
    the idea of putting the spying affair behind him once and for all. It was
    certainly one way to ensure that he would never be caught. On the other
224   £    CYEERPUNK
hand, as time passed it looked more and more as if they were going to
get away with the whole thing anyway.

Dob was asleep on the floor mattress in his tiny one-room apartment in
Berlin when he was awakened by a loud snap, the sound of someone
breaking down his door. He sprang up in his bed to see four pistols
directed at his head. Behind each gun was an officer in civilian clothing.
"What is going on!" Dob demanded.
   "Surrender your weapon!" they ordered back, although there was no
gun in sight. He was arrested at once.
   Markus Hess had just settled into a comfortable exercise regimen.
Five mornings a week, he arose at 6:45 to go for a swim before work. On
March 2, 1989, he was just returning, refreshed and cheerful, his hand
in his pocket reaching for his keys to open the front door to his apart-
ment, when he heard a quiet, polite voice at his back, the tone of
someone who might be asking for directions. "Herr Hess?" He turned
around to see eight well-dressed men staring at him. He knew immedi-
ately what was happening, and although he wasn't surprised when they
informed him at once that he was under suspicion of espionage, Hess
began to pray.
   Upstairs in Hess's apartment were more officers. Hess insisted on
calling his lawyer's office. When police had escorted Hess to his apart-
ment in June 1987, his lawyer had been on the scene within a few
minutes, and the lawyer's acumen had emboldened Hess to treat the
situation with a cavalier attitude. But this time, the police were frighten-
ingly well organized. When Hess called his lawyer's office, there was no
answer. On his second attempt he reached a secretary, who told him his
lawyer would be out for most of the day. Hess had no choice but to let
the search begin. After a thorough scouring of Hess's apartment, the
officers took him for an interrogation at police headquarters in Han-
nover, where the authorities had rented out an entire floor of the build-
ing for the handling of the hacker-spy case. It was Hess's first
introduction to Ekkehard Kohlhaas, the prosecutor in charge of the case.
   Markus Hess's first impulse was to admit to hacking, and in particular
to breaking into the computers at the Berkeley laboratory. He admitted
that he knew Hagbard and the others and that he had given Hagbard
three or four logins for the purpose of hacking. He even admitted to
providing software to Dob and receiving money for it from Carl, and to
giving software later to Carl directly, then getting money for it. But he
denied any espionage activity and any knowledge of it.
   "I've never consciously worked for a hostile secret service," he de-
clared in a carefully crafted phrase. "Carl never said for what or for
whom he needed the software and I never asked."
   "Are you familiar with someone by the name of Sergei?"
   "That name is unknown to me," Hess lied.
   "Sergei is believed to be from the trade mission in East Berlin. Sergei
is believed to be the one who ordered this software."
   "I have absolutely no knowledge of this," Hess insisted.
   But as the questioning wore on, it became clear to Hess that someone
in the group had been talking. The next question unhinged Hess com-
pletely. "According to information we have from the federal police,
Sergei was mentioned openly in conversations among Brzezinski [Dob],
Koch [Hagbard], Carl, Hubner [Pengo] and you, wherein the logins you
procured were meant to go to Sergei. What do you say to that?"
   There was no point in continuing to deny things. Everyone's name
was out. "That's right," Hess replied. "Those people did speak of Sergei.
And it was clear to me that Sergei was in the East and he must belong
to the Russian secret service."
    Hess went on to explain that after he had heard the others talk of log-
ins going to Sergei, he immediately went back to LBL and altered the
passwords into the Berkeley computer.
   When his interrogators read him several lists-supplied by Hagbard
-of dozens of systems for which Hess had allegedly supplied logins,
including a computer at the Anniston Army Depot, the Optirnis data
base at the Pentagon and several computers of the Environmental Pro-
tection Agency, Hess denied it. In fact, from what Hess could tell,
Hagbard had been feeding the authorities a damning array of confused
stories. He was telling of logins to computers that dated back to late
1985, when Hess and Hagbard had hacked together. But these logins,
Hess tried to explain, had nothing to do with Sergei.
   The story that Hess told in those first few hours was the one he would
stick with for the next year, even as the prosecutors built their own
compelling case against him. By the time the ten-hour interrogation
session had ended, Hess had confessed to delivering at least half a dozen
programs to Carl with the full knowledge that the software was headed
for the East. Mostly, he said, it was software he more or less chanced
upon at Focus. But he insisted that he had never given computer logins
226 •     CygERPUNK
and passwords to Hagbard to give to Peter Carl, and that not once had
he given information from computers to Carl to sell to Sergei. Hagbard,
he claimed, had psychological problems, and that fact should be taken
into account when weighing the credibility of anything Hagbard said.
   Unlike the search of the Focus offices in 1987, this time Hess sensed
he would remain in custody. He spent the night in jail. The following
morning, he was flown by helicopter to Karlsruhe for a deposition before
a magistrate. It was an uncomfortable and awkward ride, as Hess and
Kohlhaas were forced to sit side by side on a narrow seat. During the
thirty-minute trip, Kohlhaas sat silently and read through his papers.
The two men didn't exchange a word.
   Because Markus Hess, a young man with a clean record and stable
job, presented no flight risk, he was freed later that afternoon. His first
call was to his father, who drove to Karlsruhe to pick him up. Hess spent
the weekend with his parents, an hour's drive from Karlsruhe in Fulda.
He had a lot of explaining to do, mostly as to why an otherwise perfectly
well-adjusted son, the very picture of decency and conservative values,
had engaged in such anomalous and unseemly behavior, placing his
bright future in jeopardy, and all for the sake of 15,000 marks. Hess had
always had a close and warm relationship with his parents, but he had
never discussed his computer break-ins with them, not only because they
were illegal, but also because his parents simply would not be able to
understand his general obsession with computers. In the end, his parents
made it clear that they would stand by him.
   Peter Carl had been arrested the previous morning as he was pulling
out of a parking space near his Hannover apartment. He was just starting
out on one of his car deliveries to Spain and there was no telling if this
was to be his permanent move there, so the police acted swiftly. Two
unmarked cars blocked Carl as he began to drive away. One officer leapt
out of his car and into Carl's. The arrest took approximately thirty
   For both Dob and Carl it became apparent after an hour or so of being
questioned that Pengo and Hagbard had gone to the authorities. Even-
tually, both of them confessed to espionage. But they weren't to be
accorded the same leniency that Markus Hess got. Dob and Peter Carl
had previous arrests on their records, and both were considered flight
risks-Carl for his plans to go to Spain and Dob for his avoidance of
military duty. Both were taken into custody. Carl's ex-wife came forward
to say she would take out a loan for 1,000 marks for bail, but prosecutor
Kohlhaas felt uneasy about setting Carl free and urged the judge to deny
bail. Kohlhaas saw his case strengthen when, during the search of Carl's
apartment, a Casio pocket calculator was found. It contained the tele-
phone number for one Sergei Markov.

It wasn't just three isolated arrests that took place that day. The Bundes-
kriminalamt was a model of preparedness. That morning, units of the
state security branch of the Bundeskriminalamt were dispatched
throughout cities in the Federal Republic. A total of fourteen house
searches were conducted. Hackers and friends of hackers, some from the
old Hamburg crowd and some from the Hannover hacking group, were
rousted from their beds and taken to police headquarters in their respec-
tive cities and questioned about their knowledge of the espionage ring.
It didn't take long for the news to circulate that everything had been set
in motion. That morning, Obelix, who had also been taken in for
questioning in Hamburg, had called the journalists to say that arrests
were under way. The journalists in turn started the process of substituting
the special half-hour "Panorama" segment they had prepared weeks in
advance for the evening's regular programming. After the regular eve-
ning news, the North German television station broadcast a teaser for
the evening program, hailing the upcoming show as a scoop not to be
   When the Panorama show aired at 11:00 P. M., three million West
German television viewers were tuned in. The show opened with a
dramatic shot of the Glienicker Bridge in Berlin, a classic point for
exchanging spies from both East and West. It framed the hackers' images
in silhouette and, switching to a world map thick with arrows traveling
from Silicon Valley to Moscow, described this as the first major case in
a new era of high-tech espionage. Cliff Stoll recounted the troubling
keywords the LBL hacker had sought. He stared into the computer,
concern etched on his face. "Somebody was inside my computer looking
for Star Wars information."

Pengo had suspected something might be afoot. Earlier in the week, his
secret service contact had planned to fly from Cologne to Berlin to meet
with him, but then had canceled the appointment at the last minute.
Pengo had moved out of his father's apartment a few months earlier
without filing the necessary registration papers with the local police to
account for his whereabouts, a bureaucratic procedure required of every
    228 •     Cyg£RPUN~

    citizen. Sensing that arrests might be imminent, and worried the police
    might show up at Gottfried's apartment first, he warned his father that
    there might soon be a reprise of the December 1986 bust. Pengo's parents
    were only vaguely aware of the trouble their son had fallen into-and
    was attempting to escape from. They knew that he had an investigation
    pending against him for something rather serious, and related to his
I   hacking, but they both shied away from asking specific questions. Renate
    had long since given up trying to understand what her son was doing
    with computers and had no way of gauging what the problem really was.
    Gottfried was distressed to hear that the police might be back.
       "What are they going to do and when are they going to show up?" he
    quizzed his son.
       "I don't know," Pengo replied with the fresh impatience that only an
    exasperated son can express to his father. "Excuse me, but I don't know.
    I don't know how many will come or when they'll come or whether
    they'll come and order a strip search or just say, 'Have a nice day.' " If
    nothing else, the conversation prompted Pengo to file his change of
    address with the police the following morning.
       Three days later, at about 9:00 in the morning, he was awakened by
    the sound of his apartment buzzer. It was the West Berlin police, who
    were there to conduct a search and question him. They confiscated
    miscellaneous pieces of paper, some containing computer printouts, oth-
    ers with Hagbard's handwriting, but nothing terribly incriminating. It
    was a pro forma exercise, of course, and it was clear from the moment
    they began that Pengo would be spared a real arrest. The police also
    conducted a search at the offices of Pengo's company, which had grown
    to five people. Pengo was surprised by the extent of the police action.
    Still, he was tremendously relieved to have the whole episode burst open
    at last. As far as he could tell, his name hadn't surfaced. And the
    "Panorama" program had shielded him too, referring to him as Frieder
    Sell, a Berlin student.
        During the broadcast, Joachim Wagner, the head of the "Panorama"
    program and the show's anchor, had described the discovery of the
    hacker spies as the biggest case since 1974, when Gunter Guillaume,
    then Chancellor Willy Brandt's friend and aide, was found to be a
    captain of the East German secret service. The Guillaume scandal had
    toppled a chancellory. Gerhard Boeden, the head of the Verfassungs-
    schutz itself, declared: "What we have here is a new form of hostile
    infiltration of our data networks." And for the work his agency did in
    uncovering the operation and controlling the damage, a spokesman for
the secretary of the interior called the arrests a "severe blow to the
   The "Panorama" program set off a chain reaction of media reports.
Espionage was something the German people didn't take lightly. And
West Germans were particularly appalled by the reports of a new and
insidious form of espionage, of teenage malcontents who had stumbled
upon computer vulnerabilities and not just exploited the weaknesses they
found but used them to endanger the military security of the West.
Various officials and computer security experts stressed repeatedly
that some of NATO's most sensitive military data resided on the
computers allegedly penetrated by Hess and the others. Even if the
data that had been taken from the computers were not classified,
or even sensitive, security experts in West Germany and the U.S.
argued that classified information could be inferred from unclassified
   But it didn't take long for the ministry of the interior to retreat from
its original pronouncement that the KGB had been delivered a severe
blow. Once the initial sensation had died down, skeptical journalists
began to question just how much damage to national security the group
of hacker spies had done. Had highly classified material changed hands?
Or was it merely some harmless public-domain software? Moreover, why
the public outrage when the real menace of computer espionage is most
likely not from the Soviet Union but in sophisticated schemes perpe-
trated by both foreign and American corporate spies secretly tapping
into their competitors' systems in search of trade secrets and market

Pengo didn't enjoy his anonymity for very long. Within a few days,
rumors about who was involved in the espionage ring began to circulate
furiously through the press and in various hacking circles.
   Wau Holland, Chaos's founder and keeper of the ethical flame, was
beside himself with anger. If it was true that Pengo was involved, it was
an unforgivable betrayal of hacking in general and Chaos in particular.
Wau's first impulse was not to believe such a thing could have happened,
and certainly not that Pengo could have done it. He had always con-
sidered Pengo somewhat naive, but certainly not capable of doing some-
thing so unconscionable. It wasn't so much the selling of software to the
Soviets that bothered him, but that the hackers might have been selling
software that came from Chaos people. Worse, they had no doubt been
230   •   CygERPUNK
informing on Chaos and its members to West German intelligence agen-
cies. In Wau's estimation, both of those transgressions were far worse
than what they were being charged with. To see for himself, he called
Pengo in Berlin.
    "Just say yes or no," Wau said when Pengo answered his telephone.
"Is it safe to talk?"
    "Is it true what is being said about you?" was Wau's next question.
    "That's all I wanted to know," and with that Wau ended the conver-
    It took just a few days for true names to start surfacing to the general
public. A few days after the "Panorama" show, Der Spiegel, the Federal
Republic's largest serious news magazine, identified Hess and Hagbard by
their real names. In a second article in the same magazine the following
week, everyone's name came out. Then, more gradually, the fact of
Pengo's and Hagbard's cooperation with the authorities inched its way
out as welL
    Pengo was immediately on the defensive. In response to the uproar
over his spying, his betrayal of the others and his work with the author-
ities, he submitted a lengthy posting in English to the Risks forum, an
international computer network discussion group about potential hazards
of computerized technology. Risks is read by computer scientists and
other technical experts throughout the world. Peter Neumann, an
American computer scientist and the editor of Risks, was so surprised to
see the blunt note in English coming over the forum's electronic transom
that he decided to publish it:

Date: Fri, 10 Mar 89 18:09:25 MET DST
From: Hans Huebner <pengo@netcs.SMTP>
Subject: Re: News from the KGB/Wily Hackers

I have been an active member of the net
community for about two years now, and I
want to explicitely express that my network
activities have in no way been connected to
any contacts to secret services, be it
western or eastern ones. On the other hand,
it is a fact that when I was younger (1 ' m 20
years now), there has been a circle of
persons which tried to make deals with an
eastern secret service. I have been involved
in this, but I hope that I did the right
thing by giving the german authorities
detailed information about my involvement in
the case in summer '88.

For my person: I define myself as a hacker.
I acquired most of my knowledge by playing
around with computers and operating systems,
and yes, many of these systems were private
property of organisations that didn't even
have the slightest idea that I was using
their machines. I think, hackers--persons
who creatively handle technology and not
just see computing as their job--do a
service for the computing community in
general. It has been pointed out by other
people that most of the 'interesting' modern
computer concepts have been developed or
outlined by people which define themselves
as 'hackers'.

When I started hacking foreign systems,
was 16 years old. I was just interested in
computers, not in the data which has been
kept on their disks. As I was going to
school at that time, I didn't even have the
money to buy an own computer. I enjoyed the
lax security of the systems I had access to
by using X.25 networks.

You might point out that I should have been
patient and wait until I could go to the
university and use their machines. Some of
you might un~erstand that waiting was just
not the thing I was keen on in those days.
Computing had become an addiction for me,
and thus I kept hacking. I hope this clears
the question 'why'. It was definitely NOT to
get the russians any advantage over the USA,
232   A    Cyg£RPUNK
nor to become rich and get a flight                          to the
Bahamas as soon as possible.

For punishment: I already lost my current
job, since through the publications of my
name in the SPIEGEL magazine and in RISKS,
our business partners are getting anxious
about me being involved in this case.
Several projects I was about to realise in
the near future have been cancelled, which
forces me to start again at the beginning in
some way.
                               --Hans Huebner

   It was a fairly bold statement to send out to nearly thirty thousand
computer scientists, students and researchers-most of them American
-who perused the Risks forum every day. Perhaps Pengo underestimated
the number of people who would see his posting. They could hardly be
expected to muster much sympathy for the self-confessed spy. Those
Risks regulars who were willing to give Pengo the benefit of the doubt
were impressed by his honesty. But the general reaction was one of
horror. At best, this was someone who didn't understand what he was
getting into, who was naive to the point of stupidity. If someone as
unscrupulous and self-serving as this was frequenting the networks, what
other dangers might lurk there?
   Wau, too, wasn't to be swayed very easily by Pengo's defense of his
actions. He saw it as a self-justification that was curiously lacking in any
kind of moral reflection or thought as to how Pengo's cloak-and-dagger
adventure might affect others. Wau wanted to divorce himself and his
organization entirely from this young troublemaker. He instructed people
from the West German hacking scene to hang up the phone if Pengo
should call. A joke began to circulate that when Hans chose the sobri-
quet Pengo for himself, it was to prove more fitting than he knew: Pengo,
that is, suggested not just the heroic penguin of the video game, but one
who sprang from ice block to ice block, saving himself just as each was
about to sink.
   Pengo's parents, at least, were understanding. Renate, in particular,
was quick to excuse her elder son's escapades as an understandable search
for adventure. Moreover, both parents suspected that just as their son
had always managed as a small child to pull his head out of the noose in
the nick of time, somehow he would get out of this tight spot as well.
But Renate's septuagenarian mother, whose early life had consisted of a
series of unpleasant encounters with East Germany's Communist regime,
was horrified that her grandson could be doing business with such people.
   Hagbard came under some sharp criticism as well. But rather than go
public with his own defense, he kept to himself and quietly entertained
journalists' overtures. His price for an exclusive interview: 30,000 marks,
or $15,000.
   It looked as if Hagbard was beginning to straighten out his life. Before
the news of the KGB activity broke, a friend had helped him get a job
as a messenger in the Hannover office of the conservative Christian
Democratic Union. It was menial, low-paying work, but by all accounts
he was reliable and well liked by others in the office. Even after his
spying came to light, along with his dependence on drugs, the CDU
office kept him on, convinced that he deserved another chance. Some
of Hagbard's friends viewed the CDU job as further proof that this
erstwhile social democrat's political perspective had gone completely
awry. Others saw the job as Hagbard's first small step toward folding
himself back into society. His life, at least to outsiders, seemed more
stable. After years of rootlessness, he was finally planning to move into
an apartment of his own. And a recent embrace of conventional religion
had probably added to his calm bearing. His fellow workers would hardly
have supposed that this quiet young man had been in and out of psychi-
atric hospitals, battling his drug addiction.
   In reality, Hagbard's life was still a mess. Contrary to appearances, he
was still struggling to free himself from drugs. His rent in the new apart-
ment was being paid by the West German authorities, engendering a
strange dependence. And being in service to the West German authori-
ties was probably more than someone so invested in paranoid fantasies
could withstand. Hagbard was increasingly plagued by the notion that
he was under constant surveillance by one or another real or imagined
organization. And he was convinced that the officials were reading his
mind and using his thoughts to their own end. Perhaps just as difficult
for Hagbard was the realization that his commercial value was a direct
function of his newsworthiness, which was beginning to wane. Within a
month or two of the arrests of the others and the disclosure of the full
extent of Hagbard's and Pengo's involvement, the story had lost its sheen
and journalists had all but forgotten the young hacker with the unusual
   On the morning of May 23, Hagbard set off in the COO's Volkswagen
234   •   CygERPUNK
station wagon to make a delivery to a government office in Hannover,
and he didn't return. After lunch, friends went to look for him, and by
4:00 that afternoon the Verfassungsschutz had dispatched a search party
as well. After a week, the friends abandoned their search.
   Hagbard was missing for nine days.

The first time farmer Ernst Borsum saw the VW Passat parked in an
isolated patch of woods outside Ohof, a small village of six hundred
people north of Hannover, he thought it belonged to a jogger. But for a
few days in a row, each time the farmer went out to irrigate his fields he
saw that the car was still there, leaves gathering on its hood. He called
the police. Not far from the car they discovered a fully charred corpse.
N ext to the body were the remains of a gasoline can. All vegetation in
the surrounding three or four meters was black. The driver, the police
concluded, had taken the gasoline can, poured its contents over himself,
soaked the surrounding earth as well and lit a match. He was consumed
in flames immediately. Even if he had screamed out in pain, no one
would have heard him. When he was found, he was lying facedown, one
arm under his stomach, another over his head. That was a sign that he
might have changed his mind at the last minute and dropped into a roll
to try to put out the flames. Or, if someone had done this to him, he
might have been trying to save himself. He was wearing no shoes.
   It was, of course, Hagbard.
   The news of Hagbard's death brought the case of the hacker spies
once again into public view. West German magazines and newspapers
carried lengthy, speculative articles about the death. Just how sensitive
was the material that landed in the hands of the KGB? And just who
might have had an interest in seeing Hagbard dead? Could such a hid-
eously painful death really have been the result of suicide? Unlike a drug
overdose or suicide by gunshot, death by burning usually leaves the cause
more ambiguous. One complicated theory began to make the rounds
that Hagbard was murdered by left-wing terrorists who had commis-
sioned him to hack his way into state computers to retrieve information
and wanted to kill him before he could talk.
   Hagbard's attorney, Johann Schwenn, regretted that he hadn't done
more to help someone whose cries for help he had somehow failed to
recognize. Schwenn didn't subscribe to the murder theories. But he was
quick to lash out at the authorities, whose only concern was to build
their case, not to care for the troubled young man. Others began to
 conclude that Hagbard had been murdered, or at least pushed to his
 suicide by pressure from journalists and from the West German author-
 ities, all of whom wanted something from him and pursued it with-
 out regard for his condition. If Hagbard had been driven to suicide by
 his own paranoia, speculated one friend, he might have chosen self-
 immolation because he believed it would prevent his thoughts from
 being misused further. The fact that he died on the twenty-third day of
 the month was also no coincidence: In The Illuminatus! Trilogy, twenty-
 three is a number of high significance. "All the great anarchists died on
 the 23rd day of some month or other," explains one of the Illuminatus!
 characters to another.

 A group of Hagbard's friends placed a notice in the Tageszeitung.

                        We are angry and sad about the death
                    of our friend.
                        We are certain that he would still be
                    alive if the police and media hadn't,
                    through their criminalization and un-
                    scrupulous sensationalism, driven him
                    to his death.

     Pengo was shaken by the news of Hagbard's death. It forced him to
  confront a horrifying reality. More than his first confession to the jour-
  nalists in the noisy Berlin cafe, more than the interrogations by the
  West German authorities, and more than the bursting open of the story
  with the arrests three months earlier, Hagbard's death forced him to
  acknowledge the serious consequences of the case. Pengo knew, of
. course, that Hagbard was so unstable that his involvement in espionage
  probably had little to do with his suicide. Nonetheless, the pretense that
  none of this actually hurt anyone was now so obviously false. Moreover,
  Pengo sensed that this could put more pressure on him as the sole witness
  in the case. Turning state's evidence in July had not yet given him full
  protection because his case in the espionage incident had been separated
  from the others'. It could still result in a trial against him.
     Kohlhaas, the prosecutor in Karlsruhe, had different cause for concern
  about what effect Hagbard's death might have on the case. Hagbard's
  statements had been sufficiently tenuous, with enough assertions contra-
  dicting the others, to make Kohlhaas wonder if he could use Hagbard's
  testimony at all. Kohlhaas had little choice but to base his indictment
236   A   CYEfRPUNK
on what Pengo and Hagbard-and the three defendants themselves-
said. The gathering of matching testimony was of utmost significance.
    The prosecution had to establish the connection between Markus
Hess and the laboratory in Berkeley. Kohlhaas's most promising witness
was Cliff Stoll. In fact, if it hadn't been for the efforts of the woolly
scientist, there would have been just the barest of threads holding the
case together. Sergei's telephone number had been found in encoded
form inside Carl's small pocket computer, but it wasn't enough on which
to build an espionage case. Stoll's most valuable contribution had been
his girlfriend's invention of the SDInet material, which had not only
cinched the telephone trace but also caused the appearance of the Laszlo
Balogh letter. In fact, Kohlhaas viewed the Balogh letter as the most
solid piece of evidence his office had to establish a link between Hess
and the KGB. Someone in Moscow must have dispatched Balogh to seek
the SDInet information, he reasoned. And since Stoll and the LBL
hacker were the only two who could see the SDInet file, Hess must have
taken the information and passed it to Carl, who in turn gave it to Sergei
Markov. As far as Kohlhaas was concerned, there was simply no other
    Stoll was a witness of special importance because Kohlhaas had no
physical evidence of Hess's hacking. When Hess's apartment was
searched on the day of his arrest, the police found no printouts that
implicated Hess as having been inside American computers in general
and LBL in particular. In fact, the police found none but the most
remote signs that pointed to Hess's hacking. It was the Bundespost's
work in completing the trace to Hess's apartment that proved Hess was
in the LBL computers.
    In June, the West German federal police summoned Cliff Stoll to
Meckenheim for two full days of questioning. The police investigators
were polite and deferential toward their American guest during the
painstaking process of translating Cliffs account first into German that
a judge unschooled in technical terms might understand, then into offi-
cial police German. For the most part, it was a matter of matching Stoll's
sitings of the hacker with the dates and times of the Bundespost's tele-
phone traces. They were aided in their efforts by a meter-high stack of
printouts from the dozens of hacking sessions Stoll had monitored. Stoll
was then taken to Karlsruhe to meet the prosecutor and repeat what he
had said in Meckenheim before a magistrate. Stoll's deposition went
smoothly. Kohlhaas was encouraged when, at one point, Stoll said that
he saw data flowing over his monitor, then two weeks later saw the
hacker break into another computer using complex commands he must
have copied down or stored from the earlier session. That indicated to
Kohlhaas that the hacker had been storing information.
    But there was one crucial point at which the prosecutor misconstrued
Cliff Stoll's words. When Stoll was explaining what he saw happening
on the computer screen at LBL, Kohlhaas concluded that Stoll had a
piece of equipment at hand to prove that the hacker was actually trans-
ferring and saving the information he was seeing. "I saw him printing
out screens full of military data" was one typical assertion of Stoll's,
when what the Berkeley scientist really meant was that he saw screens
filled with such data scroll by on his screen at the laboratory, with no
way of knowing what was happening to the information on the other
end. What Kohlhaas, with his limited computer knowledge, failed to
understand was that it quite possible that the hacker was doing nothing
at all with the data he saw. It was in fact likely that the hacker disre-
garded the bulk of it, that he neither went through the time-consuming
task of transferring it from Stoll's computer to his nor stored it on a
floppy disk. But with Stoll using such words as print and download and
copy interchangeably, Kohlhaas's mistake was understandable.
    The prosecutor concluded that he was pursuing a solid case. From
what he could tell, the Berkeley computer manager had just provided
the prosecution with ample proof not only that Hess was the one hacking
into LBL's computers, but that Hess's intrusions into LBL and other
U.S. computers played an integral role in the dealings with the KGB.
So Kohlhaas decided to call Stoll to the stand as the prosecution's star
    But Stoll was no expert in military matters. Kohlhaas's next step was
to find a witness, preferably from the U.S. military, to corroborate Stoll's
claims that what Hess had taken from LBL and other computers was of
a highly sensitive nature. The names of the places whose computers Hess
had compromised sounded sensitive enough. All Kohlhaas needed was
for someone more qualified than Stoll to peruse the pile of printouts and
offer an assessment. He contacted the Verfassungsschutz and asked that
someone from that agency seek a U. S. official who would be willing to
travel to West Germany and take the stand as an expert.
    It turned out to be a far more difficult request to fill than Kohlhaas
had imagined. Although the Verfassungsschutz official assured Kohlhaas
that he got in touch with the appropriate U.S. officials promptly, to the
prosecutor's bewilderment no one from the U.S. stepped forward to lend
some expertise to the case. The National Security Agency had made its
238   ~   CYG£RPUNK
own assessment of the damage to national security that might have
resulted from the West German hackers. It came in the form of a brief
memo from Robert Morris, an agency scientist and the father of Robert
Tappan Morris, the Cornell student who had released the computer
worm throughout the Internet the previous year. Morris's summary:
"Looks like the Russians got rooked." But even Morris's opinion wasn't
sent to Karlsruhe.
   Still, with the confessions of the three defendants themselves, and
the promise of provocative testimony from Stoll, by late July Kohlhaas
felt confident enough to issue a seventy-three-page written indictment
of Carl, Hess and Dob.
   Officials at the U.S. Justice Department were quietly mulling their
own plans for a possible prosecution. They couldn't charge the hackers
with espionage, since the information that was sold didn't appear to be
classified. Instead, the government wanted to charge the West Germans
with unauthorized access to U.S. computers. Mark Rasch, the same
prosecutor who was working on the Robert Morris case, was considering
ways to lure the West Germans onto American soil, make an arrest and
try them in U.S. courts. With their frequent electronic travels inside
U.S. computer systems, Hess and Pengo were the Justice Department's
main targets, and Pengo was the only one who came close to ensnare-
ment. With plenty of time on his hands while he waited for his case to
be decided, he toyed with the idea of going to the States in the summer
of 1989 to visit some friends he had made over the network, but his
lawyer pointed out the possibility of an arrest in the United States and
he stayed home.

By the time the case came to trial, it was a new era in Europe. For nearly
thirty years, Berlin had been two cities divided by a wall. Two months
before the trial, in one spontaneous burst in November 1989, the border
dissolved overnight. Thousands of jubilant Germans clambered atop the
wall and danced. That night marked the beginning of German reunifi-
cation, a process that was to be surprisingly swift and peaceful. Democ-
racy was breaking out all over Eastern Europe, and the toppling of the
Berlin Wall was the most dramatic event signaling the end of the Cold
War. The long-standing restrictions on the export of technology were
diminishing quickly.
   But the prosecution didn't intend to allow world events to influence
its case against the three hackers. As Kohlhaas saw it, if the defendants
had been working for Stasi, the East German secret service, a trial would
have been of dubious relevance by now. But these young men were
accused of getting their pay from the Soviet KGB, which was still very
much in business. The trial would take place at the beginning of]anuary,
with a score of witnesses.
    Celle, a small, conservative city just north of Hannover and one of
the dozens of picturesque towns that dot the West German landscape, is
the seat of the Higher Regional Court for Lower Saxony. All half-
timbered houses, bakeries and pedestrian malls lined with shops selling
knickknacks, Celle seems an unlikely judicial terminus. The trial took
place inside a courtroom that the Federal Republic spent $8 million to
build for trying terrorists. Despite its inviting blond wood paneling, pea-
green carpeting and comfortable seats, the Celle courtroom is designed
to be highly secure.
    A panel of five judges, all middle-aged men, none possessing so much
as a rudimentary acquaintance with computers, was appointed to sit in
judgment of the three young computer hackers. The presiding judge,
Leopold Spiller, a soft-spoken man in his middle years, had a tenuous
grip at best on the technology that had played such a crucial role in the
case. Sitting on the evidence shelf behind the judges in the space that
usually held more traditional weapons were the spoils of the previous two
years' house searches: Hess's Atari and Apple II computers, Dob's bat-
tered Rainbow, Carl's miniature Casio containing Sergei's phone num-
    On the first day of the trial, January 11, 1990, curious spectators and
at least a dozen reporters sat among stony-faced officials from the police
and the Verfassungsschutz. As the three men stood trial for the same
crime, they were seated together with their lawyers before the judges'
bench. Dob was the first to give his statement. First, the presiding judge
wanted a short lesson in computers. He asked a reluctant Dob to explain
what hacking meant.
    "As I understand it," the judge ventured, "hacking is an attempt to
get into a computer system. How does it work?"
    During the months Dob had spent in jail awaiting trial, a legal form
of suspended animation in Germany known as Untersuchungshaft, or
pretrial confinement, his mood had plunged so markedly that prison
guards began to worry they might have a suicide attempt on their hands.
He had let his trim goatee grow to an unkempt mass of beard, and his
240   A   CygERPUNK
walk had degenerated to a shuffle. The time in jail-much of it spent in
solitary confinement-had only heightened Dob's natural antipathy for
West German authority.
   From Dob's slouched bearing in the courtroom, it was evident that he
didn't feel up to giving a lesson in computers. "Hacking is the modern
art of telephoning," he said.
   The judge was puzzled. "How does it work?"
   "Well, you have to have a telephone line in any case," he answered.
   The judge wouldn't be deterred from his attempt to understand this
strange new world. "What if I want to get into the University of Han-
nover and read a book in the library?"
   "I was never interested in the contents," Dob answered. "Just in the
computer itself."
   And so it went, each of Dob's answers more listless than the one
before. The other four judges grew restless. One played with his watch,
another began shuffling papers. A third began to leaf through a copy of
Hacker fur Moskau, a colorful, breathless account of the espionage story
written by journalist Ammann and his colleagues and published shortly
after the March arrests. The book was to become a major reference for
the confused judges as they struggled to understand some of the com-
puterese that dominated the triaL
    Dob's truculence came as a bit of a surprise. Despite his low mood,
Dob had offered more cooperation to the prosecutors than either of his
fellow defendants at the prison. He had served as a technical consultant
and helped the investigators restore the Rainbow computer Pengo had
damaged and retrieve the Digital Securepack security program from the
hard disk.
    "Okay," said the judge. "Let's say we're in a hacker's apartment. What
happens now? How long does one of these sessions last?"
   No answer.
    Despite his apparent indifference, Dob didn't waste any time before
trotting out what was to become known as the "world peace" defense.
The group's motive, he said, was to equal the balance between the world
powers. Hence the code name Equalizer for the operation. They adopted
as their role model Manfred von Ardenne, a leading German physicist
who immigrated to the Soviet Union after the Second World War and
worked on nuclear fusion. In his testimony, Dob perpetuated a popular
myth that Ardenne was an atomic spy who wanted to bring world peace
through espionage. With technology transfer, the hackers wanted to do
something similar, Dob explained. Better that the Soviet Union should
have more reliable Western software controlling its missiles than Soviet-
made software that could inadvertently trigger a nuclear war. What he
and his friends were doing was one more step toward world peace.
   It was an inventive line of reasoning, to be sure, a twist on the code
name Equalizer.
   When his turn came, Peter Carl continued with the world peace
defense, but it got a cold reception from the judge. "World peace?" the
judge asked, looking a bit annoyed now. "Nowhere in your one-hundred-
ninety-three-page testimony to the police did you mention anything
about world peace. "
   The judge instructed Carl to describe to the courtroom his trip in the
late summer of 1986 to the trade mission in East Berlin, his first conver-
sation with Sergei and his subsequent role as courier.
   "So you were responsible for marketing?" asked prosecutor Kohlhaas.
   "Did you ask for someone from the KGB specifically?" asked the judge.
   "No," Carl answered. "It was all the same for me. KGB. Trade mis-
sion. Ambassador. It was understood, but not said."
   If nothing else, the defendants' ostensible concern for the world's
military stability proved a successful exercise in manipulating newspaper
headlines: "Hackers Wanted to Assure World Peace" was the headline
in the Berlin T ageszeitung the following morning.
   On the trial's second day, it was Markus Hess's turn to tell his story.
In contrast to Dob's phlegmatic demeanor, and Carl's occasional coarse-
ness, Hess presented himself as a bright, articulate, wholesome young
man. Whereas each time Carl was asked a question he leaned toward his
lawyer as a shy toddler might retreat from a roomful of strangers into his
mother's skirt, Hess took command of his own defense. His lawyer was
there, it seemed, as a mere formality.
   What Hess's testimony did more than anything was remind everyone
that this was an espionage trial. Nobody had been accused of illegal
entry into computer systems. No one, that is, was on trial as a hacker.
In fact, Hess seemed only too willing to describe his exploits inside the
LBL computer. Hagbard had first introduced him to hacking, and in
August of 1986, when Hess discovered LBL, which he described as being
"open like a barn door," he began his most intensive search for military
information. It was "a bit titillating" to go into military computers;
mostly, he liked them because other hackers tended to stay away from
them. He explained that he sometimes got passwords by guessing them,
242   •    CygERPUNK
sometimes by transferring a file of encrypted passwords out of a computer
and comparing it with words in a standard dictionary using a standard
decoding technique.
   Hess rattled off the list of computers he had been in. In his hundreds
of attempts, he got into about thirty computers, with full privileges on
at least half a dozen of those. And he acknowledged his hacking as
something he had been irretrievably drawn to.
   "How is that?" the judge asked.
   "Hacking is an addiction, Your Honor," Hess said. "And as long as
you stay out of trouble, the addiction is difficult to shake." When he did
get into trouble in June 1987, Hess testified, he stopped hacking entirely.
    Hess's principal defense was that his hacking had never had anything
to do with the espionage. Yes, he had made copies of innocuous soft-
ware, software that anyone could buy, none of it on the COCOM list of
technology forbidden to cross into the East Bloc. And he had given that
software to Peter Carl with the knowledge that a KGB officer was buying
it on the other end. But that business arrangement was a separate matter
altogether from his hacking adventures. The only material to have gone
from him to Carl to Sergei, Hess insisted, was the harmless public-
domain software he had copied.
    In a way, it was a brilliant defense strategy. From the day of his arrest,
once confronted with the testimony of others, Hess knew he couldn't
deny that he had been involved in espionage, but there was no proof
that he had sold any of the fruits of his electronic wanderings to the
Soviets. Hess had long since decided that the Laszlo Balogh letter was a
flimsy piece of evidence at best. He didn't express this theory in court,
but Hess was convinced that the frizzy-haired Stoll must really be an FBI
operative who had worked together with the agency to fabricate the
SDInet material, then feed it to Balogh directly and frame Hess. Even
with the Balogh letter in hand, Hess would argue, he didn't see how the
prosecution could establish a strong connection between Balogh and
    Even Kohlhaas was impressed by Hess's self-confidence, but he also
believed Hess was lying. Kohlhaas had his own careful plan for proving
that Hess did transfer information out of LBL and other computers. Cliff
Stoll would be the witness to establish that. As far as Kohlhaas was
concerned, if Stoll simply repeated the story he had told during his
interrogations, especially the part about watching data travel from his
computer to Hess's, the case was open-and-shut. It didn't take much
imagination to conclude that Hess was storing the information for a

Pengo was to testify in Celle on January 19. He had received ample
warning from his lawyer that he should tell the absolute truth, and he
planned to comply, in his own way. But rather than get nervous, he
decided to approach his time on the witness stand as pure theater. So he
gathered a small group of friends-most of them from Berlin's punk
scene-and drove from Berlin to Hannover. Once the group arrived at
another friend's place in Hannover, an all-night party ensued, with the
consumption of copious hash. Pengo and his entourage were in some
disrepair when they arrived at the Celle courtroom. Dressed in his stan-
dard black jeans, black turtleneck and heavy black boots, Pengo strode
to the witness table and began to answer questions.
   "How old are you?" asked the presiding judge.
   "Uh, twenty-one," Pengo replied haltingly.
   "Your profession?"
   "Programmer." Since 1985, Pengo said, he had been working as a
programmer and studying computer science at the technical university
in Berlin.
   "What semester are you in?" the judge asked.
   At a loss, Pengo reached into his pocket and pulled out his student
identification. "Eighth," he pronounced after consulting the document,
and laughter rippled through the courtroom.
   Pengo might have figured that he had little to lose by being cheeky in
the courtroom. He was there strictly as a witness, and not as a defendant.
In his own investigation it looked as if he would soon be in the clear,
the charges against him dropped. Besides, he was happy to put on some-
thing of a show for his friends. One of them, a young woman with hair
dyed violet, was the object of a budding romantic interest. Pengo's
refusal to yield in the slightest to those in authority, especially in a
nation so conscious of rules and regulations, carried with it a certain
perverse appeal: Pengo was someone who wasn't frightened to speak his
mind, to shrug off a rebuke.
   Those authorities born and raised in West Germany who were in-
volved in the case-Kohlhaas, Judge Spiller, even Pengo's own rela-
tively conservative lawyer-might have explained away Pengo as a
product of the chaotic sensibilities that dominated West Berlin. But they
244 A     CygERPUNK
could just as easily have rationalized his behavior by saying he was part
of a generation that had managed to divest itself of the exhausting and
burdensome thing known as the German past. If postwar Germans hated
the Nazis and their children hated the Soviets, many of those in Pengo's
generation could only be described as indifferent.
   Pengo recounted his first meeting with Hagbard at the Chaos gather-
ing at the end of 1985. "For Hagbard, hacking was a skill with political
relevance," Pengo said. "His belief was that you have just one chance in
your life to influence world politics, and with all this hacking knowledge
in hand, this was it."
   Pengo also told of the seeds of the idea, of his trip with Peter Carl to
East Berlin and of the meeting with Sergei. "We went to a restaurant
afterward and talked about God and the world," Pengo testified.
   After that first trip, Pengo testified, he waited for Sergei's feedback.
To his great disappointment, none came, and by the beginning of 1987
Pengo's role in the enterprise had fizzled. Carl continued to drop by "for
coffee" and ask him for source code, but Pengo couldn't deliver them.
"The whole thing was more hot air than anything else," he declared.
After a while, Carl stopped calling. By that summer, Pengo said, he and
Dob were no longer speaking, and he had never had much contact with
Hess. Hagbard, he said, was completely "outgespaced," talking of noth-
ing but conspiracies and having religious hallucinations. Once, Pengo
recounted, when he and Hagbard were on a train together that was
traveling 180 kilometers per hour, Hagbard wanted to open the train
doors and jump out.
   Pengo turned himself in to the authorities, he said, not because his
conscience weighed on him or because of a recognition of a moral
breach, but because he "had to find a way to get myself out of this mess."
   It was the first time Pengo had had any direct contact with his erst-
while friends Carl and Dob in nearly two years. Both defendants sat next
to their lawyers and frowned. Dob barely moved, only occasionally
reaching up to play absentmindedly with his beard.
   The judge's questioning turned to Pengo's hacking career. "Where
have you hacked?"
   "I've been all over the world," Pengo answered.
   "Which companies have you hacked?"
   "I don't know," Pengo said. "The computers don't tell me what com-
panies they're from."
   "Did you share the passwords you got with others?"
   "No, I always kept the passwords I had for myself."
   "How many did you have?"
   Pengo took a few moments to consider this one. "In the course of my
active hacking time? Maybe fifty or a hundred."
   Pengo then decided to bring up the science fiction that had inspired
him in this esoteric enterprise. He described John Brunner's Shockwave
Rider and William Gibson's Neuromancer, and their cyberpunk anti-
   "Did the people in the books get a lot of money?" the judge asked,
clearly missing whatever point Pengo was trying to make. Pengo's friends
in the back of the courtroom giggled.
   "The chief thing for me was the adventure, suddenly being inside a
   Then it came time to ask about the Securepack software that Pengo
had stolen from Digital's Singapore computer. A witness from Digital
had already testified that it was difficult to put a market price on the
security program, as it was restricted to Digital's internal use, but it had
cost $375,000 to develop. Digital had also sent a lawyer from its Munich
office to sit at the trial every day. She was particularly interested in
Pengo's testimony.
   There seemed to be some confusion on the judges' part over whether
Pengo had copied the software once in late 1986 and a second time a
few months later. The computer kept track of the last time the software
had been worked with and not the time it was copied out of Singapore.
So when Securepack was found inside the reconstituted Rainbow, Febru-
ary of 1987 referred to the last time Pengo had looked at the software
and nothing more. But he couldn't seem to make that clear to the judge.
   "Did you copy Securepack twice?" asked the judge.
   Like Dob, Pengo was short on patience when it came to explaining
the fine points of modifying software. "Why should I copy it twice from
Singapore when I have a complete version in Berlin?"
   Pengo thought he had done pretty well. His friends thought so, too.
In fact, he seemed to be emerging barely scratched from the espionage
ordeal. It was an ordeal that had landed two of the others in jail for
nearly a year, plunging Dob into irretrievable depression and causing
Peter Carl's ex-wife to cut off what little remaining contact he had with
her and their daughter. A third had taken his own life. Markus Hess,
the wily hacker of LBL, was the only other one whose life had proceeded
on a relatively steady course. Though Hess viewed Pengo as ice cold,
and Pengo wanted little to do with someone of Hess's conservative
nature, they were in some ways similar. Their middle-class upbringings
246   •   eygERPUNK
and concerned parents had given them a sense that they could negotiate
their way out of tight spots. Their styles were different, but they were
both able to enlist sympathy and help. Hess had had the more traditional
upbringing, but both knew better than the others how to manipulate the
system to their own ends.
   Cliff Stoll was to testify on January 30, 1990. It was Stoll's testimony
that Markus Hess at once dreaded and awaited eagerly. Depending on
what Stoll might say, the American's testimony could incriminate Hess
more than Hess's own testimony. On the other hand, Hess planned to
engage Stoll in a match of technical wits.
   By late 1989 Stoll was in some disfavor among his colleagues in the
U. S. Some of his peers questioned his talent as a computer scientist,
which shouldn't have been surprising, since he had never claimed to be
one. Stoll was by no means the first system manager to have successfully
beaten a hacker at his own game. And he certainly wasn't the first to
impart a small but invaluable warning to the computer community at
large: when choosing passwords, don't choose something obvious such as
your spouse's name or your dog's name, or even an English word. Other
security experts had been saying that for years. But Stoll somehow struck
a nerve. His unorthodox personality had made him a perfect media icon.
He had written a breezy first-person book about his zealous hunt for the
odious hacker that became a best-seller and for several weeks he was a
fixture on television talk shows. There was no avoiding his message: the
nation's computers were at risk.
   When Stoll arrived in Celie, he was flanked on one side by an FBI
special agent, whom the West German spectators assumed was a body-
guard, and on the other by an American public television crew filming a
documentary on the amazing hacker catcher from Berkeley.
   First, the judge asked Stoll simply to tell the story of what had hap-
pened at the Berkeley laboratory in late 1986. Stoll had already spent
the better part of a year refining his story and was happy to tell it once
again, starting with the seventy-five-cent discrepancy and ending with
the news from West Germany in June 1987 that the hacker had been
   Gesturing wildly and speaking through an interpreter, Stoll relived
for the judges, the prosecutor and the spectators his alarm and concern
as he saw the hacker break into military computers and search for specific
military information. As soon as Stoll began to tick off the military sites
the hacker had been in, the judges started scribbling notes: the Jet
Propulsion Laboratory, Mitre Corporation, Anniston Army Depot, the
Redstone Army Missile Base in Alabama and an Army computer at the
Pentagon. The hacker searched for such hair-raising words and phrases
as nuclear, Strategic Air Command and stealth bomber.
   "Excuse me," interrupted the judge. "What does stealth mean?"
   "I consider myself lucky that I don't know what it means," answered
Stoll. And just as he had done the previous summer before the police in
Meckenheim and the magistrate in Karlsruhe, in court in CelIe Stoll
used such words as copy and print out to describe what he saw happening
on his computer screen in Berkeley. "As an astronomer, I wasn't all that
familiar with military stuff," Stoll said. "So it surprised me to see space
shuttle information being copied out." Once he had planted the false
SDInet inside the LBL computer, Stoll testified, the hacker "spent sev-
eral hours copying it back to his own home."
   This was the very point on which Hess and his lawyer were poised to
attack Stoll's credibility. Hess's lawyer spoke up. "We seem to be having
some trouble with the language," the defense attorney said. "Herr Stoll,
what is really meant with print and copy?"
   Stoll began to bounce in his seat. "Excellent question! Excellent!
Excellent!" he exclaimed, as if he were in front of a freshman computer
science class, encouraging a clever student. "To print means 'to look at,
to list, to see.' "
   Once that point had been set straight, it was clear that what Stoll
had said in his testimony during the summer, and what he was repeating
in the Celle courtroom, was scant proof that the hacker was downloading
and storing anything at all. It could well have been months of browsing,
mere electronic joyriding and nothing more. Prosecutor Kohlhaas saw
his case weaken.
   Stoll's testimony lasted nearly three days. Most of that time Stoll
spent reciting page after page of his logbook. Exact dates and times of
the break-ins were of particular importance, as German telephone offi-
cials were expected to come to court immediately after Stoll's testimony
with corroborating evidence matching the dates and times of their tele-
phone traces to Markus Hess with the dates and times in Stoll's logbook.
   On the final day of Stoll's testimony, it was the defendants' tum to
interrogate the witness. Neither Dob nor Carl had much to ask Stoll.
Hess, on the other hand, was ready for some confrontation, a battle of
technical prowess. Hess first asked Stoll if it wasn't true that he had little
way of knowing that it was the same hacker inside the computer each
time. Stoll agreed that there were some break-ins that occurred on VMS
computers, not UNIX computers, and that with the exception of the
248   •   Cyg£~PUNK

traces made directly to Hannover, he had no way of knowing that it was
the same hacker each time. Hess also wanted to know if Stoll's SDInet
file was his and Martha's own invention entirely. Hess's motive, appar-
ently, was to provoke Stoll into admitting that he was in league with the
FBI or the CIA. Hess would maintain to the end that the FBI had
assigned Laszlo Balogh to ask for the information, in order to speed up
the pace of the investigation.
    Stoll's answer was curt. "It was fictitious," Stoll responded, "but I
tried to make it look as realistic as possible."
    Hess wouldn't give up. He questioned Stoll's "typing rhythms" exper-
iment. How could this prove anything given the great distance? Any
unique characteristics would certainly be lost, he suggested, in the laby-
rinth of networks between Germany and California. Stoll countered that
he had believed the validity of his experiment.
    The interrogation at times required that Hess and Stoll gather in front
of the bench. Together they bent over computer printouts to examine
the details of their arguments. It was odd to see the hunter and the
hunted analyzing documents like two academic colleagues. One morning
before court was in session, Stoll and Hess happened to arrive before
anyone else. It was an awkward but cordial encounter. Hess suggested
that the two get together for a beer sometime. Stoll, an outspoken
nondrinker, nodded politely and excused himself.
    By the time Stoll's testimony in CelIe had ended, it was unclear
whether he had helped or hurt the prosecution. Ask Stoll what he
considered to be the ultimate value of his sleuthing and he would say,
"Chasing the bastards down." For his part, Kohlhaas was grateful for the
existence of Stoll's logbook, and pleased that Stoll had kept such careful
note of the times the hacker had been inside LBL. Otherwise, the Bun-
despost's telephone traces would have been difficult to present in court
as incontrovertible evidence. Then again, Kohlhaas had to admit that
establishing Hess as the LBL hacker wasn't as relevant to the espionage
charge as proving that the information he had seen had gone to the
Soviets. And no one would know until the verdict was delivered how
the testimony was viewed by the judges. For the judges to impose stiff
sentences, the prosecution had to demonstrate that sensitive military
information had changed hands. Prosecutor Kohlhaas's efforts to bring a
U.S. military expert to Celle to assess the sensitivity of the material had
ended in frustration. And in the end, Hagbard's testimony to the police,
in which he had claimed that hundreds of logins to sensitive U.S. and
European computers went to the East, was stricken from the record for
lack of credibility.
   At the time Markus Hess was caught, the media were making him
and his friends out to be half the KGB, when in reality Hess was a fairly
conventional young man who claimed to take inspiration from a movie.
As notorious as the hacker spies became, the judges would have to admit
that the defendants really did very little damage to Western security. In
the end the only person who was damaged irretrievably was Hagbard.
   So it wasn't a great surprise when the sentences turned out to be light.
Peter Carl, whom the court viewed as the one most actively engaged in
espionage itself, and the one who exhibited the most "criminal energy,"
got two years and a 3,000-mark ($1,500) fine. Hess was sentenced to a
year and eight months and ordered to pay 10,000 marks; Dob got a year
and two months and a 5,OOO-mark fine. But instead of having to serve
prison time, all three defendants were put on probation. At the time of
their actions, the judges remarked, Dob and Carl were operating in such
a drug-filtered haze that they were in no position to recognize the gravity
of their deed.
   In his conclusions to the court, presiding judge Spiller said he believed
the hackers had indeed sold information out of military computers to the
KGB, and that the KGB had probably found the information very inter-
esting. But, he added, Sergei couldn't have seen it as terribly valuable
because he didn't yield to the hackers' demands for a million marks. In
the end, all that hacker know-how went unappreciated, even by the

     P   hi! Lapsley, an engineering student at the University of Califor-
nia at Berkeley, was puzzled. No sooner had he logged in to a Sun
Microsystems workstation than it was clear something was amiss.
   Computers such as the Sun run dozens of programs at once, so it is
routine for people like Lapsley who maintain them to peek periodically
to see which programs are currently active. But on November 2, 1988
he saw, hidden among dozens of routine tasks, a small program con-
trolled by an unusual user named daemon. Daemon is not the name of
any particular human, but an apt label conventionally used for the utility
programs that scurry around in the background and perform useful tasks.
But this program was not one that Lapsley recognized.
   "Is anyone running a job as daemon?" he asked the others in the
"fishbowl," room 199B at the Berkeley's Experimental Computing Facil-
ity. People shook their heads. Then somebody else in the room pointed
to one of the screens, where a program that monitored the status of
various other computers in the department was displayed. Lapsley looked
more closely and discovered that a number of people appeared to be
trying to log in to other Berkeley computers. He decided it must be an
attempted break-in. At least once a year, someone tried to break into
the computers in Cory Hall, which houses the school's prestigious elec-

254   A   QygERPUNK
trical engineering department. The school year wouldn't be complete
    Whoever this intruder was, he was apparently quite intent on getting
in, trying time after time to log in to Berkeley's computers. So Lapsley
started to jot down the names of the machines from which the break-in
attempts were coming. But he was startled to see that they were scrolling
by faster than he could write them down. In fact, they were coming so
rapidly they were scrolling straight off the screen before he could even
read them. At that point, Lapsley realized it wasn't a person at all who
was trying to break in. It was a program. When it wasn't running as
daemon, it was running under the names of other users.
    The program kept pounding at Berkeley's electronic doors. Worse,
when Lapsley tried to control the break-in attempts, he found that they
came faster than he could kill them. And by this point, Berkeley ma-
chines being attacked were slowing down as the demonic intruder de-
voured more and more computer processing time. They were being
overwhelmed. Computers started to crash or become catatonic. They
would just sit there stalled, accepting no input. And even though the
workstations were programmed to start running again automatically after
crashing, as soon as they were up and running they were invaded again.
The university was under attack by a computer virus.
    Lapsley called Mike Karels, a programmer a hundred yards away in
Evans Hall, an imposing concrete tower and home to the school's com-
puter science faculty. As the principal programmer at the Computer
Systems Research Group, Karels was the scientist most knowledgeable
about Berkeley UNIX, the operating system widely adopted by universi-
ties and research institutions everywhere. If anyone would have good
advice, it would be Karels.
    All Lapsley got from Karels was a short, stiff laugh, then, "So you've
got it too, huh?"
    After another thirty minutes of puzzling over the enigmatic intruder,
Lapsley and others in the fishbowl discovered that the program was
expanding beyond Berkeley. Peter Yee, another undergraduate working
with Lapsley, logged in to a computer at NASA's Ames Research Center
fifty miles to the south and saw it there. And when Lapsley logged in to
a computer at Berkeley's sister campus in San Diego, he saw it there,
too. By the time a call came from a system manager at Lawrence liver-
more National Laboratory to say it was on his machines, there was no
doubt that this was no local problem. It was all over the nationwide
network known as the Internet.
                                                            RTH    T   255

The people who care for the networks of computers used on college
campuses and scientific research centers had spent many years preparing
themselves for various eventualities. And for years, computer scientists
had spoken theoretically of the possibility of a program running loose in
the network. But no one was prepared to cope with the massive assault
on November 2, 1988.
   Within minutes of each other, computers all over the nation felt the
presence of the rogue program. Shortly before 6:30 P.M., computer man-
agers at the Rand Corporation in Santa Monica, a famous think tank
where Daniel Ellsberg once photocopied the Pentagon Papers, noticed
that their computers were unusually sluggish. There appeared to be a
program running that was robbing the computers of speed and slowing
them to a near standstill. Fifty-five minutes later, across the country in
Cambridge, Massachusetts, computers at the MIT Artificial Intelligence
Lab were under attack. Almost immediately after penetrating MIT, the
program struck Lawrence Livermore National Laboratory and a computer
at the University of Maryland. Then it struck Stanford, Princeton and
the Los Alamos National Laboratory in New Mexico. Once inside a
computer, the program propagated to other computers much like a bio-
logical virus. On some computers there were hundreds of copies of the
program running, slowing the machines to a halt. Even when its at-
tempts to get into a new computer were unsuccessful, this electronic
virus's repeated knocks on the door were often enough to cripple the
machine. And even after it was killed, it would reappear almost imme-
diately. Moreover, once it entered a workstation, the program had a
mysterious way of finding other computers to attack. Throughout the
night it hopped back and forth through the network, setting off havoc
wherever it touched down.
   System managers around the country, responding to frantic calls from
night operators, were racing to their offices at 2:00 A. M., 3:00 A. M., and
4:00 A. M. to wrest back control of their computers. Others noticed it
when they had trouble logging in to their institutions' computers from
home. Still others wouldn't learn about the program until they arrived
at work on Thursday morning to find their computers besieged. Program-
mers at the University of Illinois at Champaign-Urbana were convinced
they were going to have to rebuild the software for their entire campus
computer system from the ground up.
   Worse than what could be observed about the program was the fear
256   A   CYJg£RPUNK
that it might be a Trojan horse program-apparently innocent, but
carrying a string of code instructing the computer to carry out a specific
damaging instruction at some later time. System administrators at an
aerospace company in San Diego got so frightened by the threat of a
malevolent string of code that they pulled everything off their computers
and installed their most recent set of backup tapes.
   When the program started entering computers shortly before midnight
at the Army's Ballistic Research Laboratory in Maryland, system man-
agers feared invasion by a foreign power. And since the program came
in over the network, they were afraid it might also be taking Army data
out over the network. Assuming the worst, at 10:00 on Thursday morn-
ing Mike Muuss, the chief Ballistic Research Laboratory system program-
mer, did what dozens of other managers across the Internet had already
done: he disconnected his computers from the network. The laboratory
would stay off the network for nearly a week.
   Taking computers off the network stopped the program from coming
in or leaving, but it had the unfortunate side effect of cutting off com-
munications among people accustomed to staying in touch with elec-
tronic messages. Few people thought to pick up the telephone, and those
who did were at a loss: the electronic network had become the sole form
of communication for most computer experts, who seldom bothered to
give out their telephone numbers.

If any place could grapple with such a bizarre and troubling situation,
Berkeley could. The university was the birthplace of the very version of
the UNIX operating system that the rogue program was targeting when
it broke into computers on the network. During the evening it became
apparent that the intended targets of the mysterious program were com-
puters made by Sun Microsystems and the Digital Equipment Corpora-
tion, two of the most common machines on the Internet.
   Fifteen minutes after Lapsley first noticed it, the program had broken
into at least thirty workstations on the Berkeley campus. From the way
it was acting, it appeared to be a selective beast, setting its sights on
machines that were connected to as many other systems as possible. It
used simple, quick and powerful methods to break in immediately. Two
Berkeley computers were especially attractive targets. One, called
CSGW, which was a gateway to local area networks on the Berkeley
campus, crashed after dozens of copies of the virus arrived. So did
UCBVAX, a major gateway to the Internet. It seemed that infecting
                                                          RTH   ~   257
such a vital organ had strategic value for the program, increasing many-
fold the number of computers it could reach in just a single hop. Thus,
UCBVAX was under constant attack. Still, the team of Berkeley de-
fenders decided against pulling their computers off the network. That
would have been admitting defeat. The challenge, they decided, was to
stay connected to the network and still kill off the program.
   One of Berkeley's first tasks was to capture ~ snapshot of the program
as it was running, in effect to catch it in freeze-frame, and then to
analyze it. From there, they could examine strings of code and try to
figure out what the program was doing. But most of it was encrypted, as
if whoever wrote it knew someone would take such a snapshot. The
Berkeley programmers found that the coding scheme used to obscure the
program was a simple one-unscrambling the data was much like
the child's game of translating words from pig latin. The programmers
quickly uncovered the original instructions. The snapshot of code also
told the programmers that the program was trying to crack passwords
using what was known as a dictionary attack, comparing encrypted pass-
words to an on-line dictionary that had been encrypted. The snapshot
also showed them that the program was using cracked passwords to get
onto one system, then go from there to other computers by taking ad-
vantage of the fact that a password validation on one computer often
grants access to other computers across the network.
   Other things quickly became obvious. The Berkeley programmers
soon figured out that the program was exploiting a subtle flaw-or
bug-in a communications program called sendmail, which it used to
send messages and data between computers over the network. The flaw
in sendmail arose from the subtle concatenation of two features of the
program, much as a binary poison gas is deadly only when two inert gases
are combined. One feature made it possible for someone at a remote
location to embed a program in a message. Instead of being handled as
an electronic letter, the message fooled the computer into running it as
a program.
   The second feature allowed those programmers who needed to
"debug" or maintain the mail program to examine mail connections over
the network. This "debug" feature made it possible to switch on the first
feature from a remote location. Once the first feature had been switched
on, a program embedded in electronic mail could be sent to run on
another computer immediately. The combination of the two features,
known to only a few, proved to be a glaring loophole in the mail pro-
258   ...   CYEERPlINf<
    Whoever had written the rogue program made use of this obscure flaw
to send a small "scout" or "grappling hook" program across the network.
This program in turn immediately called back and brought over the main
body of the virus. Having taken hold of each new computer, the process
would repeat itself indefinitely. That much, at least, was obvious.
    Their first look at the program suggested to the Berkeley group that
the invader had no intention of destroying data. Apparently, it exam-
ined actual information inside computers only in order to find ways of
breaking into other systems. But they realized that was just a superficial
impression. The possibility of a Trojan horse still lingered. The only way
to determine what it was actually doing would be to pick it apart line by
line, a painstaking task that could take days or weeks. Until the program
had been thoroughly analyzed "with microscope and tweezers," as at the
Massachusetts Institute of Technology titled a later paper on the virus,
there was no knowing what kind of dangers lurked inside.
    For the next three hours, programmers at both the Experimental
Computing Facility in Cory Hall and the Computer Systems Research
Group (CSRG) on the fourth floor of Evans Hall worked simultaneously
at shaking the program out of their systems and building a dearer under-
standing of how it worked. If it were a playful hoax, they reasoned,
wouldn't it have come with a set of instructions on how to get rid of it?
But there- were no such instructions, and all potentially useful details
hidden inside were encoded, sheltered from prying eyes. The program
tried to remain hidden by giving itself the name of an innocuous com-
mand that its author obviously hoped would avoid scrutiny. Apparently,
the idea was that to anyone taking casual stock of the computer's activ-
ities, nothing would appear out of the ordinary. And to further elude
detection, like a chameleon the program constantly changed its identi-
fying number, taking on new aliases to make itself less conspicuous.
    By 11:00 P.M., most of the Berkeley programming staff had congre-
gated in one of the two computer labs. Keith Bostic, a twenty-eight-
year-old programmer at CSRG, was in his office at Evans Hall, working
with Mike Karels. As two of the principal software engineers behind
Berkeley UNIX, they had good reason to take the attack personally.
Bostic had seen his machines hacked and crashed by outsiders before,
but this episode was on an entirely new scale. Meanwhile, Lapsley and a
group of others gathered in the fishbowl at Cory Hall. At 11:30 P. M.,
Peter Yee sent a message from Berkeley to an electronic mailing list on
the Internet: "We are currently under attack from an Internet virus,"
                                                           RTf1   ... 259
the message began. "It has hit UC Berkeley, UC San Diego, Lawrence
Livermore, Stanford and NASA Ames."
    Fueled by adrenaline, sugar and caffeine, the Berkeley group was
meeting the break-in head-on. A software invader that brought scores,
perhaps hundreds, of computers to their knees was just the sort of night-
mare that every computer manager feared. At the same time, it was as if
someone had just handed the group at Berkeley an imposing crossword
puzzle to solve-the ultimate challenge. And the possibility that the
program could contain virulent code infused the evening with tension.
Someone made a sign that read, "Center for Disease Control," and taped
it to the door of the Experimental Computing Facility.
    Sometime after midnight, Lapsley walked back to a machine room
containing most of Cory Hall's largest computers and began the arduous
task of going from machine to machine, plugging the holes the invader
was using and killing off all copies of the foreign program. He recon-
figured each system with a patch that blocked the sendmail loophole.
And in Evans there were dozens more. One by one, the Berkeley com-
puters were immunized from the attacker. Berkeley had survived the
    By 3:00 A.M. Thursday, the tired programmers knew enough about
the program to issue a broad alert to other computer sites. Bostic sent
messages to several electronic mailing lists describing how to fix systems
in a way that would stop the program. His message reached some parts
of the network, but, unfortunately, not the centers that had already cut
themselves away from the network and were working in solitude to stop
the rogue program. The Internet sites that stayed connected found their
messages bogged down by a form of electronic gridlock; the program had
clogged some of the network's crucial mail machines. In some cases it
took messages hours, even days, to travel routes that normally took a
few minutes.
    Exhausted, Bostic went home to sleep. Lapsley stayed in the fishbowl.
He knew the program was somehow using other methods to break in,
but he hadn't yet been able to figure out exactly what they were. At 8:30
A.M. Thursday, Lapsley finally went home, too.
    It had been one of the most harrowing nights that anyone in the
computer science community had ever faced. For others, though, it was
just the beginning. The following morning, news of the program spread
around the country. Word of the previous night's invasion had circulated
not just among computer scientists but in the national press as well.
260   •   CygfRPUNK
During the day, Lawrence Livermore National Laboratory, one of the
nation's leading weapons laboratories, held a press conference to describe
the attack in detail.

The anonymous caller to The New York Times on Thursday afternoon
made it clear that he didn't want to disclose who had written the Inter-
net virus. He just wanted to let the Times know that the person who had
written it was a well-intentioned soul who had made a terrible mistake
in the code.
   The switchboard first routed the call to the paper's national news
   "Uh, I know something about the virus that's going around," said the
   "What virus?" The editor sounded confused.
   "The computer virus that's crashing computers all over the country."
   "Give me your number and someone will call you back," said the
   The editor gave the message and a telephone number to John Mar-
koff, the paper's computer reporter. Markoff had already heard about the
incident. He had received a call at 10:00 that morning from Cliff Stoll,
the Berkeley astronomer who had gumshoed his way to the bottom of
the West German hacker case a year earlier. Stoll, who was now working
at the Harvard-Smithsonian Center for Astrophysics, told Markoff he
had been up the night battling the program, which had swamped fifty of
the center's machines. The reporter then spent the morning calling
universities and research centers to see if they, too, had been infected.
One of his calls was to an occasional contact at the National Security
Agency. Markoff had called the NSA in the past on security-related
stories, and he thought his contact there might tell him something about
what was going on. But his contact wasn't there and his call wasn't
   Nobody Markoff spoke with at universities, corporations or military
sites seemed to have any inkling of the program's origin. Theories ranged
from prankster to foreign agent. So the anonymous call to the Times was
intriguing. When Markoff returned the call to a number in the Boston
area, it was immediately clear to the reporter that the caller, who would
identify himself only as Paul, knew a great deal about the program and
how it was written. The excited-sounding young man said he was a friend
                                                          RTH .., 261
of the program's "brilliant" author. The author, Paul said, had meant to
write a harmless virus, but had made a small error that caused the
program to multiply around the network.
    By Friday, Paul and Markoff had talked on the phone several times.
Paul referred to the author only as Mr. X. The two went back and forth
about what kind of trouble the program's author might be in. By this
time, news of the program, which by now was being described alternat-
ingly as a "virus" or a "worm," had been on the front pages of newspapers
and on the nightly television news around the nation. It was the first
wholesale assault ever on the nation's computer systems. More disturb-
ingly, military computers had been infiltrated. The program had been
contained, but there was still no full assessment of the damage it had
   Then Paul made a mistake. During one conversation, instead of say-
ing "Mr. X," he slipped and referred to his friend by the initials rtm.
Markoff was close enough to the computer scene to recognize rtm, in
lower case, as a likely computer login. After hanging up, he phoned Cliff
Stoll, with whom he had been trading information all morning. From
his computer at home, Stoll used a network "white pages" directory and
finger, a utility program that acts as a computerized directory-assistance
tool giving users limited information about others on the network. When
Stoll fingered rtm on the Harvard University computers, he retrieved the
name Robert Tappan Morris, identified as a graduate student at Cornell
University. A phone number and an address were included. Stoll called
the Times.
    Now Markoff had a name, but he still didn't have a story. Without
an independent confirmation, he couldn't be sure if Paul was telling the
truth, or if the rtm the caller was referring to was the same person as
the rtm Stoll had just found on the Internet directory. Markoff called
the Cornell telephone number. Nobody answered.
    Late that afternoon, the NSA finally rang back. The caller was Bob
Morris, a computer security expert who was the chief scientist at the
agency's National Computer Security Center.
    "I think I know the name of the person who wrote the virus," Markoff
    "Who is it?" Morris shot back.
    Markoff bridled. "I'm not going to tell you. You're the computer
police. "
    A strained conversation followed and before long it became clear that
262   A   CYEfRPUNK
Morris knew exactly who had written the program. In fact, the man
from the security agency appeared to know more about the event and its
perpetrator than the reporter did.
    Finally Markoff said, "I think the program was written by Robert
Tappan Morris."
    "You're right," Morris answered, giving the paper the confirmation it
needed. They talked for a while longer. It had been a chaotic day of
sifting through dozens of sometimes contradictory reports on the event.
Markoff still wasn't certain what to do with this new information. If the
program had been written by a Cornell graduate student, how was it that
computer security experts at the National Security Agency already knew
that? What was going on, anyway? Just as he was about to hang up, he
had a sudden thought. "Isn't that a funny coincidence," he said. "You
both have the same name."
    Without missing a beat, Morris replied, "That's no coincidence. He's
my son."

Bob Morris entered Harvard in 1950 as a chemistry major. His father
had been a salesman for an etching and engraving company and Bob
thought he too might end up a salesman, perhaps at a place like Du
Pont. He interrupted his studies to spend two years in the Army, and
when he returned to Harvard he decided his options would be broader
with a degree in mathematics, so he switched majors in his senior year
and completed all the remaining math requirements in time to graduate
that same year. He earned his master's in 1958 and embarked on his
Ph.D., with plans to write his dissertation on number theory. In 1960,
he took a summer job at Bell Labs.
   By the time Bob Morris got there, Bell Labs had as many scientists
with Ph.D.'s as did most universities. Since its inception in 1925, Bell
Labs has been a monument to the degree of innovation that can spring
from within the walls of a monopoly. From its considerable investment
in both basic and applied research AT&T has seen ample reward. Bell
Labs scientists hold nine Nobel prizes. There are few other industrial
research institutions like it.
   Some areas of research were of uncertain commercial value, but it was
rare that a project would be blocked simply because its immediate prac-
tical benefit to telephony was unclear. AT&T managers were astute
enough to recognize that significant breakthroughs are born not of rules
                                                           RTH   T   263

or plans but of people, and those in the upper echelons at Bell Labs
governed their hiring practices accordingly. Bob's summer job was to
stretch into two summers and eventually into a twenty-six-year career.
He started out in telephony, working on data transmission, but he was a
frequent visitor to the mathematics department, some of whose scientists
were developing computer software. Morris made the acquaintance of
Doug McIlroy, a mathematician who was to become his boss and close
friend for many years. After a year or so Bob transferred to the mathe-
matics department, took over the job of someone who left and became
so entrenched that there didn't seem to be much point in returning to
his Ph.D. work. Besides, he was pushing thirty, an age at which almost
all of the best mathematicians have already done their major work.
    It didn't take long for him to migrate into computer research. By
then, in the sixties, the field of computer science had begun to burst
open with new discoveries. Computers were appearing everywhere
within the scientific community. Practically every mathematician who
came near a computer had a chance of doing something original. None
of those people had been brought up on computers, of course, but all
were captivated by them. Mathematical solutions to problems were more
and more often complemented by a computer's ability to calculate rap-
idly. The notion of programming, of being able to instruct a machine to
do almost anything, of inventing artificial worlds on a computer, was a
source of absolute fascination for Mcllroy.: Morris and their colleagues.
Morris, in particular, seemed to have an uncanny understanding of com-
puters. Whenever a convoluted computer problem presented itself,
bringing it to Morris's attention was almost certain to produce a creative
    Morris established himself early as a programming wizard. One of his
first displays of such mastery came when he joined in on a simple game
called Darwin. Darwin was the 1962 invention of McIlroy and a colleague
named Victor Vyssotsky. Vyssotskv thought it would be fun to have a
computer game in which the program played against other programs
rather than against people. The idea was to create a program that tried
to kill opposing programs. At the end of each round, the winning design
would be shared with the group. A concept well ahead of its time,
Darwin was a predecessor to a later program called Core Wars, a simple
computer game that became popular after the advent of the personal
computer. Core Wars came with its own simple programming language.
Players designed tiny software "warrior" programs, then turned them
264   ~   CYG6RPUNK
loose on an imaginary playing field in the computer's memory. The
winner was the program that disabled the opposing program and still
remained functioning at the end.
   The three young scientists had been playing around with Darwin for a
week when Bob came in one day with the toughest survivor of all. Bob's
program was composed of just thirty instructions, and its power lay in
the fact that it was adaptive: it learned how opponents were protecting
themselves and devised its attacks accordingly. Bob's program was un-
failingly lethal, and the game ended.
   Even in an environment where quirkiness abounded, Bob was re-
garded by his colleagues as an original. Shortly after arriving at Bell
Labs, he grew a beard, which was to remain aggressively unkempt for the
next three decades. An iconoclast by nature, Bob had a habit of chal-
lenging others' assumptions and would go to any length to make his
point. But he carried out his challenges playfully, never dogmatically. In
answer to a colleague's assertion that all the equipment in the computer
room was fireproof, he took a lit match to a computer tape's write-protect
ring (an attachment that prevents computer information from being
erased) and tossed it in the wastebasket, setting off smoke alarms and
causing pandemonium throughout the lab. And so acute was his sense
of a system's vulnerable points that as his colleagues buzzed proudly
around the computer lab on the day a new operating system called
Multics was first unveiled, Bob strode in and typed two specific charac-
ters he suspected would confound the system. They did. The computer
    It was such stunts that earned Bob his reputation as a scientist whose
biggest strength was his capacity for offbeat thinking. When he wrote a
program designed to spot typographical errors, the program contained no
dictionary and knew no English. Instead it was based on statistical prob-
ability; it sifted through a document searching for uncommon sequences
of characters. And it found many of the mistyped words.
    Bob loved having inside information, and he enjoyed possessing in-
sight into arcana that others were only vaguely familiar with. He often
subjected his colleagues to intellectual popquizzes. If he discovered that
someone down the hall had a passing interest in, say, relativity, he would
learn all that he could on the subject and start asking questions. It was
less a show of intellectual bravado than a sign of Morris's constant
    In a report for the twenty-fifth reunion of his Harvard class in 1979,
Morris wrote, "A long time ago I promised myself that I would learn to
                                                           RTH    T   265

read Greek, learn in some detail how the planets move in their orbits
and how to decipher secret codes. I have gone a long way toward keeping
all three promises." In his thirties, he had taught himself ancient Greek.
And one project that occupied him for nearly a year at Bell Labs was an
astronomy program for predicting planetary orbits. But the promise with
the most relevance to his work at Bell Labs was the third one.

In 1964, Bob was one of the first people at Bell Labs to have a terminal
in his home. His modem carried data back and forth at an excruciatingly
slow rate of 135 bits per second, about a tenth the speed of the slowest
of today's most common modems. Retrieving or sending even a small
amount of information was a process that left plenty of time to go get a
cup of coffee while the modem churned away.
   The terminal itself, called an IBM 2741, looked like an oversize IBM
typewriter with an IBM type-ball mechanism. The typewriter perched
atop a pedestal and inside the pedestal was a mass of electronics. Later
came a slightly faster terminal, the Teletype Model 37, a cumbersome
affair that was roughly half the size of a standard desk. The Teletype
terminal had a mechanical encoding grid of rods under the keys that
converted keystrokes into binary signals, which in turn made their way
over a modem into the Bell Labs central computer. Any Bell Labs sci-
entist with a terminal got to know the repairman pretty well. Every
repair visit finished with an oiling of the encoding mechanism; the next
time the terminal was used, fresh oil often dripped onto the user's legs.
   In the early 1960s computer security wasn't a problem. Locked doors
sufficed. It first became an issue with time-shared computers. The origi-
nal idea of time-sharing was simply that everybody could apparently have
his own computer when he needed it, with the cost shared among many
people. This was the first time people had thought about sharing the
power of a computer. With the development of time-sharing, there arose
the need for accounting and security mechanisms of some kind, because
more than one person at a time could use a computer.
   Multics was one of the first time-sharing systems that paid real atten-
tion to security as an explicit design goal. The main goal of Multics, a
joint research project of MIT, Bell Labs and General Electric, was to
make time-sharing commercially available. The dream was that Multics
would be a computer utility with capabilities far beyond those of existing
commercial time-sharing systems. It had to permit cooperation among
users who wanted it while guaranteeing privacy to others.
266 •     CYJgfRPUNK
   The earliest self-described computer hackers, those at MIT who
abhorred computer security, or anything else that would inhibit the
sharing of information and free access to computers, had it in for Multics
from the start. MIT hackers often tried to bring the system to its knees,
and occasionally they succeeded.
   But ultimately, Multics developed to the point of becoming too un-
wieldy. As Morris would describe it many years later, continuing to
support its development was "like kicking a dead whale down a beach."
Bell Labs pulled out of the project in early 1969, after which Multics was
adopted by Honeywell as a secure operating system to run on military
computers. But the "tiger teams" of the 1970s-groups of people who
were authorized to probe the security of Defense Department computers
by trying to break into them-put Multics computers through rigorous
tests and eventually got in. The teams even managed to confound the
system's meticulous audit trail, modifying it so there would be no trace
of a penetration of the computer.
   Breaking into computers in order to improve security was an impor-
tant tactic used by people who worked in the field of computer security.
Members of a tiger team were allowed to have at least limited access to
a target computer. That was one thing. But what about those would-be
invaders with no legal access at all? People like Bob Morris and Ken
Thompson, another talented computer scientist, thought about such
problems extensively. The first step, of course, would be for intruders to
find out what telephone numbers dialed in to a computer, perhaps by
using a scanning program that could dial every possible telephone num-
ber sequentially. Ten years later, it would be common for twelve-year-
old computer hackers to write programs similar to those they saw de-
picted in the movie WarGames. If a hacker's modem detected another
computer, signaled by a high-pitched tone, his next step was to log in
and identify himself to the computer's satisfaction by supplying an ac-
count name and a password. Unless the perpetrator already had inside
information of some kind, coming up with the correct password could be
the most difficult step. But once he had logged in, it was possible for
him, depending on the level of privileges achieved, to enter other com-
puters over a network illicitly.
   Bob's interest in computer security grew with the development of
UNIX, the successor to Multics. UNIX was a play on the name Multics.
Where Multics was complex and its name referred to computing in
multiples, UNIX signified simplicity and uniformity. UNIX began as a
backlash to the Multics system; it was developed for a small computer,
                                                              iCTH ... 267
and programmers grew to like it for two primary reasons: its flexibility let
them tailor it to suit the needs of whatever program they were working
on, and it was designed to be "portable," meaning it could be made to
work on computers of many different brands. Future versions of the
system grew slightly in complexity as new capabilities were added, but
each new edition of UNIX remained faithful to the principles of simplic-
ity. UNIX would bring fame to a few Bell Labs programmers and become
a fixture at universities and research institutions around the world.
   The UNIX development team consisted of two principals-Ken
Thompson and Dennis Ritchie-and a peripheral group that made
smaller contributions. Bob's work on UNIX involved the mathematical
functions of the software. Something as simple as asking for the time,
for instance, involved a calculation. But his main interest lay in writing
the encoding algorithm used in UNIX, the procedure that transformed
the uncoded, plain text in a file into encrypted text.
   When Bob wrote the crypt program, his fascination with ciphers in-
tensified. He was a mathematician, and his deepest interest lay in num-
ber theory, which typically involves the study of prime numbers and
creative uses of randomness. Cryptology is a natural extension of number
theory since it requires turning a message of clear text into a code
through manipulation of numbers. Cryptology is more than a mathemat-
ical discipline; it requires linguistic skills as well. To do exceptional work
in cryptology requires remarkable intuition and leaps of imagination.
Morris had that. He also had an ability to see the security holes where
others saw protection.
    In the mid-1970s, Morris was working on a method of cracking the
encryption machines developed during the 1930s by a Swedish cryptol-
ogist named Boris Hagelin. The machine, known as the M-209, was
considerably more sophisticated than the earlier German Enigma ma-
chine used by the Nazis in World War II, which was decoded in 1939 by
British cryptanalysts, including the famous mathematician Alan Turing.
The M-209, which looked like a cash register with letters as its keys,
coded messages in such a way that each letter was turned into one of
more than a hundred million possible substitutes. Morris devised an
elegant method for taking a passage of text encoded by the M-209 and
transforming it into clear, readable English without relying on machines.
At the same time, Jim Reeds, then a mathematician at the University
of California at Berkeley, came up with a different method for breaking
the code that could be done with a computer program. Reeds and Morris
learned of each other's work and, with the help of Dennis Ritchie,
268   •   CYE£RPUNK
created a program that would read encoded text and generate a clear
translation. The trio then wrote a joint paper describing their feat, and
submitted it to the academic journal Cryptologia. At the same time,
however, as a courtesy Bob sent a preprint of the paper to the National
Security Agency, whose mission-indeed, whose very existence-was
largely unknown to the general public at the time. The NSA spreads a
far-flung net for gathering communications intelligence in every corner
of the world. For example, when a Korean airliner strayed off course in
1983 and was shot down by a Soviet interceptor, NSA monitors captured
the radio conversation between the Russian pilot and his flight control-
lers. And in 1989, when the United States accused a German company
of selling materials that enabled Libyans to build a poison gas plant,
intelligence on the matter was gathered by a massive and permanent
NSAcommunications surveillance operation in Europe.
   The mission of the NSA, which was classified until recently, also
required that the agency maintain the world's best cryptographic capa-
bilities. The NSA maintained that it was not in its interest for the most
advanced cryptographic research to be widely disseminated to the public.
So it was perhaps not surprising that shortly after submitting their paper
to the agency for review, the three Bell Labs researchers received a visit
from a retired Virginia gentleman to discuss the impending publication
of the paper. He was, in fact, a former intelligence officer and still had
close ties with his former employers.
   The agency was divided, he told them. Some didn't see a problem
with the article, but one conservative group was opposed to any publi-
cation of information that would advance the public knowledge of cryp-
tography. The initial contact over lunch at Bell Labs led to other
meetings. The researchers traveled to visit agency officials several times.
In the end, the Bell Labs scientists decided to withdraw the paper.
   As Ritchie remembers the incident, it was at this time that Bob
Morris's flirtation with the NSA began. What went on inside America's
most secret intelligence agency held a certain fascination for all of them,
and for Morris in particular. Already, the NSA was a customer for UNIX
and the accompanying C programming language that the Bell Labs group
had designed. Morris was offered a summer appointment at the Institute
for Defense Analyses, the NSA's classified think tank. But at this point
all three still felt that if they took security clearances, it would mean
sacrificing much of the freedom they enjoyed as outsiders. They decided
to keep their contact with the computer spooks informal.
                                                         RTH    T   269

Anne Burr Farlow came from a long line of New Englanders. Moonfaced
and slightly plump, Anne was a music graduate fresh out of Bryn Mawr
College in 1959 when she moved to Cambridge to work as an office
assistant in the geology department at MIT. On occasion, a Harvard
graduate student in mathematics named Bob Morris would stop by
Anne's apartment to visit her roommate, but Anne didn't take much
notice of him until one day when he asked the roommate to a concert.
When the young woman declined the invitation, the serious young
student turned straight to Anne and asked if she would be interested.
She accepted. Their two-year courtship consisted of frequent ski trips in
the winter and long sailing trips in the summer. In June of 1962, Anne
and Bob married.
   When Bob decided to settle permanently into the Bell Labs job, the
young couple went house hunting. Bob, who had grown up in farming
country north of Hartford in the Connecticut Valley, wanted ample
privacy. They settled on a farmhouse in the small town of Millington,
New Jersey. The house dated back to 1740; it had few modem amenities
and abutted a steep wooded hillside, a wildlife preserve that served as a
woodlot for the family. The house sat on a nine-acre triangle of land on
a dead-end road. A two-acre field lay between the house and the Passaic
River. Unless the temperature dipped below twenty degrees, the nine-
room house was heated entirely by a wood stove and a fireplace.
   Meredith, the first child, was born three weeks after Bob and Anne
moved into the Millington house. Gradually the house filled up with
three large dogs and two more children. Robert was born in November
of 1965, Ben two years later. On the property, farm animals were accu-
mulating: sheep, chickens and geese. At least a dozen cats, "working
cats" as Bob called them, roamed freely. When Meredith asked for a
horse one year, Bob compromised and got her a pig. The entire house-
hold later joined Meredith in her hobby of training Seeing Eye dogs. A
large vegetable garden provided much of the family's fresh produce, and
within a few years nearly half the family food came from the animals and
the earth. Every lamb they owned was named Lambchop, lest the chil-
dren lose sight of its fate.
   Anne would always describe marriage to Bob as "complex." He was
completely lacking in conventional traits. For stretches at a time, in
fact, he was missing in action. He kept odd hours; for years his pattern
270   •   QYE6RPUNK
was to work half the night and sleep until eleven the next morning. He
believed firmly that if he was going to work hard, it should be at some-
thing he enjoyed doing. The children occasionally had trouble under-
standing that their father wasn't going to be like the fathers of their
friends, conventional workaday men who would spin through the
kitchen at 8:00 A. M., briefcase in hand, returning promptly at 6:00 P. M.
Bob's erratic schedule was dictated by the nature of his work.
    But once Bob finished a major project at work, he would spend several
weeks at home working on a domestic job that he found equally absorb-
ing. Bob had a natural inclination to integrate his practical skills with
his broader cultural knowledge. One of his more ambitious projects was
to design and build a sheep pen. Dissatisfied with the conventional
designs he found in home construction books, he turned to the February
leaf of a famed fifteenth-century illuminated manuscript, the Tres riches
heures du duc de Berry, a wintry scene of peasants working on a farm,
depicted in exacting miniature in blues and golds and whites. A center-
piece of the picture is a simple yet elegant sheep pen. The sheep pen on
this page of the fabled Book of Hours proved the most practical and
aesthetically pleasing design. Bob built its precise twentieth-century rep-
lica in rural New Jersey.
    Bob's good salary at Bell Labs enabled Anne to do what was important
to her rather than work just to supplement the family income. For the
first few years, that meant raising the children. Then she involved her-
self in local and state environmental work; eventually she became exec-
utive director of the Association of New Jersey Environmental
Commissions, a statewide umbrella organization for municipal environ-
mental commissions. For his part, Bob became chairman of the local
planning board.
    Sending the children to the best schools possible was all part of the
plan, too. Bob and Anne considered the quality of the local public
schools inadequate to the task of educating their children, so they sent
them all to private schools. Not only was it a financial burden, but for
Anne it meant she would have to drive them to school every day for
fourteen years. In order to pay the hefty school bills, Bob and Anne
remained frugal. The house was furnished with pieces both inherited 'and
discovered. Seldom was a new household appliance purchased. Washing
machines and other large appliances were mostly other people's discards
that were old but still functioning. Bob kept a stockpile of appliances in
various stages of disrepair in the barn, with at least one at the ready
should the one on duty fail.
                                                           RTH    ~    271
   Bob and Anne Morris provided their children with an idyllic, if some-
what eccentric life. Their range of options and their exposure to life in
general were far broader than those of the vast majority of children their
age. And the family was unusually close-knit. They played in orchestras
together, sang in choirs together and took regular trips to Manhattan.
And when the family went on vacations together, it wasn't to island
resorts, but to Iceland for a month, or to England for canal boating. Bob
Morris's work was on the cutting edge of a discipline that was defining
the future; yet when friends came to visit from more suburban commu-
nities, they felt as if they had stepped into a time warp. And to complete
the picture, the computer terminal resided in the basement next to an
enormous eighteenth-century beehive oven.
   Bob left the task of child rearing to Anne. She believed that children
should be exposed to a full range of experiences and should be able to
draw on that broad spectrum when choosing the direction of their adult
lives. Ben would eventually pursue a life outdoors, working as a tree
surgeon in Millington. Meredith would choose liberal arts and become a
researcher at the Library of Congress. And from an early age, Robert
seemed destined to follow his father into science.
   It was an admirable approach, but it also required a disciplined house-
hold. Anne instilled in her children a strict work ethic: each had morn-
ing chores to do outside, animals to feed, eggs to collect, wood to gather.
For the most part, the three children discharged their chores without
complaint. Some of the work was also a great deal of fun. Gathering
wood in winter, for instance, meant cutting a path up the icebound river
with shovels and skates, and carrying the wood back down on sleds.
Anne also made it clear that they were wholly responsible for their
assigned work around the house. If Robert neglected to feed the sheep
in the morning, he would return from school in the afternoon to a chorus
of bleats.
   Through the years, changes were made to the house that reflected the
Morris way of life. Bookshelves went up everywhere. In time, the family
library included six thousand volumes, their subject matter ranging from
theology to natural history to sailing and navigation. Every book had
been read by at least one member of the family. One day, Bob brought
home one of the original Enigma cryptographic machines. On one of
what had become regular visits to Fort Meade, Bob had simply walked
out the front door of the NSA, accompanied by the agency's deputy
director, with the machine stuffed into a brown paper bag. Eventually it
became yet another Morris household curio.
272   ..   CYE£RPllNK
   The children weren't given an allowance. Instead, they were paid for
work they did around the house outside of their normal obligations, such
as digging drainage ditches and building fences. Anne was always careful
not to pay them very much, in order to get them to see that they could
earn more money by working elsewhere. Other kids always had more
money than the Morris children, and later, when others had the use of
their parents' cars, Bob and Anne told their children that if they wanted
to drive, they had better figure out a way to buy their own cars. Robert
and Meredith accepted the arrangement perhaps more easily than Ben,
who groused mildly at the restriction.
   Like many children in rural settings, the Morris children grew up
without a gang of neighborhood kids to run with, a nearby shopping
mall or a video arcade. When the children were very small, the family
had no television set. But when it turned out that six-year-old Meredith
was a mass-culture "illiterate," as Anne described it years later, the
family bought a tiny black-and-white set specifically to view "Sesame
Street." An upgrade in size came only because all three children had
trouble seeing the screen at the same time. Still, television wasn't so
much prohibited as quietly discouraged. The black-and-white set with
poor reception competed for space in the living room with the computer
terminal, which had migrated upstairs. When Anne voiced complaints
about having a computer terminal in the middle of the living room, Bob
gently reminded her that he could have put it where some of his col-
leagues had theirs-in the bedroom.
   The young Morrises were early and voracious readers. Meredith was
already reading at age four. By the third grade, Robert had read T olkien's
Lord of the Rings trilogy and memorized its many poems. By age nine he
was devouring back issues of Scientific American, and by the time he was
in his early teens his reading list had expanded to include the classics,
history and copious science fiction.
   Robert's intelligence was especially apparent from an early age. As a
preschooler, he built working-scale models of cars out of whatever tools
he found lying around-paper clips, cardboard and file folders. Before
long, following his father's example, he was pulling electronic equipment
apart and piecing it back together.
   Anne saw that Robert sensed that he was different from others his
age. But he recognized only that he was different, not why he was
different. In fact, he once confided to his mother that he thought he was
"weird." She occasionally tried to inquire just enough to test whether he
understood that his abnormality lay in his intelligence. But even though
                                                           RTH   T    273

it was clear to his parents that Robert was more intelligent than his
schoolmates, he seemed only confused, and occasionally frustrated, by
the difference between him and his peers.
   Robert and Ben started out at the Country Day School in Far Hills.
Robert was easily bored and his performance suffered accordingly. When
Robert got to the fifth grade, Bob took matters into his own hands: he
went to the headmaster and suggested that Robert skip into sixth grade.
The headmaster refused, citing school policy. Bob's response was to keep
Robert home for four days. The headmaster relented, put Robert in sixth
grade, and his grades improved immediately. Nonetheless, unhappy with
the direction of the school's curriculum, Anne and Bob pulled the boys
out and enrolled them in the Peck School, twelves miles away in Mor-
   Following the Peck School's more traditionally rigorous curriculum,
Robert improved his scholastic performance dramatically. Still, he was
well beyond his classmates on most subjects. By the time he was in
seventh grade, Robert was reading science fiction at a clip of two or
three volumes a day. Bob and Ben were avid science fiction fans, too,
but Robert was the one who seldom went anywhere without a science
fiction novel tucked under his arm. When Anne went to parents' day at
the Peck School one day, she saw her son seated in the front row of his
math class, his nose buried in a science fiction book. When called upon,
Robert simply looked up from his book, recited the correct answer and
returned to his reading. It was clear to Anne that this wasn't cheek on
her son's part. It seemed a perfectly suitable arrangement between the
teacher and a student who could read his books and still stay a step ahead
of the class.
   High school meant yet another private school. Delbarton was an
exclusive boys' school, also in Morristown, run by Benedictine monks.
Delbarton was known for its excellent music department, and after Rob-
ert's third week there he came home one day and announced that he
planned to learn the violin. Once Robert started, Ben took up the viola,
Anne played the bassoon, Meredith started on the French horn and Bob
dabbled, starting out on the oboe, then turning to the cello. Anne and
Bob made music a family focal point. Each child was introduced to grand
opera at age ten, with a trip to New York. For years, Bob's annual
Christmas present to the family was Hdnsel und Gretel at the Metropoli-
    Ben and Meredith liked using the computer well enough. They logged
on mostly to play games. But of the three children, Robert was the one
274 •     eygERPUNK
to fasten onto computers most earnestly. When Bob stepped away from
the terminal it was only a matter of minutes before Robert logged on. In
front of the terminal was a cavernous and comfortable old armchair, its
back facing the rest of the living room. Whoever sat in the chair was
enveloped by it, and the young Robert nearly disappeared.
   Then there were the electronic friendships computers created. To give
the children a sense of what was possible with computers and communi-
cations, some of the parents gave their children their own accounts on
the computer at Bell Labs. Aside from some fairly strict ground rules
about behavior on the network, the kids were allowed, and even en-
couraged, to explore the world of computers firsthand. Ken Thompson's
son Corey was a regular on the network. At times, there were up to
twenty-five kids using the Bell Labs computers and communicating with
each other. Many of them, in fact, developed strong electronic friend-
ships before they ever met in person.
   Robert's was the first generation to grow up with ubiquitous computer
networks. Using the computer gave the fourteen-year-old Robert his first
taste of the power of instantaneous communications, and the social
equality that computers made possible. Tapping the computational
power of a machine ten miles away presented an irresistible lure. Robert
became a regular, making friends on-line and exchanging homemade
computer adventure games. Not only were they cleverly programmed,
but the kids' games also required a fairly sophisticated knowledge of data
communications. They were similar to early adventure games such as
Zork and Adventure. These games were really vast puzzle-solving exercises
played at a computer terminal. They were interactive, permitting the
player to explore by typing commands at the keyboard. The games re-
volved around treasure hunts and magic words. One of the teenagers
wrote a game called t4c (The Four Corners), complete with underground
passageways. The best thing about the game was its interactive, multi-
user nature. Players ran into each other while playing.
   Robert then wrote a game called Run-Me, an enhanced spinoff of t4c.
In t4c, characters could only talk to one another. Run-Me players could
also hug, kiss, hit and tickle. With Run-Me, Robert established himself
as the games master of his group.
   Not only were the teenagers learning about computers, but they were
learning the rules of the computer community. For some of them, the
Bell Labs computer was a telephone and television rolled into one,
fulfilling their social needs and their need for entertainment.
   One of Robert's best friends on the network was the unusually bright
                                                           RTH   ~    275
daughter of a Bell Labs scientist, one of the few girls on the network.
Robert set up some features of the Run-Me game specifically for her. For
example, the altar in the church would shimmer when her character
entered. One of her most impressive achievements was her own war-
drobe program, which told her what to wear each day. Her automated
decision was a function of the articles of clothing in her drawers and
closet. Each morning when she called up the program, it would tell her
which pants and shirts hadn't been worn recently and would select
several possible combinations for her. Though they lived just eight miles
apart, she and Robert carried on an electronic courtship for a year before
actually meeting.
   Only rarely did the children of the network overstep their bounds.
One day Bob arrived at work and stormed into an office where some
colleagues were sitting. He announced to the group in his trademark
summary manner that all the kids' accounts had to be taken away im-
mediately. Deciding there must be a story behind this sudden decision,
the others prodded him into telling them that one of the kids had been
operating as a superuser on the computer.
   "Well, then, just take away that kid's account," suggested one in the
   Bob shook his head.
   After more probing, Bob broke down and said it had been his own
   "How did he manage to get the root password?" someone asked.
   "He didn't."
   "Well how did he get in there?"
   It finally surfaced that Bob had absentmindedly walked away from the
terminal at a point where he had access to everything on the Bell Labs
computer, leaving Robert the run of the system. Robert had just walked
up to the machine and started using it.
   Robert was clearly interested in more than just playing games on the
computer. By the time he entered junior high, his father had introduced
him to UNIX and he was already finding holes in it. He was soon writing
his own UNIX "shell," a sophisticated program for carrying out user
commands. As soon as the UNIX source code was on line, Robert started
to study it with a special zeal. In his mid-teens, Robert was showing his
best friend, Doug McIlroy's son Peter, how it was possible to get super-
user privileges on one computer, then parlay those privileges into a tour
of various computers at the lab. Robert even modified a few files before
alerting his father's colleagues at Bell Labs to the security hole he had
276    &   CY£fRPUNK
 found. If researchers at Bell Labs were amused, or grateful to a teenager
 for pointing out weaknesses in their own handiwork, they didn't let on.
 He was told to stop and that was that.
    Even as a ninth grader, Robert was more his father's colleague than
 his disciple. Bob was careful never to sit Robert down and say, "Here,
 I'm going to give you a lecture." For weeks at a time, father and son
could be steeped in an ongoing discussion of a technical problem. A
conversation could last for hours or for days, and while they were talking
about whatever it was-it could be a discussion of a security flaw in
UNIX or of building an electronic circuit together-they remained
oblivious to the rest of the family. As the one with more knowledge to
impart, Bob could occasionally be hard on Robert, and extremely chal-
lenging. In overhearing some of these exchanges, Anne could tell from
the tone of Bob's questions and Robert's quiet responses that Bob was
pushing Robert. But that was Bob's way. He was accustomed to quizzing
everyone anyway, asking his questions in short, clipped phrases that
might seem abrupt and impatient to an outsider but were unthreatening,
at times even playful, to those who knew him. Mostly, the part of their
relationship that involved computers centered on theoretical questions.
Yet Bob always encouraged Robert to refine his practical programming
    To outsiders, it seemed that Bob might even be encouraging Robert
to break into computers. In 1982, Gina Kolata, a writer for Science
magazine working on a story about computer crime for Smithsonian mag-
azine, went to interview Bob Morris about security. He told her about
tiger teams and smugly predicted that after a few minutes of looking in
her wallet he would know enough about her to guess her computer
password. When she asked him if he knew of any young hackers she
could interview, he suggested she speak to his son on an anonymous
basis and invited her to the house. The sixteen-year-old Robert struck
Kolata as unusually shy, almost intimidated by the reporter. Anne Morris
supervised the interview, and while Anne appeared to be protective of
Robert, Kolata got the impression that father and son were a duo, egging
each other on. Young Robert told her that yes, he had read private
computer mail and had broken into computers that were linked together
in networks. "I never told myself that there was nothing wrong with
what I was doing," he told her. But, he said, he continued to do it for
its challenge and excitement. In an ironic coda, that year Robert placed
eleventh in a state high school physics competition. His prize was a
                                                          RTH •     277
subscription to Smithsonian, and the first issue he received was the one
with Kolata's article in it.
   The Smithsonian article came out at a time when awareness of com-
puter security was growing gradually. By the 1980s, hundreds and then
thousands of personal computers were linked together via networks and
one user, one machine was the new computing philosophy. But then
another idea began to form: why not create a computing system that
wasn't found in a single computer but was spread throughout a network
of computers? Could the system itself be so intelligent that when a
particular computing task needed to be done, it could be distributed
automatically to the geographic point that had the best available re-
sources? A computer revolution that is still only partially realized was
under way.

When Robert was growing up, networks were for the most part private
laboratories used by computer scientists who were experimenting with
new ways of using computers. The things he observed his father do, and
the research he heard and learned about, served only to reinforce that
perception. But the world was changing rapidly, and the most powerful
instrument of change over the next decade was the Arpanet. Its name
derives from ARPA, the Pentagon's Advanced Research Projects
Agency, which was renamed Defense Advanced Research Projects
Agency during the 1970s. This agency was run by scientists rather than
soldiers and it was charged with exploring high-risk ideas. For American
computer science in university and corporate research centers, DARPA
created an entirely new world. During the 1960s and the 1970s DARPA
funding was crucial to the most significant advances in computer science.
Personal computers, networks, artificial intelligence and voice recogni-
tion all in one way or another were the fruit of DARPA-funded experi-
   The Arpanet network in turn was the brainchild of a community of
computer scientists who, during the late 1960s, were among the first to
envision permitting scientists and engineers to share computers and ex-
pensive resources instantly and easily no matter where they were. That
a computer network could serve as both a means for instantaneous com-
munication among researchers and an experimental communications
laboratory was a revolutionary notion.
   At the beginning of the 1960s, Paul Baran, a scientist at the Rand
278   •    CYG6RPUNK
Corporation, was searching for ways to make telephone networks more
reliable in the event of nuclear war. Out of his research came the idea
of breaking digitized messages up into "packets of numbers." Each packet
would carry an electronic address, and each could be routed by the most
efficient route. Packet switching dramatically lowered the cost of data
communications, making low-cost computer networks possible. The no-
tion of actually linking computers to share these networks came from
J. C. R. Licklider, a psychologist who went on to become the first
director of DARPA's information processing and technology office.
   The original Arpanet was built around separate message-passing com-
puters known as Interface Message Processors, or IMPs, which were the
backbone of the network. Later small computers, known as TIPs (ter-
minal interface processors), which handled connections with slow dial-
up terminals, were added. Each of the IMPs would be connected to
another IMP on a leased phone line and was capable of sending and
receiving at what then seemed like an extremely high speed. Contem-
porary networks routinely carry data at twenty times that speed and
network designers are working to build a "national data highway" that
would increase the speed of today's fastest commercial links by up to
seven hundred times.
   The first Arpanet node was installed at the University of California at
Los Angeles in late 1969 and the next three nodes were placed at the
University of California at Santa Barbara; Stanford Research Institute, a
California think tank; and the University of Utah. The next year three
more nodes were added on the East Coast: at the Massachusetts Institute
of Technology; Bolt, Beranek and Newman, the Cambridge, Massachu-
setts, think tank that designed the Arpanet network; and Harvard Uni-
   Other research projects had linked computers experimentally, but the
Arpanet was to grow into the first nationwide computer network. The
Arpanet connected research centers, military sites and universities. Ini-
tially, virtually all of the computers on it were identical (almost all were
Digital PDP-lOs), and virtually all of the people at those sites were
government-funded computer science researchers. By 1973, the Arpanet
consisted of twenty-five machines.
   To be at a site connected to the Arpanet was to be among an elite.
So coveted was a connection to the network that academic job offers
were sometimes accepted or turned down on the basis of promised access
to the network. For some computer scientists, access to the network was
a requirement for doing their jobs. For these scientists, going to a uni-
                                                          RTH ...    279

versity without a network connection would have been like a research
microbiologist accepting a job at a school with no microscopes.
   In its early days and even into its middle years, the Arpanet had the
feel of a private club. "Are you on the net?" was a question heard among
the most elite computer scientists. Getting into the club wasn't easy, but
once you were in, you were given free rein. There was no concept
whatsoever of security. Anyone anywhere could read a file anywhere in
the network. At Carnegie-Mellon University, for example, every file on
every computer, save th~se that were explicitly protected, was available
for examination or copying by anybody on the Arpanet. Graduate stu-
dents at those places spent many happy hours cruising around through
the files on outside computers to see if there was anything worth reading.
    In 1975 the operation of the network was turned over to the Defense
Communications Agency, a Pentagon organization that is responsible
for military voice and data traffic. By then, there were more than sixty
sites on the network, and the amount of data traffic carried by Arpanet
had increased dramatically.
    Part of the clubbishness that defined the early Arpanet grew up around
the technical limitations of the network. Through the 1970s, the
Arpanet could only support 256 computers. But by 1982, a new
network-addressing scheme was developed to allow for exponential
growth of the network. By the mid-1980s, the Arpanet had become the
seed for a complex of networks called the Internet, which touched down
in more than fifty countries. It was no longer just an engineering exper-
iment. Computer centers used the network for technical support, re-
searchers sent papers back and forth in an instant and software of all
kinds flowed around the globe. Commercial enterprises adopted the
technology of the network to create their own private versions of the
network based on the same set of communication protocols. These cor-
porations also used the Internet itself to stay in contact with operations
spread around the world. The Internet in turn was connected through
gateway computers to hundreds or thousands of other networks. Some
began to speak of an even broader concept of interconnected net-
works. They referred to it as the Matrix, taking the name from the all-
encompassing computer network in William Gibson's Neuromancer.
    It was with some indignation that the Arpanet pioneers watched their
 network be appropriated by society at large. Whereas in the early days it
 could cost as much as $250,000 a year to maintain a connection to the
 network, the base of support had since grown to the point where the
 cost was minimal. Universities and corporate research centers still com-
280   &    CygERPUNK
posed most of the links, but by 1988 the function of the network had
broadened considerably. The Arpanet was supposed to be primarily a
computing laboratory, but mostly it was used for sending electronic mail
about every topic imaginable. The network pioneers were puzzled and
not a little miffed to see newspaper reporters, of all people, with accounts
on machines linked to the net. That was nearly as preposterous as the
notion of walking into the campus chemistry lab and seeing a bunch of
reporters wielding Bunsen burners and pipettes.
   Gradually, over a period of years the original Arpanet network links
were supplanted by faster data paths, and by 1990 the Arpanet ceased to
exist as a separate entity, having been absorbed into the Internet. By
one current estimate, several hundred thousand different computers are
currently on the network, from supercomputers to personal computers.
The best guess is that there are more than two million Internet users.
This data highway already carries the work of scientists, students, sol-
diers and businessmen, and many now argue that connecting it to mil-
lions of American homes and businesses will revolutionize the country
with new business, educational and entertainment services.
   But at first, the reason for the existence of networks was to carry out
experiments that explored the reach and power of the networks them-
selves. In 1971, Bob Thomas, a scientist at Bolt, Beranek and Newman,
was working on distributed computing software. His group designed an
air traffic control simulation that was intended to model different airports
on different computers. The idea was to be able to move control of an
airplane from one computer to another and tell all the other computers
so that they would know the location of a particular aircraft had been
changed. To do this, Thomas wrote a clever program whose mission was
to crawl through the network and pop up on each screen, leaving the
message, "I'm creeper! Catch me if you can!" Some time later, as word
of the program grew within the early network community, other hackers
wrote similar programs-some of which multiplied as they worked their
way around the net (Creeper didn't reproduce itself, it simply moved),
and others of which included "reaper" programs that sought out and
destroyed creepers. Writing such programs became a minor fad for a few
months and then died out.
    In the early 1980s, two computer researchers at Xerox's Palo Alto
Research Center started experimenting with programs they called
"worms" that were able to run on many computers in a local-area net-
work. (The Arpanet was a wide-area network, connecting computers
over long distances.) The term worm was taken from the book The
                                                         RTH   T   281

Shockwave Rider, a science-fiction classic written by John Brunner in
1975. It describes an authoritarian government that exercises power
through an omnipotent computer network until a rebel programmer
infests the network with a program called a "tapeworm." In order to kill
the worm, the government has to turn off the network, losing its power
in the process.
   Brunner became a cult figure, as the book swept through the world-
wide community of science fiction readers. It had a strong influence on
an emerging American computer underground-a loose affiliation of
phone phreaks and computer hackers in places like Silicon Valley and
Cambridge who appeared simultaneously with the development of the
personal computer. John Shoch and Jon Hepp, the Xerox researchers,
were looking for a way to make shared computing power more widely
and easily accessible over a local area network. They came up with five
or six useful worms. One was called a "town crier worm." Its job was to
travel through the network posting announcements as it went. Another
was a "diagnostic worm." It was intended to hop from machine to ma-
chine, constantly checking to see if anything was amiss. Certainly the
most dramatic distributed program the two conceived was the "vampire
worm." Such a program, they suggested, would take advantage of the
almost limitless free processing power in a network of computer worksta-
tions. After all, these machines spent many idle hours that could be
harnessed for useful work. The Xerox vampire worm automatically
turned itself on at night when people had gone home, setting to work
on complex problems that required vast amounts of computing power.
In the morning, when the computers' human owners returned to reclaim
their machines, the vampire program would temporarily store the partial
solution computed so far and shrink back to wait for the next evening.
   But early on, Shoch and Hepp were also to learn of the potential
dangers of worms. One night a malfunctioning program went out of
control on a local area network at the Palo Alto Research Center. In
the morning, when scientists arrived, they found that computers
throughout the building had crashed. They began to restart their sys-
tems, but soon found that each time they attempted to start a machine
the defective worm caused it to crash again immediately. The problem
was that many computers were behind locked doors and couldn't be
reached. Finally they wrote a "vaccine" program that traveled through
the network and electronically inoculated each computer in the network
against the worm.
   The Arpanet was also a resource to Bell Laboratories scientist Ken
282   ~   CYGERPUNK
Thompson, who used it for a computer security experiment. In the late
1970s, when Thompson was working on a paper about breaking pass-
word security, he used several network sites, such as Harvard, MIT,
Carnegie-Mellon and Berkeley, on which password files were publicly
accessible or on which he had accounts with access to these files. His
password-cracking program was successful and he discovered that he had
inadvertently captured passwords used by some of the Arpanet's key
administrators, people with accounts on many machines throughout the
network. He tried the passwords and discovered they worked. In the
hands of those whom Thompson and Bob Morris thought of as network
"bad guys," such a security flaw was dangerous. So Thompson sent mail
to the people who owned the passwords to tell them about the problem.

Robert leapfrogged entirely the process of learning computers in school.
The Delbarton School had early Apple computers, but from age twelve
Robert had access to a machine ten times more powerful. While the
school was handing out computer achievement awards to other students
for mastering the Apples, Robert was already writing complicated pro-
grams and technical papers.
    Yet few of Robert's friends and teachers at Delbarton even suspected
that the diffident sophomore had such a level of expertise in computers.
Robert had launched his computer career entirely from home. During
Robert's senior year, Bob's old friend Fred Grampp hired Robert part-
time at Bell Labs. Robert at sixteen behaved like any of the dozens of .
college students who took part-time jobs and internships at the Labs.
Unlike his father, he was inordinately quiet, but he shared his father's
tremendous curiosity about the world around him.
    Robert had already made something of a name for himself at Bell Labs
with his earlier unsanctioned tours of Bell Labs computers. But he was a
hard worker. His project there was his own idea: to write a more secure
and efficient implementation of UUCP, the program used for copying
files from UNIX machine to UNIX machine. The challenge was to write
a UUCP implementation that could cope with the volume and variety
of traffic that had evolved on the network over the years. It's not every
high school student who can redesign a major piece of software. Despite
a few problems, Robert's program was so good that it became the model
for UUCP that Bell Labs eventually adopted. He even produced a tech-
nical paper on results of his work, titled "Another Try at UUCP."
    Yet he wasn't single-minded in his devotion to computers. Robert
                                                           RTH ,. 283
distinguished himself early in other ways at Delbarton. He swam on the
school team and sang with the chorus. Yet he remained shy and, as far
as his parents could tell, still unaware of his intellectual gifts, so Anne
took it upon herself to have a talk with the headmaster at Delbarton.
She explained that she thought Robert might benefit from a boost from
his superiors at the school. Apparently in agreement with this concerned
parent, the headmaster went out of his way to praise Robert in the
presence of other students. On the day the school received the results of
that year's SAT exams, the headmaster greeted Robert in the front hall
of the school and, in front of a dozen other students, told Robert that
his scores-a perfect 800 in verbal and a 790 in math-were the highest
in the school's history. From that point on, Robert's self-esteem seemed
to soar. For college, he set his sights on Harvard. Not only had his father
gone there, but for several generations back the Burrs and the Farlows
on Anne's side of the family had as well. Robert applied for early accep-
tance and got in.
   When Robert entered Harvard in the fall of 1983, he was still shy and
socially awkward. But he knew of one place he could go where he would
have a good chance at fitting in quickly: the Aiken Computation Labo-
ratory. Most college campuses have a research computer center, distinct
from where the university's own central data processing is performed. At
Harvard it's Aiken, which takes care of the computing needs of the
university's Division of Applied Sciences. The central computer center,
with operators who don't need to know much more than how to feed the
printer, is across the campus at the Office of Information Technology.
Aiken is a little faster and looser with its computers, and hence a more
interesting place to work. When Robert arrived at Harvard, there was
in fact no computer science department per se at the school. Instead,
there was a computer science faculty, a group of seventeen faculty mem-
bers within Applied Sciences. Serious computer science students at Har-
vard gravitated, more often than not, toward Aiken, where computer
science faculty spent their time.
   An all-brick monument to an architectural aesthetic grounded in
common sense, Aiken stands diagonally across from the magnificent
law school building on the Law School Quadrangle. Inside the Aiken
lobby, spanning an entire wall, stands Howard Aiken's fifty-one-foot-
long, eight-foot-tall legacy to modern computing-the Mark I Auto-
matic Sequence Controlled Calculator. During the thirties, the mathe-
matics professor had a dream of building a large-scale calculator, a
switchboard-mounted device that would do arithmetical operations
284   •   CYJg6RPUNK
without the intervention of an operator. In 1944, in collaboration with
IBM, Aiken completed the Mark I, at a cost of $250,000. It was the
world's first large-scale electric calculator. A typical problem that would
have taken a team of four experts three weeks to solve occupied the
machine for only nineteen hours. By the late 1980s, a problem that
would have really challenged the Mark I could be done in a second or
two on a $40 programmable hand-held calculator. Still, the Mark I was
a revolutionary development in its time and Aiken's place in the history
of technology was duly secured inside the building named for him.
   Across from the Mark I is a glass-enclosed room of terminals and
workstations, a place where students and Aiken staffers work. In 1983,
the hard-core computer people at Aiken who didn't have their own
offices spent most of their waking hours in the room and others drifted
in and out. In the 1950s, Bob Morris had also spent time at the com-
putation lab, helping to build the Mark IV, the fourth generation of
Aiken's calculator.
   Shortly after arriving at Harvard, the younger Morris walked into the
Aiken administrator's office and asked for an account on the lab's com-
puter. Eleanor Sacks, the Aiken administrator, patiently explained that
freshmen weren't given accounts at Aiken, that Aiken was the exclusive
province of faculty and more advanced students. She gently told him to
join the other freshmen a few doors down at the Science Center. But
Robert didn't especially want to join the masses in the basement of the
Science Center, a sea of computer terminals and personal computers
that resembled a word-processing pool more than a computer science
lab. Not only was Aiken a more civilized place to sit and program, but
there were more computer resources available there. But instead of trying
to argue about it, Robert wandered back out of her office. A few days
later, he took care of the problem himself by turning the Aiken VAX
into a single-user machine, creating an account for himself and then
returning the VAX to multiuser status. His login, which he had used
since the days on the Bell Labs computer, was rtm. Shortly thereafter, a
faculty member who knew Bob Morris saw to it that Robert received a
legitimate account.
   Nick Horton, the Aiken manager, knew little about UNIX, and be-
fore long Robert was a permanent fixture there. He wasn't one to learn
a little about a lot of things. Like his father, Robert learned a lot about
a lot of things. He could handle hardware emergencies as well as software
problems. Once his expertise became know at Aiken, his services were
in great demand.
                                  q         1')1;
                                                                   I. ,

   The number of computer science majors at Harvard'Is' relatively small
                                                                             ~     285
                                                                            c- -_,'. , . .'
                                                                               "- ,

-each year, about thirty students get their undergraduate degrees in
computer science-but they pride themselves on being more well
rounded than their counterparts down the street at MIT. Robert may
have had an unusual aptitude, but he was certainly no freak. Everyone
around him had a dozen other interests. The spirit of a place like Aiken
was personified by the students who worked there as part-time staff
programmers. One professor who needed a student to do some program-
ming for him was amazed by what he saw: there the student sat, and as
he waited for output from the computer he appeared to be reading two
books at once, one in French and one in German, just to keep himself
    Perhaps what made Robert stand out most was his impressive knowl-
edge of UNIX. He could sit for hours just reading UNIX manuals. The
UNIX documentation had grown to consist of more than two thousand
pages, and on each page was a new set of minutiae concerning the
workings of the operating system. Most people kept the manuals on hand
purely for reference, but Robert appeared to enjoy simply reading them.
Before long, he was regarded as the most knowledgeable UNIX techni-
cian on campus. And he had more than a theoretical understanding. He
had a tremendous capacity for remembering details. If someone had a
question about UNIX, it was often easier just to ask Robert than to look
it up. While some people were in awe of his capacity for minutiae, others
wondered if that was all he thought about. He was, in computer par-
lance, a systems hacker through and through.
    Like his father, Robert was especially talented at putting together a
quick program that would solve a pressing problem. On one occasion,
when a professor got a new computer, he needed some software written
for it. He asked Robert if he could do it, and without needing to pick up
a pen to sketch an outline of the program first, Robert sat down at the
computer and just typed. He was finished within a few hours, and the
program, while not the most refined of code, certainly did the job.
    By the end of his freshman year, Robert was spending almost all of his
time at Aiken. He was doing odd programming jobs and technical sup-
port, all of it gratis, and he had become indispensable. When others
asked him why he didn't apply for a job at Aiken so that he could at
least get paid, Robert replied that his father had told him not to take a
job right away so that he could concentrate on his schoolwork. This
way, he could keep to his word but still have fun at Aiken. And since
he wasn't on the payroll, he could work on projects of his own choosing.
286   A   CygtRPUN~

   Robert spent the summer of his freshman year living at home in
Millington and working at Bell Labs. A second technical paper came
from the summer's work, calling attention to a security hole in Berkeley
UNIX. By this time, Robert's expertise was so well appreciated at Har-
vard that a special data line was set up between Harvard and the research
machine at Bell Labs so that Robert could perform remote diagnostics
and maintenance from New Jersey during the summer. Robert's notes
were always terse and they almost always fixed the problem. Even when
he wasn't asked directly for help, he offered it anyway. For example,
while browsing around in the Aiken system he noticed that some hard-
ware had been installed improperly. A message showed up in Nick Hor-
ton's mailbox one day: "Try swapping the two boards in Positions A and
B." It was from Robert. The problem was fixed.
   When it came time to declare his major, Robert started out in math,
but soon switched to computer science. In the first semester of his soph-
omore year at Harvard he was hired as a staff programmer. He wasn't
actually doing much more than he had done when he worked for no pay,
but now he spent even more time at Aiken, to the exclusion of nearly
all else, including his coursework. His academic performance fell to the
point where the college ordered him to take 1985 off. Rather than tell
his parents right away that he was in trouble with Harvard, Robert lined
up a full-time job as a programmer at Convex, a hot new computer
company in Dallas, and presented the news of academic probation to his
parents as a problem for which he already had the solution. Once again,
while in Dallas, Robert was a remote diagnostician and consultant for
Aiken. When Nick Horton asked him a technical question, he would
send back not just a paragraph of explanation, but a lengthy example
and tutorial.
   While at Convex, Robert helped run the company's time-sharing
systems, and wrote software that would analyze and simulate the perfor-
mance of Convex hardware. He was also flown out to customer sites as
the company's troubleshooting whiz kid. It was a pretty lonely time for
someone so young and so shy. When people gave him projects to do, he
invariably finished them early and waited for something else to do.
Outside of work he learned rock climbing and scuba diving, and he
played Photon, a high-tech version of Capture the Flag. When he went
to visit his friends at Aiken, he said he was looking forward to coming
back. One of the conditions of Robert's return to Harvard in early 1986
was that he not work again on the Aiken staff, at least not right away.
                                                          R1H    ...   287

So although he continued to spend time there, he was no longer a formal
   The Aiken staff came and went, but in early 1986 the esprit de corps
was especially strong and the group especially diverse. There were Nick
Horton, a psychology major and social activist; Andy Sudduth, a tall,
red-haired Olympic rower; Steve Kaufer, captain of Harvard's fencing
team and in the midst of starting a software company; Karen Beausey,
on her way to law school; and David Hendler, one of Robert's closest
friends, a linguist and history of science major. It was a group that
somehow clicked together particularly well, doing things outside the lab,
taking trips to museums, going on ski vacations and eating dinner to-
gether. They all knew a lot about computers but they also knew a lot
about other things. David was a gourmet cook who found some of his
best recipes on the USENET Cookbook, a network recipe exchange.
Nick Horton was also an avid subscriber to the recipe exchange and, as
a Christmas present one year, Nick printed out the archives of the
cookbook, bound it and sent it to everyone at the lab. Around his friends
at Aiken, Robert's shyness melted away. Engaged in a technical discus-
sion, Robert was quite animated. In his element, he could be positively
    Working at Aiken meant keeping irregular hours and adjusting to a
hectic and occasionally demanding environment. Aiken staffers did so
uncomplainingly. Dozens of computer start-ups were eager to use places
like Aiken as a test-bed for their new hardware and software. There was
also a push among the Aiken staff and some of the faculty to procure
equipment that was new and interesting, if not entirely reliable. The
popular sentiment was that if it didn't work, then it could be made to
work, so Aiken staffers spent a lot of their time trying to fix things. If
there was a problem in an area that no one knew anything about,
someone would volunteer to become an expert in the course of an eve-
    Robert always managed to find time for some harmless pranks. Ex-
ploiting people's tendency to type "mial" by mistake when asking for
their electronic mail, Robert wrote a program so that each time someone
made the error, instead of mail a Dungeons & Dragons-like adventure
game appeared on the screen. He excluded senior faculty members from
the prank; when they made the typing error, the system simply said it
did not recognize the command. The "rnial" prank was clever and harm-
less, but after a while people became annoyed with the game and Robert
288 •     CygfRPUNK
was told to remove it from the system. Then, as an April Fool's joke,
Robert wrote a program that made it appear to anyone who logged in
that Harvard had gone back in time ten years and was using a long-
obsolete operating system on equally obsolete hardware. Whenever Rob-
ert was asked if he was the source of a prank, he would look down with
a shy smirk.
    Then there was the Oracle. Anyone logging on to the computer was
told to ask any question of the Oracle. But before the question could be
asked, a question from the Oracle had to be answered first. Some ques-
tions tested one's knowledge of technical trivia; others were just silly
("Why do we have 8:30 a.rn, classes?"). It took everyone a while to
figure out that it wasn't the computer itself generating questions but
others using the system. Whenever someone logged on, he or she ful-
filled the computer's request to ask a question, which was sent on to the
next person to log on. That person's answer was mailed to the user who
posed the question, and so on. The Oracle's cleverness lay in making it
look as if the computer were doing everything, when in reality people
were both asking and answering questions and the computer was just
mailing messages back and forth.
    Those who knew Robert well were aware that he had a special interest
in computer security. He wasn't one to boast about his computer security
expertise, but it helped to explain his preoccupation with studying UNIX
line by line. Careful examination of the code itself was the best way to
unearth security flaws. But he didn't flaunt his detailed knowledge of the
operating system, and he certainly didn't announce plans to follow a
career in security. Nonetheless, one of his favorite refrains was how
many holes there were in Berkeley UNIX.
    At the same time, Robert had a sense for where to draw the line when
probing security. Once, he and David Hendler were discussing a partic-
ular way of logging in to machines around the network. Taken with the
notion, David considered logging in to Brian Reid's computer at Digital
Equipment's research laboratory in Palo Alto, but Robert advised him
strongly against doing that. David knew Reid for his network recipe
exchange, but Robert knew him to be an especially conscientious net-
work sentinel who would notice something amiss immediately if some-
one logged on to his computer. Robert made a practice of breaking into
only the computers of people he knew wouldn't mind.
                                                          RTH    T   289

Paul Graham, a hyperactive and pink-cheeked computer science gradu-
ate student, had always considered himself more intelligent than vir-
tually everyone else. In his twenty-one years he hadn't seen much
evidence to indicate otherwise. Then Paul heard from a friend about
someone who, the friend said, was on another plane altogether.
   One day at an Aiken party shortly after Robert's return from Dallas,
someone pointed out the brilliant young Morris.
   Paul went up to him. "Hey, aren't you Robert Morris?"
   The young man lowered his head and grinned, then pointed across
the room at someone else and said, "No, that's him."
   It wasn't until several days later that Paul learned that he had been
duped, if only because the same person who had disowned the name
Robert Morris was always at Aiken Lab, always working until at least
3:00 A.M. and always working on something that seemed complex.
When Paul started spending time at Aiken, Robert Morris was writing a
program called a ray tracer for a graduate-level computer graphics course.
A ray tracer produces images of three-dimensional scenes. Given a model
of the scene in geometric shapes, the ray tracer follows the path of
individual rays of light from their source as they bounce off objects in
the scene and eventually enter an observer's eye. The most impressive
thing to Paul was that even though the course had ended and Robert
had already received his grade, he was still working on perfecting the
program for the sheer intellectual challenge of it. In fact, Robert's pro-
gram was so interesting that it caught the interest of his roommate, Greg
Kuperberg, a math student, who helped him with some of the more
complex mathematics needed for constructing solid shapes.
   Ray tracing requires vast numbers of computing cycles and Robert
took them wherever he could find them. But he was scrupulous about
not affecting the other users on the system. So he wrote a program a bit
like John Shoch's vampire worm for making use of perfectly good cycles
on computers around the lab that would have otherwise gone to waste.
When a user sat down at a workstation and began to type, the computer
stopped doing Robert's work and went to work for its rightful owner.
Robert extended his clever redistribution of cycle wealth so that other
students could use it, too.
   Paul came to call him by his login: rtm. There was no limit, it seemed,
to rtrn's knowledge. He not only knew about the workings of the VAX,
but he also knew about graphics, and he had read all of the UNIX source
code. In Paul's view, rtm was no single-minded geek. This guy had read
290   ~   CYGERPUNK
all of the Norse Sagas. And he liked to go to the opera, of all things. He
was nothing like Paul's suburban contemporaries, who had grown up
addicted to video games, television and junk food. When Paul left Mon-
roeville, a Pittsburgh suburb known for its immense shopping mall (it
served as the set for the cult film Dawn of the Dead), his years in front of
the television haunted him through college. When he got to college, he
tried to make up for lost time by going cold turkey. One glance at a
television screen could well turn into a week-long binge. But here was
someone with no interest in that electronic drug, nor in video games.
Paul felt that, compared ro Robert's, his childhood had been wasted. He
envied Robert's upbringing: the rustic environment, the private-school
education, the adventurous vacations, the prominent father. Paul was
in awe. He was a reluctant computer scientist who would rather have
studied painting and looked upon others in the graduate program as
irredeemably narrow-minded digit-heads. Meeting rtm was the best thing
that had happened to him all year.
    Paul knew that he and rtm were going to be good friends when he
discovered one thing they had in common: neither of them liked to sit
in classrooms, and if a course failed to challenge them, both were in-
clined to skip class frequently. One day, Paul was sitting outside on the
steps of Aiken reading a book when he was supposed to be sitting inside
absorbing a lecture on artificial intelligence. When rtm walked up to
him and took a look at Paul's book, the historian Jacob Burkhardt's
history of the Italian Renaissance, he smiled. Both agreed that reading
Burkhardt was a far better way to spend one's time.
    Others at Aiken considered Paul too unrestrained, but Robert was
more willing to be his friend. He once took Paul along to a relative's
house on an island off the Maine coast. As they were in a boat headed
for the barren island, which had no electricity or telephones, Robert
said, "You're going to like this. From now on, things are done right." It
was Robert's appreciation for things that had nothing to do with material
possessions that impressed Paul.
    But Paul was concerned that his friend rtm didn't have girlfriends. "If
you like someone, rtm, you've got to say something to her," Paul would
insist. "You can't expect her to read your mind."
    "But what are shy people supposed to do?" Robert would reply.
    More than once, Robert and his friends got the itch to make a killing
from their specialized knowledge. After completing their much-praised
ray tracer, Robert and Kuperberg gave brief thought to launching a
                                                           RTH    T   291

computer graphics firm. And with David Hendler, Robert mulled over
the idea of a computerized method for predicting the commodities mar-
ket. Robert hatched his most farfetched get-rich plan with Paul Graham,
when the two decided they could make a bundle predicting the horse
races at Suffolk Downs. Paul kept copies of the Racing Form locked in
his desk drawer and the two spent hours entering reams of data about
past races into the computer. But after two depressing afternoons of
mingling with the crowds of desperate middle-aged men as they walked
from the subway station to the racetrack, they decided that it wasn't
worth the effort.
   The summer after his junior year in 1987, Robert worked at Digital
for the second year in a row, this time in Palo Alto. The previous
summer he had spent at Digital's engineering facility in Nashua, New
Hampshire-the same facility Kevin Mitnick would later break into
electronically-working on routine programming tasks, which he found
only moderately interesting. But the Palo Alto summer was wonderful.
While there he worked on graphics programs and programming lan-
guages, trying things that had never been done. The work was extremely
challenging and Robert thrived.
   The Morris family, in the meantime, was uprooting itself from New
Jersey and Bell Labs after twenty-six years. Bob had gotten frustrated at
the labs. He had been waiting for months to be appointed to a new
position that would oversee the creation of a secure version of UNIX, a
version with no security flaws. The job got stalled in the bureaucracy,
and while Bob's patience was wearing thin, the National Security
Agency came to him with an offer he couldn't refuse: to be chief scientist
at the National Computer Security Center, the unclassified component
of the NSA. The center had been established to improve the security of
computers within the military services but was later given a broader
mandate that encompassed establishing computer security standards in
the commercial world as well. The job was particularly attractive to Bob
because, while most of his work revolved around the more public center,
there was a classified aspect as well. Part of the time he worked in the
arcane intelligence-gathering world of the NSA.
   Bob and Anne sold the old house in Millington and moved to Arnold,
a small Maryland suburb along the banks of the Severn River. Bob had
crossed the line from theoretical research into a real game with real
players. Anne was sad to give up her job as director of the Association
of New Jersey Environmental Commissions, but she knew this was pre-
292   •   CYJgERPUNK
cisely the career boost Bob wanted. She eventually took a job with an
environmental group in Washington, which required a long commute
each day.
   Meanwhile, Robert's senior year was another period of both intense
work and good fun at Aiken. Schoolwork was shunted aside yet again-
one geometry course that Robert found excruciatingly dull he scarcely
attended at all. Despite a lot of cramming for the final, he failed the
course. He spent little time in his room at Dunster House (the Harvard
dormitory where his father had also once lived) preferring to sleep on
the couch at the group house where David Hendler was living. Many
evenings were spent cooking elaborate dinners and baking cookies to
send to out-of-town friends. During his spring break, at his father's
suggestion, Robert gave a talk at Bob's division of the NSA on every-
thing he knew about UNIX security. The following day, he repeated the
lecture to a group at the Naval Research Laboratory.

When Robert applied to graduate school in computer science, Stanford
was at the top of his list, followed by Cornell and Harvard. Stanford has
the most rigorous program; while Harvard fosters a more nurturing at-
mosphere for its students, it isn't uncommon for first-year graduate stu-
dents at Stanford and Carnegie-Mellon to fail their qualifying exams.
Stanford's program is also by far the most difficult to get into. Of 1,000
applicants each year, the graduate program admits just 30. Cornell,
which ranks among the nation's top ten graduate programs, is also very
difficult to get into. Of 550 or so applicants, the school admits just 40
into each entering class.
   Robert gathered letters of recommendation from some of the most
respected figures in computer science. Doug McIlroy from Bell Labs
wrote one. And Mark Manasse, for whom Robert had worked at Digital's
Palo Alto Research Center, wrote an effusive letter. "I fully believe that
Robert will succeed at almost anything he undertakes," Manasse wrote.
Nonetheless, Stanford rejected him, partly because of his spotty aca-
demic record, possibly because his score on the math section of the
standard Graduate Record Exam, though high, wasn't a standout in the
fiercely competitive Stanford applicant pool.
   But both Harvard and Cornell accepted him. His thesis adviser rec-
ommended against his staying at Harvard for graduate school. Perhaps it
was time for a change, and Cornell was a renowned center of computer
science theory. If Robert was to be faulted for anything, it was his
                                                           RTH    T   293

tendency to allow himself to be seduced by the machines themselves, at
the expense of a theoretical understanding. Robert's father, on the other
hand, had such a strong mathematical foundation that he would instinc-
tively bring mathematics to bear on problems that didn't appear mathe-
matical at first. Cornell would be a perfect place for Robert to gain a
better theoretical foothold on computer science. So Robert decided to
go to Cornell.
   He spent his last summer in Cambridge working at a plum job. At the
recommendation of Jamie Frankel, the same adjunct professor who had
recommended Robert for one of the summer jobs at Digital, Robert spent
the summer of 1988 at Thinking Machines Corporation in Cambridge.
One of the most interesting companies to work for, Thinking Machines
had developed a supercomputer based on "massive parallelism," applying
thousands of small processors-rather than one or a handful of processors
running at tremendous speed-to divide up the burden of particularly
numerically intensive tasks. The "Connection Machine"was being used
for such applications as picking out ground structures from satellite pho-
tographs, predicting the behavior of molecules and making three-dimen-
sional maps. The company was doing so many interesting projects that
any job there promised to be fun.
   Robert's principal project at Thinking Machines was to refine one of
the sophisticated languages that took special advantage of the Connec-
tion Machine. On the side, he wrote a crossword puzzle generator, which
took a blank pattern and a list of about fifty thousand words and filled in
the grid. The only manual labor was in writing clues for the words. It
was a perfect use of the Connection Machine's ability to tryout millions
of combinations of words very quickly. By the end of the summer, he
had a working puzzle generator. He was pleased enough with his work to
send one of the puzzles to the crossword editor of The New York Times.
To Robert's disappointment, the puzzle was rejected.
                                  ... T ...

Robert enrolled at Cornell in the last week of August 1988.
   Cornell is among the most isolated of major universities. Its campus
is in Ithaca, a small city with a population of twenty-nine thousand at
the southern tip of Cayuga Lake, one of the five large Finger Lakes in
upstate New York's lush farming region. In the first week he was there,
Robert skipped most of the computer orientation talks given by Dean
Krafft, the campus computer facilities manager. It seemed unnecessary
to attend basic lectures on logging in to the system and sending elec-
294   &    CYJg£RPUNK
tronic mail. Krafft had handed everyone a copy of the computer science
department's computer use policy, which prohibited the "use of ...
computer facilities for browsing through private computer files, decrypt-
ing encrypted material, or obtaining unauthorized user privileges."
While Krafft was giving his talks, Robert was already logged in to the
   He didn't make many friends at first. He moved into an old house
about a mile from campus with two other graduate students, but the
students kept to themselves. It was nothing like the easy communal
atmosphere at the house in Cambridge.
   Upson Hall at Cornell didn't seem to foster the camaraderie and
closeness of Aiken at Harvard. It was larger and more anonymous. Rob-
ert shared an office with seven other graduate students on the building's
fourth floor. The office had just two terminals. From his desk, Robert
had a southerly view that looked out on Cascadilla Creek gorge and on
Ithaca College across the hills on the other side of the valley. Interesting
science was happening all around him. One floor above in the building's
newer wing, the plasma physics group of the electrical engineering de-
partment was working on space plasma physics using data from the space
shuttle program. And in Cornell's computer science department there
were major research efforts in physical modeling and simulation, robot-
ics, and computer vision, as well as in reliable distributed computing,
which looks at ways of building systems to survive the failure of individ-
ual pieces.
   Robert started out taking basic graduate courses. In one, a small class
on microprocessor design, the professor noticed that Robert had an un-
usual curiosity about how things work. He seemed less interested in
concentrating on his assigned piece of the project of building a micropro-
cessor than in the bigger problem of chip design. If something didn't
captivate him right away, Robert was blunt about it. When another
professor gave him a paper to read, he returned it, saying it hadn't
interested him. He had spent enough years staring out the window when
something bored him. Now he spoke his mind freely.
   Robert was lonely and slightly distracted. He was late in submitting
one of his first mathematics papers, and received only a fair grade. He
was spending a lot of time at the computer but he wasn't necessarily
concentrating on his schoolwork. He made friends with one of his office-
mates, Dawson Dean. Dawson had gone to MIT and was just the kind
of one-dimensional digit-head Paul Graham so often complained about,
but Robert was also quick to give someone the benefit of the doubt, and
                                                           RTH   Of   295

he thought Dawson was an okay guy. He enjoyed having technical
discussions as much as Robert did. One night while both were working
late at Upson Hall, Robert and Dawson started talking about network
security. Robert pointed out that he had figured out several ways to
bypass security on local area networks.
   "Are you one of those people who breaks into computers for fun and
then gets hired?" Dawson asked.
   Robert smiled and nodded. He told Dawson' that he had given lectures
on security at the National Security Agency and the Naval Research
Laboratory. But, he told Dawson, he wasn't particularly interested in
making a career of computer security. "It's too boring," he told his
   Robert exchanged lots of electronic mail with his old friends from
Cambridge, most of whom had scattered to other places. A great deal of
mail came from David Hendler, as well as from Janet Abbate, a house-
mate from the previous summer in whom Robert had taken a romantic
interest. A graduate student at the University of Pennsylvania, Janet
was getting ready to return to Philadelphia. She checked in regularly
with Robert, sending him warm and high-spirited notes over the network
and baked goods by mail. Eleanor Sacks, the former manager at Aiken,
sent him a note saying she hoped Cornell would give him rtm, his
beloved login. He still hadn't bothered to change the morris login
Cornell had assigned him. Nick Horton, who had moved to Oregon,
forwarded Robert a half-dozen recipes for Thai dishes from the USENET
recipe exchange.
   Robert quickly developed a reputation as a talented programmer who
kept to himself. He wasn't aloof. He was just quiet. He sat apart from
others in classes and declined invitations to join professors and fellow
students for beer at a local pub on Friday evenings. Nonetheless, he took
advantage of many other things Cornell had to offer. He signed up for a
rock-climbing course, joined the computer science department's intra-
mural ice-hockey team and started singing with the Sage Chapel choir.

While the first personal computer virus probably emerged on the Apple
II computer in the early 1980s, it wasn't until 1987 that viruses exploded
into the public consciousness. A computer virus that struck at the cam-
pus of Lehigh University in Pennsylvania drew national attention that
year. A year later, viruses made the covers of Time and Business Week.
The programs were captivating because they were so mysterious and
296   A   CYJgERPUNK
because they offered such a clear analogy to their biological namesakes.
Computers became "infected." "Vaccines" were possible. Soon people
were drawing comparisons between computer viruses and the plague of
   The American public assumed that all viruses were malicious and that
they invariably destroyed data. But those who knew more about comput-
ing understood that there was no rule that such programs had to be
harmful. It would be much more interesting, in fact, to write a program
that was at once subtle and benign but still capable of spreading.
   Writing a virus that would spread to as many computers as possible
was an idea that just seemed to come to Robert, and the recent examples
emboldened him. He loved the thought of an invisible piece of software
that could propel itself through an electronic universe of thousands of
computers, spreading slowly and imperceptibly, achieving immortality
by protecting itself against anyone who might want to destroy it. And
there were certain flaws in Berkeley UNIX that he had known about for
at least two years, perhaps collecting them with the intention of someday
putting them to use. By early October, Robert was thinking in earnest
about writing the program. The goal of the program was simply to see
how many computers he could reach. On October 15, he produced a
wish list, a set of two dozen or so goals for his program, including these:

The goal is to infect three machines per
Ethernet (local network).
Only work if all users are idle.
Try to avoid slow machines.
Look through host table for the other
interfaces of known gateways, then find
hosts on that net.
Steal his password file, break a password,
and rexec.

   In Robert's mind, it was a perfectly harmless plan for probing security
of the network. It was something his father, in the early, clubbier days
of computer science, might have dreamed up to earn the respect of his
colleagues. It probably didn't occur to Robert that this was the kind of
thing computer saboteurs might generate in order to bring down an
entire international computer network.
                                                            RTH    T   297

Robert got a ride to Cambridge for the long weekend over fall break with
Dawson Dean. David Hendler was out of town on a long trip to Europe,
so Robert spent most of his time with Paul Graham. Andy Sudduth,
who was rowing that weekend in the Head of the Charles race, was
around too.
    It was like old times. Robert sat upstairs, glued to one of the worksta-
tions. Paul was sitting downstairs in the office of David Mumford, an
eminent mathematician on the Harvard faculty, whose office Paul occa-
sionally used when Mumford wasn't around. Early Saturday evening,
Robert walked into Mumford's office, wearing his trademark smirk. Paul
knew something was up. Robert was pacing furiously. He announced
that he had been reading UNIX source code and had found a big bug in
ftp, the file transfer program that enables users to copy files from machine
to machine over a computer network. The hole enabled someone to read
or write any file on the target computer.
    From the level of Robert's excitement, Paul got the impression that
he had just discovered this hole a few minutes earlier and was bursting
to tell someone. Robert's pacing in the small office picked up.
    Encountering Mumford's desk at the end of one of his paces, instead
of turning around again, Robert walked straight ahead and paced atop
the desk. It was a sure sign that he was completely lost in his discovery.
    "rtm! You're on Mumford's desk," Paul cried as he watched his friend's
sneakers shuffle the papers on Mumford's desk.
    "Oh," Robert replied, and descended from the desk.
    At first, Paul wasn't sure what Robert was driving at. It sounded like
just another way of breaking into UNIX. "Well that's an amusing hole,
but what's the point?" he asked.
    "I could use this hole to write a virus," Robert explained. He said that
for much of the fall term at Cornell he had been thinking about writing
a virus, one that would spread slowly over the Internet. As Robert
described it, the virus wasn't going to do anything malicious and cer-
tainly would not destroy data. In the end, it should do nothing at all
except spread to as many machines as possible.
    Paul was immediately intrigued. He had been badgering Robert all
semester to make more friends at Cornell and improve his social life, but
when he heard that such efforts had been deferred in favor of something
as intriguing as this computer virus, he was delighted and envious.
    "That's really great!" Paul was getting just as worked up as Robert.
 "You should do this for your dissertation!"
    Paul was, in some ways, the perfect friend. When he got excited about
298   A   CYJg£RPUNK
an idea, his support was all the encouragement one needed to keep
going. He could pull others into his excitement just by virtue of his own
enthusiasm. Especially when it came to his friend and role model rtm,
Paul was a one-man cheerleading squad.
   When Robert began to talk about the virus he was planning, Paul's
enthusiasm tripped into high gear. It must have had its effect even on
the usually placid and low-key Robert. Robert had meant to keep his
virus plans to himself. By telling Paul, of all people, Robert must have
gotten both a validation for his idea and a stronger sense of urgency.
Had he told anyone else about it, such as Nick or Andy or David, they
might have been less encouraging. They would probably have suggested
he simulate the experiment first, perhaps by running it on a smaller
network that had been disconnected from the Internet at large. Safety
measures like that would keep the virus from affecting the entire network
in case in contained an error. But a controlled environment would have
made it a more modest and less scientifically interesting experiment.
Robert wanted a large proving ground.
   Paul and Robert went to meet Andy for dinner that night at Legal
Sea Foods, a restaurant across the street from MIT frequented by pro-
grammers and engineers. While Paul and Robert were standing outside
waiting for Andy to arrive, the subject of the virus came up again. Since
neither Paul nor Robert knew of anything like this having been done
before, thinking about it required a lot of creativity. Both of them
thought the idea sounded like the kind of "great hack" that was often
dreamed up in the computer world. Robert started thinking out loud,
describing to Paul some of the more important features such a program
would require. First, of course, it would spread through the network,
secretly planting itself in many different machines. An important goal
was to make the virus as inconspicuous as possible, so as not to arouse
suspicion from system managers. Once it had taken up residence it would
need a means of knowing whether or not another copy was already
present. And it would have to regulate itself in order to limit the number
of copies in each computer. But a difficult question was still unresolved:
how to limit the growth without halting it completely.
   While they waited, Robert sketched out his ideas. The virus would
enter a computer through the UNIX loopholes he had found and look
around the system for any possible copies. If it found one, the two would
"talk" to each other and decide what to do. Ideally, one would automat-
ically stop running in order to limit growth. But what if someone discov-
ered the virus and tried to trick the incoming virus into believing that a
                                                           RTH   T    299

copy was already running on the machine it approached? A programmer
could design a decoy to fool the invader into thinking that a copy already
existed on a computer. Such a program, easy to write, could prevent the
virus from spreading-serving the same purpose as a biological vaccine.
Thinking like chess players, Robert and Paul decided that there would
have to be a countermeasure against potential defensive programs. How
could they fool the decoy?
   Why, randomization, of course! They had taken a graduate course
together in efficient algorithms taught by Michael Rabin, a prominent
mathematician and cryptologist. The concept of randomization was big
with Rabin, who told his students again and again that if a problem
seemed impossible to solve they should reduce it to a simpler one and
apply randomization. (This was the philosophy behind Bob Morris's
probabilistic typo checker.) Rabin had discussed randomization as it
applied to abstract problems, such as prime-number searches. But it
occurred to Paul and Robert that they could use the concept in the virus
program. When the program entered and detected another copy, it
would toss an electronic coin to decide which one should stop running.
   Another way of insuring the virus's survival occurred to Robert. One
in N times the virus should enter a computer, forget about the electronic
coin toss and simply command itself never to stop running. But then
came a new question: what should N be? Five? A thousand? Ten thou-
   Just as they were beginning to ponder this, Andy walked up. Al-
though he was a close friend, his job as system manager at Aiken would
place him in an awkward position, to say the least, if he were suddenly
privy to discussions of huge security holes in UNIX. Andy thought the
conversation had ended abruptly because his friends were discussing a
woman in whom both Andy and Robert were interested.
   Robert was still thinking about the virus. And he couldn't contain his
enthusiasm for his discovery of the ftp bug. The next day, he walked into
Andy's office at Aiken and casually told him about it. Just as casually,
he told him not to spread it around.
   Andy didn't waste much time before attempting to verify the bug.
When he couldn't, Robert had to give him a more explicit description
of how it worked.

The Wednesday after Robert returned to Cornell, Paul sent him some
electronic mail: "Any news on the brilliant project?" Two days later,
300   ~   CYEERPUNK
Robert sent a message back: "No news. I'm buried under legitimate
work." Paul took that to mean schoolwork. But the virus project was
still alive. One of the most time-consuming things Robert had done
during his four days up at Harvard was to decode a collection of en-
crypted password files he had taken from various machines around the
    While directly decrypting a password may not be possible, guessing
often works. It's impossible to decode a password by reversing the process
that created the coded version. However, nothing prevents the decoder
from guessing at the original password by coding, say, an entire dictio-
nary in the same manner and matching the results against the coded
password. Since many passwords are ordinary English words, a dictionary
method yields a surprisingly high number of matches. The faster the
computer, the more computers used, the less time it takes.
    On that Friday evening, Dawson Dean walked into the computer
terminal room at Upson Hall, where Robert was seated at a Sun work-
station. Dawson asked Robert what he was up to. When Robert showed
Dawson what he had on his screen, Dawson's eyes widened. It was a
long list of passwords in plain, unencrypted form. Robert scrolled down
the list to reveal passwords of dozens of Cornell students and professors.
Dexter Kozen, the graduate adviser, was on the list. His password? to-
mato. Keshav Pingali, who was teaching the course in microprocessor
design, had chosen snoopy.
    "Wow!" Dawson exclaimed. "Is my password on there?"
    His password didn't show up because it wasn't a dictionary word.
    "How about Aitken?" Bill Aitken was another graduate student whom
Dawson thought to be "really obnoxious." Robert scrolled down and
found Aitken's password, subway.
    "Isn't it kind of dangerous to have a decrypted password list lying
around in your account?" Dawson asked. The very tone of the conver-
sation-Dawson's excited questions, as if he were getting a vicarious
kick out of the illicitness of what Robert was doing, and Robert's careful
answers-implied that there were some real taboos being broken here.
    "Well," Robert replied, "you encrypt what you can. For the rest you
basically take your chances. "
    Dawson's curiosity had gotten the better of him. "Could you do this
all over the place and build a nationwide data base of passwords?"
    Robert told Dawson that he had other ways of getting into machines
without relying on dictionary encryption.
    Dawson pressed Robert to tell him about the other ways he had of
                                                             RTH    T   301

breaking into machines. Robert hesitated, but Dawson wouldn't give up.
Finally Robert said that from reading UNIX source code, he had discov-
ered several bugs. One was the back door in the sendmail program.
Another was a bug in finger that would also allow him to run a program
on another machine without logging in. Robert said he had known about
the two bugs for a year and didn't think anyone else knew about them.
   Dawson Dean wanted to skip the bugs talk and hear more about
specific computers Robert could get into. He asked about a specific pri-
vate company. Robert shook his head. "You could get in there, too," he
said, "but you really want to stay with machines that are owned by
universities. Universities are less strict about their security in general."
Further, Robert said, it wasn't really a good idea to access machines
across state lines.
   Dawson then asked about one machine at the MIT Media Lab, a
center at MIT that was studying technology and communications.
Within a few minutes, Robert had logged on to the computer.
   Dawson was amazed. "What's the account you're logged in as?"
   Robert typed a command that would tell him the account name he
was using. The computer replied with "nobody." Dawson was impressed.
Obviously, Robert had fooled the machine and was logged on illicitly.
   But Robert didn't tell Dawson Dean he had any plans do anything
with the bugs. And Dawson Dean didn't ask. In fact, aside from Paul,
Robert hadn't told anyone of the program he had been planning all
semester, and on which he had been working in earnest since just before
the trip to Cambridge. By this time, in fact, Robert had been program-
ming the virus on and off for a little more than two weeks.
   A week later, on November 2, Robert was dismayed to see a posting
on the network: Keith Bostic, who worked on Berkeley UNIX, had
posted a fix to the flaw in ftp. Since it had been just the week before that
Robert had told Andy about the bug, this couldn't be a coincidence.
Robert immediately suspected that Andy had alerted someone at Berke-
ley about the ftp bug. He fired off a note to Andy and asked him if he
had leaked the secret. No reply. This meant that Robert could no longer
use the ftp bug for his virus. But the flaws in sendmail and finger remained.
   Robert spent that afternoon and early evening putting the final
touches on the virus. He finished the work at 7:30 P.M., Eastern Stan-
dard Time. An hour later, having logged on to a computer at the MIT
Artificial Intelligence Lab, he typed in a few commands to execute the
program. He went to get some dinner.
   In the time it took Robert to put on his jacket after pressing Return,
302   •    CYJg6RPUNK
the program began to spread. Within a few minutes it was already fan-
ning out over the network. Computers started infecting one another like
toddlers in a day-care center. Any VAX or Sun machine linked to any
other VAX or Sun was instantly vulnerable. While Robert was eating
his dinner, dozens of copies of the virus were already swarming around
inside machines, vying for computer time. Machines had begun to slow
down and then crash.
   Robert had planned to go home after dinner, but he couldn't resist
returning to Upson Hall to check up on the program's progress. When
he logged in, the computer wouldn't respond. Something seemed to be
going wrong: The virus was replicating out of control.

Later that night, at about 11:00 P.M., Paul and Andy had just returned
to Aiken from a late dinner. As Paul was pulling his keys out of his office
door, the phone rang. Andy answered it for him. It was Robert, who
asked to speak with Paul. Andy put Paul on the line and went back to
his office.
   Robert sounded miserable. "I think I've really fucked up," he said. It
was the first time Paul had heard rtm utter such strong language. From
the tone in his friend's voice, which was much softer than usual, Paul
knew rtm was very upset. Paul's first thought was that it must have
something to do with a woman.
   "What do you mean?" Paul asked. "What'd you do?"
   "I started a virus and it isn't working at all the way it's supposed to,"
Robert replied. "I got one of the numbers wrong on how it should
propagate. "
   "What number did you use?" Paul asked, referring to the question of
how frequently the virus should infect a machine even if there was
already a copy present.
   "One in seven. "
   "One in seven?! rtrn, you jerk! Why seven?" At that instant, it was
clear to Paul that the number should have been higher by a factor of a
thousand or more.
   But Robert wasn't eager to sit around evaluating his error in judgment.
He told Paul that all the Suns and VAXes at Cornell were messed up,
crashing every few minutes. And if all the Suns and VAXes at Cornell
were messed up, it was reasonable to assume that lots of other computers
around the country were, too. He said he had launched the virus earlier
that evening from a computer at the MIT Artificial Intelligence Lab. He
                                                            RTH ... 303
had gone out for dinner, and when he returned to check on the virus's
progress, he saw that it was stalling machines everywhere he was able to
   They discussed ways to fix the virus. Paul's idea was to send another
program after the virus to kill it, a Pac-Man-like program that would run
after it and eat it up. As Paul got more excited about his Pac-Man idea,
Robert just grew more morose. If he had already made a mess of one
program, what reason was there to believe that he wouldn't screw up on
a second one?
   The next idea was to get Andy involved. Paul went up to Andy's
office and peered in the door. Andy was still there, working late to install
new hardware on the lab's computers. "I think you'd better get in touch
with rtm," he said. "There's something really big going on. But I can't
tell you what it is." Paul was pacing back and forth in front of Andy's
    Andy was ever skeptical of anything Paul might consider really big.
"What's going on?" he asked.
    "You'd better talk to him about it," Paul said. "He told me not to tell
    "Why don't you just tell me?" Andy was losing his patience.
   That was all the coaxing Paul needed. "Well, don't let him know I
told you, but he's written this computer program and it's taken over the
whole country. It's out of control now! It's incredible!"
    Andy was still skeptical. But then he remembered receiving an un-
characteristically conspiratorial message from Robert that afternoon,
asking if he had told anyone about the ftp bug. Andy had in fact men-
tioned it to a few people; he had even demonstrated it to them. And he
had used it to get full privileges on Nick Horton's machine in Portland
and then later told Nick about it. He figured that someone had even-
tually told the Berkeley people about it. It wasn't until Robert registered
his concern that Andy remembered Robert's having asked him not to
spread the word. Andy hadn't responded to Robert's earlier note, but
now he typed a message back: "Sorry about betraying other trusts," he
wrote. "Tell me what's going on." Andy was, above all, concerned about
what could happen to the Harvard machines.
    An hour or so later, Andy got a call from Robert, who sounded
unusually subdued. Robert told Andy there was a virus out in the net-
work that seemed to be bringing down a lot of machines. He didn't say
he had written it, and Andy didn't need to ask. Andy wanted to know
whether the computers at Harvard would be affected. No, Robert re-
304   •   CYEERPlINI<
plied, because Harvard had already patched the holes the virus was using
to get in. An hour later Robert called back and asked Andy to send an
anonymous note out to the network with directions on how to fix the
virus. Robert told him the points he wanted to make and Andy com-
posed the following message:

A Possible virus report:

There may be a virus loose on the internet.
Here is the gist of a message I got:

I'm sorry.

Here are some steps to prevent further

1) don't run finger, or fix it to not
overrun its stack when reading arguments.
2) recompile sendmail wlo DEBUG defined
3) don't run rexecd

Hope this helps, but more,                  I hope it is a

   After Robert had dictated his brief apology and cure, Andy told him
that he would make sure it was sent from a remote machine that wouldn't
be traced to Robert or Andy. Andy decided that he didn't want to be
the one to say Robert had done it. He thought Robert should be the one
to decide when and if he was going to admit to having done it. Andy
told Robert to be prepared to lie to people. If anyone asked him about
the virus, he said, Robert should try not to smirk.
   After he hung up, Andy thought about the best way to send the
message. He knew theoretically how to send out anonymous electronic
mail messages, or at least how to pretend to be another computer deliv-
ering mail. The message had to look as if it had come from some place
other than Harvard, and certainly not from Cornell. He decided to post
the message to a network discussion group on a computer at SRI. He
realized that if he sent it directly to the computer at SRI that fed the
network, it might get traced straight back to Harvard. So he created a
                                                             RTH " 305
fictitious ongm for his message-foo@bar.DARPA-and routed the
message through a computer at Brown University, expecting to see it get
sent to SRI within a few hours.
   As it turned out, Andy's message got bogged down on the first leg of
its trip. The Brown computers were besieged by the virus. Worse, Andy
had not indicated the subject at the top of the message, making it likely
that it would be ignored, or given a low priority, once it finally did arrive
at SRI.
   Andy also tried to make a few phone calls to Berkeley to tell the
Berkeley UNIX people there about the virus, but he wasn't sure just
whom to call, or even how to find their numbers. Calls to the main
number on the Berkeley campus brought no answer, and Andy decided
the whole thing probably didn't justify rousting people from bed in Cal-
ifornia, where it was already midnight.
    Andy knew that anyone, even as gifted a programmer as Robert, could
make a mistake. Andy had once inadvertently brought down two
hundred computers at Harvard because of a small error in a computer
network routing command. The blunder had been a breach of Harvard's
official policies regarding computer use, but because university officials
recognized it as an honest mistake, they didn't censure Andy. This virus
didn't seem to be as disastrous as Robert and Paul were claiming. If
Robert had indeed brought down a bunch of Cornell computers, then
perhaps some people would be upset. But it couldn't be too terrible.
Finally, satisfied that he had done what he could for his friend, Andy
went home at 4:00 A. M.
    That Robert had committed a grave transgression-something illegal,
in fact-didn't occur to Robert, Andy or Paul. Robert's worst fear was
that people in the computer community would be beside themselves with
anger. He hoped he wouldn't get in trouble with Cornell. Seeing what
he had already done to the Internet was very upsetting. And his little
program was probably still ricocheting out of control. He could only
hope that Andy's message would help stem the damage.
    But news of the virus that ate the Internet was all over Aiken by the
time Andy got to work on Thursday morning. And Robert Morris's
name, it seemed, was on the tip of many a tongue. After all, Robert's
reputation at Harvard was that of a security expert, Internet habitue and
occasional prankster. The only aspect of this incident that wasn't in
keeping with Robert was the apparent malice behind a move that would
crash computers all over the network.
306   •   CYJg£RPUNK
    It was unclear to Andy whether his message had reached people.
Reports were filtering in from Berkeley and MIT telling everyone on the
network how to get rid of the virus. No one mentioned an anonymous
message that was going around, but the instructions given were exactly
those Robert had dictated to Andy.
    It was with some difficulty that Andy told professors around Aiken
that he didn't know a thing about the incident. Paul appeared to be
having an easier time with the deception. When another computer
science graduate student asked Paul if Robert had anything to do with
this virus he was hearing about, Paul looked him squarely in the eye and
said no. During the day, a subdued-sounding Robert called Andy to ask
if he had sent Robert's message out. Andy assured him he had.

When Keith Bostic of UC Berkeley arrived at work at 6:00 A.M. Thurs-
day, after three hours' sleep, the phone was already ringing. Calls were
coming in from angry computer managers around the country, demand-
ing to know what to do about the program that had infested their sys-
tems. Bostic had already anticipated some of the wrath. Did he know
about these holes in Berkeley UNIX? No, Bostic responded, he didn't.
Computer managers at the Defense Department, one of the largest cus-
tomers for Berkeley UNIX, were particularly irate. Did Bostic have any
idea who had committed this heinous act? Had he been aware of holes
in Berkeley UNIX? Could he assure them that there were no Trojan
horses lurking in the program? Did Berkeley plan to disassemble the virus
code layer by layer?
   One of Bostic's first tasks that day was to send out Virus Posting #2,
an amendment to his first patch for the sendmail bug, providing a more
complete fix to the program. That message went out at about 8:00 A.M.
   Bostic and the others had already discussed the question of disassem-
bling the virus. It would be a time-consuming and arduous task, but it
was the only way to determine beyond a doubt whether there was any
destructive code hidden somewhere inside.
   The work at Berkeley was being duplicated in Cambridge, Massachu-
setts, by a group of MIT programmers who had also stayed up most of
the night. In the middle of the day, a message from MIT arrived in
Berkeley with news of a second method of attack the virus was using.
Exploiting a hole in the small UNIX finger program, the virus was able
to crash finger by sending it more characters than it could handle. Once
it had overflowed the storage space, the invader was able to start a small
                                                            RTH   T    307

program that called back across the network and brought the entire body
of the virus into the target computer.
    Bostic was skeptical about what MIT was telling him-finger, after
all, was such a trivial little program. He couldn't imagine that a program
only fifty lines long could contain significant bugs. But he was wrong: to
prove its point, the MIT crew sent him a sample program that demon-
strated the hole in finger. Later that day, Bostic sent out Virus Posting
#3, a fix to the finger program.
    Seeing the finger attack was enough to prove to Bostic that the only
way to find out if other dangers remained was to pore over it line by line.
The program would have to be laboriously decompiled.
    Decompiling programs is something of an arcane art that entails trans-
lating a program from ones and zeros, which a computer reads as "on"
and "off" instructions, into something a human programmer would write
and understand. Decompiling a computer program is like taking a book
that has already been translated from the original English into, say,
French, then translating it back into English, all without seeing the
original version. The new English version may not use the same words,
but good translators can convey the book's proper meaning. When a
program is decompiled, the language itself may be slightly different, but
its behavior should be identical to the original.
    Programs are normally compiled, not decompiled. That is, once a
program is translated from the original source 'code into machine-execut-
able code, there is seldom any reason to do things the other way around.
In fact, many commercial software licenses prohibit disassembling pro-
grams precisely because the people who do so might want to break copy
protection or modify the software. But having a program's original source
code is invaluable because it is a window into the author's intentions.
And since the author of this particular program had apparently gone out
of his way to hide his program, there was no choice but to decompile
the program, an arduous task that few programmers have much practice
    Berkeley, as it turned out, was ideally suited to the job. Not only was
the Berkeley version of UNIX created and still maintained there, the
Berkeley campus was the site that week of an annual gathering of UNIX
experts from around the world. A year earlier, during the same UNIX
conference, the stock market had crashed. This year it was the Internet.
Bostic even had Chris Torek, one of the nation's leading UNIX experts,
staying at his home. In addition to T orek, there was a compiler expert
from the University of Utah named Don Seeley at the conference. That

           308   &   CygERPLlNK
           much talent alone would probably have been enough to decompile the
           program. But Phil Lapsley and Peter Yee knew of yet another decompi-
           lation ace.
              Dave Pare became expert at decompiling programs as an undergradu-
           ate at the University of California at San Diego in 1985, when he got
           miffed at the author of a computer game called Empire who refused to
           distribute his source code. The twenty-two-year-old Pare put his mind
           to decompiling Empire in its entirety. It took him the better part of two
           years. Now he was in Silicon Valley, fifty miles south of Berkeley, work-
           ing for a software developer. Not only was Pare proficient at decompiling
           code, but he had also written his own disassembler, a program that tried
           to make decompilation as easy as possible by automating some of the
           more mechanical steps. So Peter Yee called him Thursday afternoon to
           say there was something they needed his help on.
              This was the first Pare had heard of the virus. "Where's Phil? Can't
           he handle this?" he asked.
              "Phil's asleep. He was up all night," Peter replied.
              That was enough to convince Pare that this was something big. He
           had never known Phil to stay up all night.
              Pare got in his car and made the hour's drive up to Berkeley. When
           he got to Evans Hall, he sat down with Chris Torek at one workstation,
           while Bostic and Don Seeley sat at another one across the room. The
           office was transformed into a disassembly line. The job of the Pare-Torek
           team was turning the raw ones and zeros of each of the program's routines
           into assembly code, then into rough code in the C programming lan-
           guage. Once they had each routine in hand, they gave it to Bostic and
           Seeley, who tried to make sense of the code's precise purpose. The UNIX
           conference on campus was alive with talk of the virus that had taken
           hold of the network the previous evening. Some who had planned to
           arrive in Berkeley Thursday morning were forced to stay home and battle
           the invader. For those who were there, discussions of the siege over-
           shadowed the workshops on subjects such as "UNIX with NPROC =
           3000" and "Kernelization of MACH. " Two of the attendees had already
           been pulled from the conference to help in the decompilation effort, and
           during breaks others wandered over to Evans Hall to see how things were
           going. When the programmers got hungry, they ordered calzones from a
           nearby pizzeria and kept working as they ate.
              Teams of programmers on both coasts continued pulling the code
           apart. But behind the spirit of cooperation, an element of competition
           crept into the exercise. Each school was privately hoping to finish first.
                                                           RTH " 309
Besides, once each group had settled into its own rhythm, it was far
easier just to do the work alone than to adapt to someone else's method.
   When Keith Bostic wasn't busy taking panicked telephone calls or
responding to electronic mail, he helped with the effort. At least once
an hour he got an anxious call from various branches of the Defense
Department, asking if the Berkeley team had finished disassembling the
   The suspense came from uncertainty over what instructions the rogue
program might contain. At one point, Pare got nervous when he saw
that there was code that appeared to have a timer on it.
   "Hey, guys," Pare called to the others in the room. "After twelve
hours it does something."
   "What?" came the chorus of replies.
   "It calls a routine called H_Clean. "
   H_Clean? Could that mean Host Clean? And if it did mean Host
Clean, did it intend to clean out the files of the host it was running on?
   Bostic rushed across the room to peer over Pare's shoulder. Anything
that was timed was not a good sign. They had no idea what would
happen once the timing mechanism was tripped. With an edge of panic
in his voice, Bostic said, "Dave. Time out. that routine. Now."
   So Pare set to work on the H_Clean routine as the others watched.
As it turned out, the routine was designed to erase the virus's own list of
the hosts it had infected during the previous twelve hours. It was nothing
to worry about.
   There also appeared to be a piece of code in the virus designed to send
a little bit of information-a signal at regular intervals-to Ernie
CoVAX, a computer in Cory Hall that computer science graduate stu-
dents used for sending and receiving mail. It appeared that this part of
the program was supposed to act as a foil to throw pursuers off the scent
by making it seem that the program was coming from Berkeley, but there
was an error in the routine that was supposed to send data to Ernie, so
nothing was ever sent. The Berkeley team came upon other errors in the
virus's code as well. They appeared to be careless mistakes. For example,
whoever wrote the program forgot, at one point, to assign a value to a
variable; the author also misdirected a message to another program. The
most mystifying thing to Dave Pare was the inconsistent quality of the
code. Parts of it were extremely well written while other parts were so
sloppily executed they appeared to be the work of someone else entirely.
   In Cambridge, the MIT group found a more significant flaw: the
dialogue between a newly arrived virus and one that was already estab-
310   A    CYEERPUNK
lished would necessarily end in disaster because the listening program
would not always listen long enough to the new infection to acknowl-
edge its arrival. Therefore, because each program thought it was alone
on the computer, the "electronic coin toss," which in most cases was
supposed to result in the self-destruction of one of the copies, would
never take place. This was a major error. As it was, the author had
guaranteed that the virus would clog the network because one in seven
times neither the listening program nor the new arrival would destroy
itself; this newly discovered flaw meant that a logjam would have resulted
even if the author had chosen a frequency of one in a hundred thousand.
    By 4:00 A.M. the next day the main structure of the program had been
reconstructed. By now, it was clear the virus was basically harmless. So
early Friday morning, Bostic sent out his fourth and final virus posting
to the network. It was a list of fixes to the virus itself. This final posting
was a bit of a joke because the Berkeley team was wagging a finger at the
author of this clever but in some ways careless program. After that,
Bostic went home to sleep soundly for the first time in two days. As soon
as they were finished with the disassembly, the Berkeley team sent a
copy to the anxious officials at the Defense Department.
    But no sooner was the program decompiled than a controversy erupted
over whether decompiled versions should be posted on the network.
Bostic and others at Berkeley were against it, arguing that they didn't
want some high school student to take it and try it out. Bostic's detrac-
tors argued against adopting this patronizing, "father-knows-best" atti-
tude.Bostic stood his ground. As far as he was concerned, sending out
the source code would be "the electronic equivalent of scattering guns
through the network." At the same time, he said, Berkeley wasn't trying
to hold back anything about what the program did. Moreover, although
the Defense Department made no specific request of Berkeley, officials
there told Bostic they were pleased that he wouldn't be sending out the
disassembled code.

Robert didn't return to Upson Hall on Thursday morning. He stayed
home for much of the day, trying to concentrate on schoolwork. He
went to choir practice that evening and on his way back from the chapel
he stopped by Upson to log in and read his mail. When he logged in, he
saw that most of the computers were working fine. In his mailbox were
notices from the Cornell staff that said there was a virus loose but that
Cornell had it under control. There were also some bulletin board no-
                                                           RTH    T    311

tices from Berkeley about patching the holes the virus had used to get
in. And there was a message from Paul asking Robert to call him.
   Andy and Paul had dinner that night with David Hendler, who had
just returned from his long trip.
   "Well, have you heard?" Paul asked David.
   "Heard what?" David asked.
   "The virus that's been going around the Internet!" Paul could hardly
contain himself. "Andy was up all night! It was out of control."
   "Oh." David smiled. "Did Robert do it?"
   The trio went back to work at Aiken, and when David walked into
Paul's office shortly after 11:00 P. M., Paul was on the phone with Robert,
telling him what a huge mediaevent the virus had become. Robert didn't
have a television and he was shocked to hear that it was one of the top
stories on all the networks. Paul was also trying to cheer him up by
reading aloud from The Oxford Book of Light Verse. Robert asked to speak
with David. David had expected an effusive welcome after his absence,
but Robert was monosyllabic. "This thing is mine," Robert mumbled.
The tone in Robert's voice was not only that of a programmer who was
upset to have erred, but of someone who was aghast at himself for having
erred so visibly.
   David wasn't surprised, but was still of a mind to joke about it. "Do
you want to meet in Montreal?" It was, after all, the city closest to
Ithaca that was outside the U.S. When Robert didn't laugh, David knew
his friend was extremely upset. So he got practical. "What are you going
to do?"
   "I don't have any idea," Robert replied.
   T en minutes later Robert called back. He had called his father and
was going to leave Ithaca the next day. He didn't say where he was

Bob and Anne went out to dinner Thursday night. They discussed the
computer virus that had been going around the Internet. Cliff Stoll had
called Bob that morning and told him about it, but Bob had been too
busy with other things to spend too much time thinking about its origins.
   At 11:30 P.M., the phone rang. Bob was already in bed. Anne an-
swered. She was surprised to hear Robert's voice. Ben frequently called
this late, but not Robert.
   "Can I talk to Dad?" he asked.
312     A   Cyg£RPUNK
   "He's asleep," Anne told him. "Is it important?" From the tone in
Robert's voice, she already knew that it was.
   "Well, I'd really like to speak to him," came the reply. He was as
insistent as she had ever heard him.
   She called Bob to the phone.
   It was a short conversation between father and son. When he heard
what Robert had done, Bob was perturbed but not angry. Robert told his
father he already had a plane ticket to Philadelphia for the following
day; he had been planning to spend the weekend with his friend Janet.
Bob told him to use the ticket, not to talk to anyone and not to tell
anyone where he was going. It was likely he would need legal advice.
   When Anne went to work the following day, the staff were milling in
the coffee room, talking about the computer virus. They barely knew
what Anne's husband did for a living, much less her children. Tables
were strewn with newspapers, all with news of the virus prominently
displayed. Queasy and unable to concentrate, Anne left work early. By
late that afternoon, The New York Times had figured out that Robert was
the author of the virus and was planning to run the story in the paper
the following morning. Anne and Bob spent the afternoon looking for
an attorney. By the end of the day, they had several names. If Paul
hadn't kept talking to The New York Times, there would have been more
time for the family to figure out what to do. But Paul's careless slip about
Robert's login had accelerated the pace of events.
RITY"  was Saturday's front-page headline in The New York Times. The
paper hadn't been able to get a photograph of Robert in time, but the
following day, photographs of both father and son appeared. Bob looked
the very picture of an eccentric scientist. His long, untrimmed, graying
beard complemented his eyebrows, whose arch resembled birds in flight.
Hair all but covered his craggy face.
   Even with the presidential election coming up the following Tuesday,
the media had an insatiable appetite for the story of the young computer
whiz, son of a computer security expert, who loosed a rogue program on
a nationwide network and brought it to its knees. By Sunday morning,
a crowd of television and newspaper reporters had settled at the end of
the Morris driveway, where they would remain on and off through the
weekend and into the following week. The telephone inside rang cease-
lessly with calls from the press. Bob's sense of humor stayed sharp. When
a friend called and opened the conversation with "This is not a press
call," Bob responded, "Oh, then you must have the wrong number. "
                                                          RTH ... 313
  It seemed wise for Robert to remain in Philadelphia and steer clear of
his parents' house and the pack of journalists, so on Sunday, Bob and
Anne drove to Philadelphia to retrieve him. On their way back, they
pulled off the highway for gas and Bob got out of the car. Just then, a
red sports car pulled up alongside the gas pumps and the driver glanced
at Bob. A wide grin spread across his face as he took a longer look.
"Hey!" he called out. "You a computer scientist?"
   The nation's press corps fastened onto the story first as an incident
that had disrupted a network of military computers, then, as the identity
of the culprit emerged, as the story of a remarkable family, and of intel-
lectual pranksterism gone awry. For several days in a row, it was front-
page news in the nation's newspapers and one of the top stories in
television newscasts. By Monday, every newspaper in the country, it
seemed, was already writing editorials. Mike Royko, the acerbic Chicago
Tribune columnist, called for a stiff prison sentence. Journalists tapped
every computer security expert and computer industry executive they
could find. "The MacNeil/Lehrer NewsHour" interviewed Ken Olsen,
president of Digital Equipment Corporation. Even though Digital's com-
puters had been a target not just of this incident but of hackers with
more felonious intent, Olsen urged the computer science community not
to respond by placing increased security on computer networks. "The
worst thing that could happen," Olsen said, "is that we clamp down on
the free flow of academic information, because that should be preserved
at all costs. "
   When Robert's name first came out, some of those who knew him
well weren't terribly surprised. Bob's old Bell Labs friend Doug McIlroy
got up early Saturday morning and read the news. The slumbering Me-
Ilroy family was awakened by a booming "Guess who did it!" There was
already a story circulating among computer scientists that as the virus
was knocking on the door at Bell Labs, some of the old UNIX hands
were chuckling among themselves, saying, "Must be Morris's kid."
    But others who knew Robert, even those who had been on the receiv-
ing end of some of his Harvard pranks, were incredulous, or at least
willing to give him the benefit of the doubt. One Harvard faculty mem-
ber who knew Robert well was hoping that he had released the virus on
a small local area network at first, had gone home and had returned the
next morning to find that it had somehow spread out over the Internet.
Another faculty member, whose courses Robert had taken, shook his
head in disbelief and asked, "Why didn't he simulate it first?"
   Andy, Paul and David Hendler, in the meantime, sat in Andy's office
314   •   CygERPUNK
at Aiken discussing ways to protect their friend. Cooking up a bit of a
propaganda campaign, the group wanted to see the press describe Robert
in the best possible light. Paul was enjoying his role as spin doctor on
the story. He told the others that he had repeatedly described Robert to
the Times reporter as brilliant, and the reporter had used that description
in his story. As they were sitting there, Robert called. He didn't tell
them where he was. He said he just wanted to check in.
   "What are you doing?" David asked when he got on the line.
   "Baking cookies to send to friends," Robert replied.

FBI special agent Joe O'Brien had moved to Ithaca from New York City
in 1984 with specific orders to lie low for a while. He had been involved
in an organized-crime case; he was the principal agent responsible for
placing highly sophisticated electronic surveillance equipment in the
home of the late Mafia boss Paul Castellano, which eluded detection
even by the experts Castellano hired to root out bugging devices.
O'Brien's work had led to a series of trials and convictions of organized-
crime figures, and the bureau suggested to O'Brien that he move to a
nice quite place where nothing much happens. Ithaca, New York, was
the ideal place. Compared with some of the other FBI field offices, the
three-man outpost in Ithaca was one of the sleepiest. Background checks
on graduates of Cornell University and Ithaca College who had applied
for government jobs made up the bulk of O'Brien's caseload. The last
thing he wanted was another high-profile case.
   When O'Brien heard of a computer virus that was sweeping the na-
tion, his only concern was whether it would affect his home computer,
an old Apple II he was using to write a book about his undercover role
in infiltrating the Gambino family. The next night, when the 10 o'clock
news revealed that The New York Times was about to identify the culprit
as a Cornell student, O'Brien realized he had much more to be con-
cerned about. O'Brien knew next to nothing about computers. He
switched off the news and called his neighbor, a Cornell scientist in
charge of the school's supercomputer center.
   Cornell had beaten O'Brien to the news by thirty minutes. At 9:30
P. M., the public affairs office received a call from The Washington Post
asking about Robert Morris. The call set off a chain reaction: the public
affairs officer called the vice-president in charge of computing, who
called the president, the provost and the computer facilities manager.
The chairman of the computer science department had already spoken
                                                           RTH " 315
with Bob Morris, and within an hour a small group was assembled in the
chairman's office. Dean Krafft, the facilities manager, combed through
Morris's computer files for telltale signs. There wasn't much to see in the
student's active account, so Krafft went to the backup tapes. Within an
hour he found enough to be virtually certain of Morris's authorship of
the virus. Among recipes and invitations to hockey games he found two
files, one called try.out, another called Stanford. Both were slightly hid-
den in a directory that wouldn't show up ordinarily. It was difficult to
tell exactly what was going on because the bulk of the files were en-
crypted. He saw that the final version of the program was last modified
at 7:26 P.M. on Wednesday, November 2. And among Morris's personal
mail from the previous two days was a message from Greg Kuperberg, the
math whiz who had helped Robert with the graphics program. Kuperberg
had written Morris a message on November 3 warning him of a virus on
the Internet.

O'Brien could have lived without a computer crime case, and would
have been perfectly happy to hand it over to Mike Gibbons, the FBI's
sole computer crime expert. Gibbons had parlayed a computer retailing
job into a position as the bureau's foremost authority on computer crime.
He had been deeply involved in tracking down the West German hack-
ers who had been plaguing Lawrence Berkeley Laboratory, uncovering
what turned out to be a new form of espionage. Gibbons was the only
FBI agent who knew how to write a search warrant tailored specifically
to a computer crime case.
   O'Brien half expected to see a special team of agents descend on
Ithaca as soon as the case broke. It was routine for the bureau to dispatch
specially trained units to handle hostage crises and difficult kidnappings.
But there was no such unit at the ready for computer crimes-if this was
in fact such a case. Gibbons was good, but there wasn't much his exper-
tise could add to an investigation that, in the end, would be carried out
much like any other, with agents talking to as many people as possible
and gathering what evidence there was. So O'Brien didn't have much
choice but to set aside all else to start getting to the bottom of what was
turning into the nation's biggest computer break-in to date.
   When O'Brien went to have a look at Robert's office on Saturday,
there were at least a half-dozen reporters already milling around, scouring
the contents of the student's desk: a squash racquet under a stack of
computer books, a pile of pennies, a computer-generated crossword puz-
316   •   CYKERPUNK
zle. A problem set done for a course on the design and analysis of
algorithms was marked late, and had earned only six points out of
twenty. But nothing seemed the least bit incriminating.
   O'Brien told the reporters to leave. Robert's personal belongings were
placed in several cardboard boxes and put aside as government evidence.
It was indeed an investigation turned on its head. There was little doubt
who had committed the act; the question was whether the act was a
crime. But that wasn't O'Brien's territory. It was something for the
Justice Department attorneys to worry about. O'Brien's task was to
gather as much evidence as he could.
   An obvious and crucial witness was the person who had placed the
anonymous call to The New York Times. At first, O'Brien was sure it was
one of Morris's officemates. Most of them, it turned out, knew little
more than that Morris was quiet and kept to himself.
   O'Brien's first helpful interview was with Dawson Dean. The nervous
graduate student recounted his trip to Cambridge with the suspect. He
told O'Brien about the passwords he had seen on Morris's screen shortly
after their return, and how he had seen Morris seated at his terminal late
Wednesday night, in the midst of a telephone conversation. Dean said
he was surprised; it's difficult to carryon a telephone conversation and
program at the same time. But it was apparent that Morris was conferring
with someone about a program he was either running or still working
on. Dean heard Morris mention Harvard, then MIT. As Dean turned to
leave, Morris waved good-bye.
   O'Brien asked Dean one last question: "Did you call The New York
   "No!" Dean exclaimed.
   One officemate remembered that Morris didn't seem to be around
Upson Hall on Thursday afternoon. He also remembered there was a
note in large letters scribbled on the office blackboard: "ROBERT: CALL
   The officemate didn't know who Paul was.
   O'Brien's next interview was with Kevin Asplen, the graduate student
who drove Morris home on Thursday morning. Asplen had arrived at
Upson Hall at about 1:00 A. M., he told the agent, planning to spend
just a few minutes sending an electronic mail message to someone. He
saw Robert Morris sitting in his office and told him that if he would like
a ride home, Asplen would be ready in five minutes. Morris said he
would appreciate a ride and sat down to wait. Then Asplen saw that
there seemed to be something wrong with the computer he was using. It
                                                           RTH ... 317
was crashing frequently for no apparent reason. He couldn't get it to stay
running long enough to get his mail out. He told Morris he was having
problems with the computer and it might take him a while longer than
he had thought. Morris didn't offer any advice. He just said he didn't
mind waiting. At 2:00 A.M. Asplen finally got his message out and took
Morris home. During the ride, Morris was his usual quiet self and didn't
say a word about the virus.
                                 ... ... ...
Tom Guidoboni arrived home in Arlington, Virginia, late Friday night,
exhausted after a two-day trip to Texas, where he had been taking
depositions in a new case. When the forty-one-year-old defense lawyer
got up the next morning, his wife was absorbed in a newspaper story
about a virus that had crippled computers around the nation. "This is
fascinating," she said, pushing the article toward her husband.
   It was the first Guidoboni had heard of the incident. "What's so
fascinating about it?" He took a quick look and went back to his coffee.
   "Well," explained his wife, also a lawyer, "whenever I call the court
to ask for something they always say they can't answer because the
computer is down. Or you get a computer on the telephone. Now thou-
sands of computers are down because of this one thing." But her husband
was already lost in another section of the paper.
   Two hours later, the telephone in the Guidoboni house rang. It was
a partner from his firm, wanting to know if Guidoboni would be inter-
ested in representing "the virus kid." An hour later Guidoboni was on
the phone with Bob Morris, and they arranged an appointment for 10:00
A.M. Monday.

    It wasn't clear to Guidoboni which law, if any, had been broken.
Normally, if he got a weekend call on a criminal case, he knew instantly
which section of which statute applied to the case. But this was new
terrain for Guidoboni. He was familiar with basic wire-fraud law, which
was the first thing to come to mind after the conversation with his
partner. Even as a criminal lawyer, he was only vaguely familiar with the
existing computer crime statute. The next day, Guidoboni went to his
office to read the law carefully and prepare for the meeting.
    The 1984 law, amended in 1986, was created in response to the
public's demand for more computer crime protection. It was Congress's
first attempt to criminalize computer trespass, making it a crime to gain
unauthorized access to computer systems. The statute also made it a
crime to modify, destroy or disclose information gained from the unau-
318   ~   CYG6RPUNK
thorized entry into a computer. As of 1988, the amended law was still
largely untested. Only one case had gone to trial, none to a jury. But
while other areas of criminal law tended to address narrowly defined,
well-understood conduct, the computer fraud statute cast its net over a
wide range of computer activities. The law's critics claimed that it was
too broad and too vague, that it failed to define adequately such words
as access and authorized. But Guidoboni, new to the subject, understood
only that he would have to meet his prospective client and hear the facts
before he could consider how the law might be applied.
   The first thing that struck Guidoboni when the Morris family entered
his office was their unusual appearance. Robert, dressed in a navy jacket
and narrow purple flowered tie, appeared to be in a state of shock. Pale
and drawn, he looked as if he hadn't slept or eaten in several days. Bob's
eccentricities were evident from his long and unkempt beard. And
Anne, heavyset and with graying blond hair piled into a bun, came
across as a forceful, educated woman concerned about her son. In con-
trast to Robert, the parents appeared to be relatively composed, although
anxious about whether or not this stranger could help.
   Bob and Anne made it clear that they were doing a little hasty attor-
ney shopping. One prominent Washington attorney had already offered
to take on the case pro bono, but the Morrises wanted to explore all their
options. Guidoboni took down some general information and asked Bob
to explain his obligations to the National Security Agency with regard
to the incident. The elder Morris responded that he felt obligated to tell
law-enforcement officials what he knew of the offense. Morris had al-
ready contacted the lawyers at NSA, and he had already spoken with
the FBI. He said that he believed Robert should speak with the FBI as
well. It is a common, but often misguided, impulse to go straight to the
authorities with the whole truth, hoping for leniency in return, Guido-
boni explained. He was quick to dissuade the elder Morris from that
   The Morris family then interviewed Guidoboni. He told them he had
attended the University of Virginia Law School, that he specialized in
white-collar crime, and that with the exception of one computer-related
case in 1981, his background in computers was minimal. They discussed
his fees.
   The lawyer then excused both parents in order to have a private
conversation with Robert. In some ways, Robert struck Guidoboni as
the youngest twenty-two-year-old he had ever met. Once the lawyer
assured Robert that federal marshals weren't about to swoop down and
                                                          R1H    T    319

carry him away, he relaxed a bit. Guidoboni had spent years learning to
gauge potential clients-from street criminals to congressmen to corpo-
rate executives-and he could see through the overwrought facade to a
young man who was not only very smart but utterly without guile.
Robert's main concern appeared to be whether he would get into trouble
with Cornell.
   Once Guidoboni established that there didn't appear to be any na-
tional security risk involved, and that Robert's program hadn't affected
any financial institutions, Guidoboni asked the young man to tell him
who else knew firsthand that he had written and released the virus, and
what they knew. Robert said that Paul Graham and Andy Sudduth knew
the most about it. Then there were others whom he had told afterward:
David Hendler, Janet and, of course, his father. Paul was the only one
who knew about it beforehand. Guidoboni also wanted to know what
tracks he had left. Robert said he had "cleaned up" a lot of files on his
Cornell account, but someone there had obviously gone into his old files
and found incriminating evidence. Guidoboni said that he considered
this an indefensible means of collecting evidence. Cornell's claim that it
had a right to search Robert's files because it owned the computers was,
in the lawyer's view, preposterous. Scouring someone's personal com-
puter files is as much a privacy breach as opening someone's personal
mail, or listening in on private telephone calls.
   As the conversation wore on, Guidoboni saw that Robert was only
gradually beginning to see that this was a serious criminal offense. Until
now he hadn't even realized that anybody outside the computer com-
munity would pay much attention to the incident. It appeared likely to
Guidoboni, in fact, that until Robert walked into the lawyer's office, he
didn't know there was even a law to violate-and he certainly wasn't
aware of the computer crime statute in particular.
   Guidoboni wanted the case. He knew it would be interesting and
challenging. It was a rare opportunity to take on a piece of the law that
had never been interpreted. He liked Robert, who was clearly intelligent
without being arrogant. And the publicity couldn't hurt. The next day,
he got a call from Bob Morris asking him to proceed.
   The session with the lawyer had taken its toll on Robert. When the
family got back to the Metro station afterward, Robert collapsed. His
anxious parents propped him up on a bench and waited for him to get
his strength back before continuing the journey. When the Morrises
returned home later that afternoon, a crowd of reporters, photographers
and television crews was waiting at the end of the driveway. Robert put


      320   ~   CygfRPUNK
      his head in his hands. At first it appeared to the journalists that the
      computer whiz might make a statement. Instead, he looked down at his
      feet, admitted in a barely audible voice that it had been a pretty rough
      day and was escorted by his mother into the house.

      Later in the week, Paul and Andy drove down from Cambridge to meet
      with Guidoboni. Robert was glad to see them, but at the lawyer's instruc-
      tions he wasn't talking about any specifics surrounding the incident, in
      case they were ever called on to testify. Conversations had to remain
      general, and Robert stayed pretty quiet, yet he was naturally curious
      about the reaction among people he knew at Harvard, and he found it
      difficult to refrain from asking his friends about it. Telling him not to
      discuss the case was like telling him not to think about the color red.
         It was clear to Robert's two friends that the incident and the attention
      being heaped on him had made him shier than ever. When he called
      Guidoboni's office one morning, he hung up after one unsuccessful at-
         "What happened?" Andy asked.
         "I guess he's not there," Robert responded.
         "What did they say?"
         "A secretary just said he's not there," Robert said.
         "Of course she said thad" Andy said. "They're screening calls from
      the press. Did you tell her your name?"
         Robert shook his head.
         "Call back and say who you are. I'll bet he's there this time."
         Andy was right.

      Among computer scientists, the incident set off a debate that was to last
      for months. Within days, analyses of the program were under way. Three
      lengthy scientific papers emerged. The Association for Computing Ma-
      chinery's technical journal, Communications of the ACM, devoted an
      entire issue to an examination of the program. The opinions of critics
      ranged from "mediocre and sloppy" to "brilliant." Some people said it
      had caused damage that could reach into the millions of dollars. One
      industry group estimated the damage at $96 million. Others said it had
      done no damage at all, that its effect was just the opposite: it had alerted
      the industry to security flaws.
         Another topic of debate was the question of how widely the rogue
                                                          RTH    T   321
program had actually spread. Estimates from MIT put the total number
of computers infected at about six thousand, or about 10 percent of the
number of computers on the Internet in 1988. But other evidence sug-
gested that the number may have been much higher. A conclusive
number was never reached.
   Among the semanticists, a debate erupted over whether the program
was a virus or a worm. The general consensus was that even though
Morris and his friends called the program a virus, the program more
closely resembled a computer worm because worms can move under their
own power while a virus piggybacks parasitically on another program. In
the biological world, a virus is an agent of infection that can grow and
reproduce only within a host cell. In strictly technical terms, so the
argument went, since the program did not need to attach itself to an-
other "host" program in order to propagate, nor did it alter or destroy
any programs, it was more like a worm. A number of computer research-
ers disagreed. Mark Eichin and Jon Rochlis, who battled the program
when it hit MIT, wrote a paper declaring that "virus" was a more accu-
rate description of the program because it was a closer biological analogy
to what the program actually did. But in the end, the "worm" label
   One of the few computer scientists whose career was actually boosted
by the worm incident was a Purdue University professor named Eugene
Spafford. Before the worm, he was a softwareengineer. After the worm,
he switched to computer ethics and began to travel around the country
making speeches on the topic of computer viruses, computer security and
the moral imperatives he perceived. He argued that there was no excuse
for what Robert Morris had done, that it violated the spirit of a com-
munity built on trust and that trespassing is trespassing, whether it is
done physically or electronically.
   The ethical controversy wasn't confined to computer scientists. It also
engaged people who knew nothing about computers but who were wor-
ried about how this new technology could be used for criminal ends. The
fact is that virtually everyone has been victimized by a computer in some
way or other, either because a bank deposits money in the wrong ac-
count, an insurance company loses a claim or an airline reservation is
booked on the wrong flight. What people often don't take into account
is that it isn't the computers making the mistake but the people who
program them and work with them. Nonetheless, some argued, if a
suitable punishment wasn't dispensed to Robert Morris, wouldn't com-
puter criminals with far more malicious intent than Morris think they
322   A   CygERPUNK
had an open invitation to set off similar programs that could topple
banks, sabotage air-traffic control systems and perhaps even start wars?
   Robert's motive became another hot topic of debate. Since he refused
to speak with the press, journalists speculated as they pleased. Newspaper
reports and magazine articles suggested that he got his inspiration from
The Shockwave Rider, John Brunner's proto-cyberpunk novel, which
Anne Morris told reporters was among the most frequently read books
on his bookshelf. But chances are that Bob Morris was more taken with
the story of the shockwave rider than his son was. Bob was an early
admirer of Brunner, and when Andy and Paul went to Maryland, one of
the first things Bob did was pull the book from a shelf with the words,
"Let's see where this all started." Robert had read the book, of course.
He had even read it more than once. But he didn't feel that it had
influenced his life any more than did any other book he enjoyed. More-
over, in planning his program, he called it a virus not a worm.
   Others imputed more sinister motives to Robert. The program, they
argued, was a hostile act, intended to bring the network to its knees.
Some even came to suspect that there was a deeper message buried in
the program. Perhaps this was a conspiracy. After all, wasn't he the son
of one of the nation's leading computer security experts, an official at
the secretive National Security Agency, America's computer spy orga-
nization? Might this be some sort of NSA-inspired exercise gone awry?
The truth, of course, was far less dramatic. Robert had never pried his
way into commercial computer networks in search of power, money or
state secrets. He was simply carrying on an intellectual tradition he had
learned from his father.

Slowly, with the help of Dean Krafft, the patient and helpful facilities
manager, FBI agent O'Brien was catching on to the computer talk every-
one at Cornell was throwing around. It gradually dawned on O'Brien
that the real evidence in this case wasn't to be found among the suspect's
personal effects. It was all on the computer. But one of the problems, he
gathered, was that Morris had encrypted nearly all of his files-and not
only had he encrypted them, but with few exceptions he had also com-
pressed them first-he shrank them by running them through a special
program. That made them even more difficult to decipher. By one esti-
mate O'Brien heard, it would take scientists at the National Security
Agency, the center of cryptography, two hundred years to decrypt Mor-
ris's files. The information O'Brien got from his sources turned out to be
                                                         RTH    T   323

out of date. Krafft's first step was to get Cryptbreaker's Workbench, a
program to break encryptions. With the help of the program, Krafft was
able to break the encryption and give Cornell access to all versions of
the Morris worm. The process took less than a day.
   The matter of the anonymous call to The New York Times was also
becoming clearer. The FBI had obtained Robert's long-distance tele-
phone record, and it appeared that that night he had called someone at
Harvard named Paul Graham.
   Special agent O'Brien couldn't muster much sympathy for this Morris
fellow, even if he was as brilliant as people said. All that talk of an
innocent backpack-carrying kid who made a mistake didn't move
O'Brien. At twenty-two, the agent figured, a man should be considered
a responsible adult. At that age O'Brien was already a full-time parole
officer putting himself through graduate school. And he didn't think
much of Morris's Harvard friends, either. When he flew to Boston soon
after the incident to interview Paul Graham and Andy Sudduth, he
found Graham first, in his office at the Aiken computer center. When
O'Brien introduced himself, Graham was unfriendly and uncooperative.
    "I don't think I want to get involved in this," he told the agent.
    "But you are involved, Paul," O'Brien responded. "You dropped a
dime on your buddy."
    Paul stood his ground. "Well, I'd rather not talk to you."
    O'Brien nodded, reached into his pocket and produced a piece of
paper. "Paul," he said, "would you mind telling me if! spelled your name
right?" He handed Graham the piece of paper.
    "What's that?" Graham asked as he took the paper from O'Brien. His
eyes widened. It was a subpoena summoning Paul to Syracuse to testify
before a federal grand jury. O'Brien had to smile at the young man's
ignorance. Most people to whom he served subpoenas knew what was
coming the instant O'Brien reached in his pocket. By reflex, they would
put both hands over their heads and take a step backward to avoid being
    "It's a subpoena, and we'll see you in Syracuse next week."
    "And what if I don't come?" Graham sounded irritated.
    "Then I'll be back with an arrest warrant. "
    Andy Sudduth was more gracious, but he too declined to be inter-
viewed. He was served a subpoena as well.
    On the morning of the grand jury testimony, Sudduth and Graham
were late. The assistant U.S. attorney, Andy Baxter, began to show
concern. "What time does their flight get in?" he asked O'Brien.
324   •   CygE~PUNK

   "What flight?" O'Brien asked. "I heard they're driving."
   The prosecutor looked confused. "It's a six-hour drive! You didn't tell
them that they can fly and the government reimburses them?"
   "Gee, that must have slipped my mind," O'Brien said, barely con-
cealing his lack of sympathy for these young men.
   When the two witnesses did arrive, they had to spend some time
warming up. The heater in Paul's car was broken. They had stopped
several times along the way to warm their hands, and once to buy Andy
a warmer jacket. When they heard that the government would have paid
their airfare, both were furious.
   In December, Robert agreed to give a proffer, a statement to the
government detailing everything he had done. It was an unusual conces-
sion from someone suspected of committing a crime, but Robert and
Guidoboni thought that full cooperation might soften the Justice De-
partment. O'Brien and Gibbons were ready for him. One of the bureau's
interrogation experts flew to Syracuse beforehand to coach the two
agents on effective questioning techniques.
   O'Brien had already seen a videotape of the lecture Robert Morris had
given half a year earlier to the NSA on computer security. Part of the
lecture had been called, "How Not to Get Caught."
   "Isn't it kind of strange," O'Brien asked the suspect, "that you're
giving a talk like that on how not to get caught and you're sitting here
right now?"
   Robert looked down at his hands and smiled. He had no answer.
   Shortly before Christmas, Guidoboni called Alan Rubin, the defense
attorney in Los Angeles representing Kevin Mitnick, who had been
arrested earlier in the month for breaking into Digital computers and
stealing proprietary software. The lawyers didn't know each other and
their clients were clearly spun of different cloth. Yet there was so little
existing case law for the computer crime statute that Guidoboni thought
he and Rubin might be able to exchange a few ideas. As it turned out,
the two attorneys could only commiserate and wish each other luck.
   With just a handful of keystrokes, Robert not only had paralyzed
thousands of computers but had brought his life to a standstill as well.
Returning to Cornell, continuing to attend classes and starting work on
his dissertation were, of course, out of the question. By Thanksgiving,
Robert had withdrawn from Cornell, his career as a computer scientist
on hold indefinitely. Between Christmas and New Year's, Bob and Anne
took him to Cornell to pack up his things. He moved in with his parents
and got a programming job at a private international development
                                                          RTH " 325
agency, accompanying his mother on her three-hour daily commute into
Washington. Cornell, in the meantime, in a measure to guard against
civil lawsuits from victims of the worm, embarked on an extensive in-
quiry. By February, the university had issued a forty-five-page report on
the incident, which concluded that Robert Morris had violated the
university's computer-use policy. As if that weren't enough, Robert had
to go through an academic disciplinary hearing. Cornell formally sus-
pended Robert. He would be eligible to reapply for admission in the fall
of 1990.
   In an aside during the academic hearing, Dean Krafft asked Robert
about the keyword he had used to encrypt his files. "It's in the dictio-
nary," Robert told him. Krafft ran the dictionary through the encryption
mechanism, compared the output and found the word. It was simple.

The U.S. attorney in Syracuse was apparently willing to send a recom-
mendation to Washington that Robert be charged with a misdemeanor,
but Guidoboni heard nothing from prosecutors directly. The Justice De-
partment was maintaining an ominous silence in the matter, placing the
case in a tense state of suspended animation. Here was someone who
had supposedly committed the computer crime of the decade, and he
had yet to be charged with it. Still more ominous for the defense, in the
spring of 1989 the case was lifted from the Syracuse prosecutor's hands
and given to prosecutors in Washington. The waiting game continued.
Beside himself with boredom and loneliness in Maryland, Robert moved
to Cambridge and started working at a small software company run by a
couple of his old friends from Aiken, then switched to a programming
job in Harvard's classics department. Robert had yet to utter one word
in public about what he had done, about his reasons or his motives, yet
in some circles he had acquired folk-hero status. One young hacker had
Robert's photograph taped to his bedroom wall; others were choosing
rtm as their password.
   In June, seven months after the incident, Mark Rasch, the Justice
Department's young computer crime expert, called Guidoboni to intro-
duce himself and say that the Justice Department planned to indict
Robert on one felony count. If Robert agreed to plead guilty, Rasch said,
the Justice Department would consider granting some concessions on the
   In one last attempt to reconcile the matter outside of court, Guido-
boni had a meeting with Edward Dennis, the assistant attorney general
326   A   CygERPUNK
in charge of the criminal division of the Justice Department. But Gui-
doboni's pitch-that his client had intended no fraud and should be
treated lightly-failed to convince the prosecutors. After the meeting
with Guidoboni, it still took Dennis some time to make the decision to
proceed with the indictment. Robert, who would have pleaded guilty to
a misdemeanor, decided he would rather take the case to trial than plead
guilty to a felony.
    The whole family was affected by the long, tense wait. Exhibiting his
own strange way of coping, Bob decided to take up a hobby of his
daughter Meredith's that he had always found intriguing and signed up
for a church-tower bell-ringing class in Washington. The class met twice
a week and demanded that Bob spend even more time away from home
than he already did. It irked Anne, who had quit her job in order to
work from home and hold things together. She was particularly edgy,
still stinging from what she considered the Justice Department's targeting
of her son when there were so many true criminals to worry about. All
her children, and Robert in particular, had grown up understanding that
you simply don't hurt people. Those who portrayed Robert as a malicious
felon didn't understand that it was antithetical to Robert's moral makeup
to do anything harmful. For his part, Robert remained quiet, asking his
parents for very little. When Anne asked him if he wanted the family to
be at the trial, he said there was no need. Anne told him everyone
would be there anyway.

The question of Robert's intent was of pivotal importance from the start
of the legal proceedings. With few exceptions, the computer science
community agreed that Robert had intended no harm. Even a cursory
look at his program was enough to show that he had designed the worm
to be as innocuous as possible while simply taking up residence inside as
many machines as possible. Moreover, he had included mechanisms in
the program to limit its growth. In Guidoboni's view, this was a strongly
mitigating point and could even win him the case. He argued that the
section of the statute under which Morris was being tried was for people
with malicious intent and not for someone like his client.
   This was a cornerstone of the Morris defense. Guidoboni's strategy
was to paint Robert as a well-meaning innocent who got caught up in
what he intended as a harmless experiment that exploded in his hands.
No jury, if allowed to consider the intentions of this young man, could
conclude that Robert Morris had meant to cause any harm. As far as
                                                          RTH   T    327

Guidoboni was concerned, it might even be enough for the judge to
dismiss the case entirely.
   So Guidoboni filed a motion to dismiss the case, arguing that the law
required proof of intent to prevent authorized use and cause damage.
Obviously, he argued, there was no such proof. In October 1989, Gui-
doboni went to Syracuse and argued for dismissal in front of the judge,
but the judge denied the motion. At that point, Guidoboni saw his case
weaken. The trial was scheduled for late November.
   Two weeks before the trial, the prosecutors surprised the defense with
a list of a dozen or so extra witnesses whom they planned to call. Then
came another surprise.
   Just how the Justice Department came into possession of the videotape
of Robert's NSA lecture wasn't precisely clear, but it appeared that the
agency had alerted Justice Department attorneys to its existence. Robert
had already told Guidoboni about the lecture. He explained that his
father had originally been asked to give the talk but had recommended
Robert instead. Robert told Guidoboni that the speech hadn't gone
particularly well and he was embarrassed about it. The following day's
lecture to the Navy, he told the lawyer, was much better.
   The videotape showed an extremely shy Robert in a blue shirt and
blue jeans, his left hand wedged into his pocket and his eyes focused on
some middle distance. For more than an hour, he rambled on about
some of UNIX's biggest weaknesses, system managers' carelessness,
sloppy user habits. Desperate, it seemed, for something to do with his
eyes, he referred frequently to handwritten notes, but the information
seemed to be so familiar he clearly could have done without them. It
would have taken someone with a particular interest in UNIX security
to stay alert through this talk.
   Under any other circumstances this would have been regarded as an
endearing if painfully awkward speech by a budding computer scientist.
But as soon as the same shy young man was the subject of a federal
indictment, the speech became an open window into the defendant's
frame of mind. Did this Harvard student think like a computer criminal?
A careful look at the videotape and the meticulously rendered transcript
that accompanied it revealed that there were indeed ways in which this
twenty-two-year-old appeared to be able to get inside the mind of a
computer criminal. Moreover, he had a clear knowledge that certain
uses of a computer were illegal. In one particularly revealing moment, in
which he discussed "how not to get caught," Robert started out by
saying, "Close to the heart of every hacker out there has got to be how
328   •   Qyg£RPUNK
to stay out of jail." He went on to list some of a computer criminal's
most useful tricks: covering his tracks, laundering calls and never return-
ing to the scene of a break-in. This was clearly someone who knew
whereof he spoke. And if there was doubt in anyone's mind about
whether Robert Morris was aware of the number of computers connected
to one another over the network, such doubt could be dispelled by the
second line of his speech: "There are thousands and thousands of UNIX
systems out there," he said.
   It's likely that a screening of the entire hour-long videotape would
have just caused the jurors to doze off. But any juror listening closely
couldn't have helped but notice some of the language Robert chose. Far
from distancing himself from computer criminals, he fairly took on a
hacker's way of thinking. "If you have foresight," he said on the matter
of outsmarting auditing mechanisms, "the right thing to do is to fill up
the disk which the audit trails reside on before you go doing bad things,
so that when programs go to log something, they find the disk is full and
nothing gets audited anymore."
   The next thing Guidoboni knew, the videotape was on the prosecu-
tion's list of exhibits. More unnerving, the prosecutors had selected to
view only the "How Not to Get Caught" portion of the tape.
   Guidoboni's first move was to call in Bob Morris and question him on
the lecture.
   "Whose idea was the lecture?" the lawyer asked.
   "Mine," Morris answered.
   "And who decided which topics should be covered?"
   After a few moments of considering the question, Morris said he
couldn't answer that without disclosing classified information.
   Guidoboni tried another question. "Who attended the lecture?"
   Morris shook his head. The answer to that one would reveal classified
information, too.
   "Were there any topics of particular interest to the audience?"
   "I can't answer that one either," Morris said.
   Guidoboni's hunch was that NSA did its share of breaking into com-
puters, or at least learned the mechanisms by which break-ins are carried
out. The defense attorney wanted the trial to reveal just who chose his
client's speech topics, and why.
   Guidoboni told the prosecutors that if they planned to show the tape,
then they had better show the whole thing or he would file an objection,
primarily on the grounds that the portion they showed had been lifted
out of context. Further, if the tape was to become a piece of evidence,
                                                           RTH    T   329

then Guidoboni was going to put Bob Morris on the stand and ask him
to disclose classified information under oath.
   The matter of the videotape remained unresolved until shortly before
the trial, when, as quietly as it had appeared on the list of exhibits, it
disappeared. Guidoboni could celebrate a small victory.
   The trial was rescheduled for January 1990.

Syracuse, a city of about 170,000 in upstate New York, is hardly a high-
tech mecca. Its largest employers include a Carrier Corporation air-
conditioner plant and Syracuse University. Prior to the Morris case, the
crimes that made the news in Syracuse didn't involve computers. In
December of 1989, local citizens were captivated by headlines of serial
prostitute murders up the road in Rochester and an In Cold Blood-style
murder in nearby Dryden. These came on the heels of the dramatic trial
of former Syracuse mayor Lee Alexander, who was convicted in 1988 of
accepting kickbacks. The Robert Morris trial might have played to a
more suitable audience in Silicon Valley. On the other hand, Syracuse
jurors would be a realistic cross section of average Americans whose lives
had yet to be touched by technology any more directly than by super-
market scanners and digitized directory assistance.
    Even if the locals weren't riveted by the case, the press turned out in
force. Local newspapers had already heralded the start of "the hacker
trial. " Television cameras were stationed outside the building, waiting
for the family to appear. Sequestered with relatives in a Syracuse suburb,
the Morris family guarded its whereabouts closely.
    The press showed up because the case tapped America's ambivalent
feelings about the power and reach of computers. It had somehow come
to symbolize the nation's collective anxiety about computer hackers and
the threat they might pose. And the case had family drama: the father
and son who belonged to a computer science elite were both obsessed
with exploring the subtle intricacies of the complex computers that had
come to control much of society.
    The case also marked the sudden awakening of a national recognition
of the fragility of tens of thousands of interconnected computers. The
havoc wrought by Morris's program symbolized as never before the na-
tion's increasing dependence on computers, and the increasing vulnera-
bility of those computers. As computers became more and more
interwoven, each link interconnected and interdependent on hundreds
of others, and as these computers became accessible to more and more
330   A   CYEfRPUNK
people, it was inevitable that something like the Morris program would
come along. Still, it was a surprise when it came.
   The two attorneys representing the Justice Department discharged
their duties with proficiency. Mark Rasch, a short man with sharp,
handsome features who was now the Justice Department's star computer
crime prosecutor, wasn't the slightest bit flustered. His command of the
complex technology under discussion was so impressive that it gave him
an air of extra authority. His partner, Ellen Meltzer, at thirty-seven a
Justice Department veteran, was actually senior to Rasch, but that
wasn't altogether clear in the courtroom. Meltzer deferred to Rasch on
most technical points.
   Even before the trial opened, Rasch and Meltzer filed a motion asking
the judge to dismiss as irrelevant any evidence involving Morris's intent.
"The evidence of lack of intent to cause loss is simply not relevant to
any issue in this case," Rasch declared in his motion. According to the
law, Rasch said, the prosecution needed only prove that Morris intended
to break in and that damage resulted from the worm, not that Morris
intended to cause damage.
   Jury selection went swiftly enough. Both the prosecution and the
defense were looking for one important criterion: none of the jurors
should have so much as a fleeting knowledge of computers. All prospec-
tive jurors were asked if they owned a personal computer, if they knew
anything about computers. "I don't know a thing," replied one woman,
and she was promptly seated on the panel. Only three prospective jurors
owned computers, and all three were rejected by either the prosecution
or the defense. One man who knew nothing about computers himself
but whose son worked at IBM was excused. In the end, two women who
used computers in their jobs as clerks were seated on the panel.

The first and most striking impression Robert Morris made in the court-
room was that of an unlikely felon. Traditional white-collar criminals
felled by their own greed or malice were usually much older than this
defendant. He was pale and thin, and his new suit seemed a little too
large. It was hard not to wonder whether the suit was an imperfect fit
intentionally, a subtle message from his defense team: If he's not even
old enough to know how to pick a suit, how much of a white-collar
criminal could he be? His seeming lack of worldliness was what the
defense attorneys wanted to convey to the jury of nine women and three
                                                           RTH    T   331

men, many of them with children the defendant's age. In court, his
attorneys referred to him as "Robert," a "kid."
   Howard G. Munson, the federal district judge in the case, was a sixty-
five-year-old Republican whose fourteen years on the federal bench had
accrued a record of fairness and neutrality in criminal cases. A heavy
smoker with a resounding cough and thunderous voice, Munson had a
sharp memory for the details of a case; at the same time, he possessed a
knack for extracting the essence from a witness's testimony.
   In his opening statement to the jurors, prosecutor Mark Rasch told
them they would be hearing some difficult technical language, but that
shouldn't distract them from what the government intended to prove:
that on November 2, 1988, there was a full-scale assault on computers
throughout the United States launched by the defendant, Robert T ap-
pan Morris. Rasch pronounced Robert's middle name "Tap-in." "This
assault was deliberate, planned, calculated," Rasch said. "It was calcu-
lated to break into as many different computers as he possibly could to
gain what the law will call 'unauthorized access.' "
    Rasch went on: "What the government intends to prove is that early
in the evening of November 2, 1988, computer scientists throughout the
country started noticing that something was going drastically wrong with
computers from California to Massachusetts to Florida. All of the com-
puters were starting to slow down. These were computers not just at
government sites, not just at military sites, but at commercial facilities,
at private companies throughout the country, and many of the people
we'll hear testimony from worked at different universities doing scientific
research. Their research was interrupted. They couldn't do their work
because of the actions of the defendant, Robert Tap-in Morris. Valuable
computer time was lost. Valuable experiments were lost. People could
not communicate with each other. They couldn't talk to each other.
They couldn't find out what was going on. Their computers, and you
will hear testimony about this, were crashing, stopped running, came to
a grinding halt because of the actions of the defendant, Robert Tap-in
    As Rasch described it, Morris had sat for hours at a computer terminal
in Upson Hall at Cornell, planning "the break-in." "It was deliberate."
Rasch repeated the word deliberate like a mantra. "As the time for release
of this worm came nearer, he spent more and more time on the computer
trying to perfect it. He worked all day on the day of November 2 trying
 to launch this worm. It was designed deliberately and intentionally."
332 •      CygERPUNK
The prosecutor stopped just short of voicing his private theory that
Robert had actually panicked upon seeing that the ftp bug had been fixed
and that he rushed to get the worm finished that day before other holes
were plugged as well. Moreover, he believed that Robert's rush at the
end accounted for some of the careless code in the program.
    Rasch told the jury that he intended to call a collection of witnesses
into the courtroom, victims from computer centers throughout the na-
tion who would describe the horrendous events of November 2 and
estimate the dollars lost in the effort to fix their systems. "You will hear
testimony from the scientists, the engineers, who use these computers
that when this thing was attacking them, they didn't know what it was
doing. It was designed to hide its effects, to pretend to be an innocuous
program. It disguised itself to frustrate the victims of its attack."
    In explaining what a computer virus was, Rasch drew parallels to the
biological flus the jurors could all relate to. "If you have just one virus
you may not get very sick. But if you get many viruses, if you have
hundreds, you will get very sick. Not only will you get very sick, you will
get other people sick." Not only was Rasch comfortable with most of the
computer arcana that was tossed around in the courtroom, but he did his
share of tossing. And even though Guidoboni had two computer experts
at his disposal throughout the lengthy preparation for the trial, he was
still struggling a bit with some of the concepts. If the case was to be
decided on pure technical machismo, Rasch would undoubtedly win.
    The prosecutor reminded the jurors that the government did not
intend to prove that Robert was an evil person, and that it did not need
to prove that Robert intended to cause the damage he caused. For the
jury to return a guilty verdict, Rasch said, the government needed only
to prove that he intended to obtain unauthorized access and break in.
    In contrast to Rasch's clipped and formal language, Guidoboni
adopted a folksy tone. "I guess you can tell from looking at me that I am
not Perry Mason," Guidoboni said. But where Rasch was self-assured,
Guidoboni was almost apologetic. "I don't have a rabbit to reach down
here and pull out of my hat," he said. "That's not the way criminal trials
work in real life, so you're going to have to bear with me. We are going
to do the best that we can." A few minutes into his speech, in fact,
Guidoboni made Robert's confession for him. He acknowledged that his
client had indeed written the worm. However, he added, evidence
would show that "this worm caused no permanent damage and it was
not designed to cause permanent damage. It didn't break any machines.
                                                             RTH    'f'   333

The virus didn't read anybody's private files and didn't steal any infor-
mation and didn't put one dollar into Mr. Morris's pocket."
    The defense attorney then listed Robert's credentials. Not only did
he work hard at summer jobs while at Harvard, but he had made impor-
tant contributions to the field of computer security. He had written
papers that were published, and he had alerted the computer community
to security holes in the past.
    Guidoboni depicted the Internet as a loosely organized, almost ragtag
collection of networks. "There was no drug czar, if you will, computer
czar sitting up in Washington who has control, who established rules."
Moreover, he said, the Internet is not a network that launches missiles.
It's used for conducting research and sharing scientific data. It's also used
for frivolous things like playing chess, sending love letters and exchang-
ing recipes.
    Guidoboni's speech had an odd air of desperation to it. He wasn't
asking so much for a judgment of innocence as for forgiveness.
    Guidoboni said he would present evidence showing that Robert's virus
was purposely limited, its intention to spread slowly and quietly, "but
you will hear that he made a mistake, a critical mistake." And in the
end, "he alerted the computer community that the system wasn't secure
and that they needed to take steps to fix it. This in turn caused a lot of
embarrassment. But a simple mistake, together with embarrassment and
some inconvenience, are not the equivalent of a federal felony offense."
    Witnesses were what Mark Rasch promised and witnesses were what
he delivered. The first witness for the prosecution was Dean Krafft,
Cornell's director of computing facilities. Krafft had been pacing outside
the courtroom, wondering aloud to another witness how he would be
able to explain data decryption to the jury. Krafft described to a mystified
jury the process of sorting through Morris's old files and finding early
versions of the program. If nothing else, the jury got from Krafft's testi-
mony that this was indeed to be a trial filled with technical terms.
    Many of the witnesses for the prosecution knew each other, if not
personally then by electronic mail, and if not by electronic mail then by
reputation. Rasch and Meltzer had selected their witnesses carefully.
They represented a good cross section of the nation's computer system
managers. They traveled to Syracuse from universities, Army research
labs and other government institutions. Local reporters joked that they
were disappointed to see the witnesses weren't wearing pocket protec-
tors, which they believed to be a computer nerd's signal appurtenance.

        334   A   CygERPUNK
            Mark Brown, the University of Southern California system manager
        who had battled Kevin Mitnick, was the second witness. He gave the
        first in what would become a blur of testimony about the devastation of
        November 2, 1988. Brown testified that when he logged in to his com-
        puter that night at about 11:00 P. M. California time, he saw dozens of
        strange programs running. Even after Brown killed them, they were back
        within minutes. The situation threw Brown and others in the computer
        center into a state of panic. "We'd never seen anything this widespread,"
        he told the court. "We had no idea if it was modifying the data on the
        files." So at 6:00 A. M., to prevent infections from coming in and going
        out, Brown disconnected the university's computers from the Internet.
        To prevent reinfection, Brown had to turn off the computers, halting
        file transfers, cutting off electronic mail and bringing all research to a
        halt. In the end, 350 Suns and VAXes on the USC campus were in-
            And so it went, in painstakingly systematic fashion. Rasch and
        Meltzer called computer managers to the stand, and one after another
        they described the mayhem that ensued after the worm hit their institu-
        tions. In endless litany, witnesses told of realizing they were under at-
        tack, of taking computers off the network, of crawling under their desks
        and pulling the computers' plugs. The prosecutors asked each witness
        how many hours he and his colleagues spent ridding his computers of the
        worm. Estimates of the damages to each institution varied widely, from
        $200 to $54,000, and the total climbed steadily until it reached
        $150,000. And how much was their time worth? If anything was to sink
        in with this jury, it was that people who work with computers make good
        money: one witness valued his time at $21 an hour, another at $22.38.
        Was Robert Tap-in Morris authorized to use those computers? No, came
        the replies.
            At times, the courtroom transformed into an introductory computer
        class, as witnesses attempted to explain complex and often baffling tech-
        nical terms to the jury. The prosecutors submitted as evidence dozens of
        documents containing page after page of abstruse computer code,
        prompting a rebuke from Judge Munson. "For those of us educated in
        the forties, most of this is totally incomprehensible," the judge told the
        attorneys. "I don't see what good it would do for the jurors."
            In his cross-examination of the prosecution's witnesses, Guidoboni
        had them state that the worm caused no damage. And he tried to
        establish that much of the time they devoted to working on the problem
        was spent in pure intellectual pursuit, analyzing the worm once it had
                                                            RTH    T   335

already been eradicated. The most brilliant moment for the defense came
in Guidoboni's cross-examination of Keith Bostic, the Berkeley system
manager in charge of Berkeley UNIX. The prosecutor had Bostic tell of
the catatonic state the Berkeley computers were in, and the exhausting
decompilation effort. But Guidoboni asked Bostic whether the worm
had in fact alerted the computer science community to security flaws in
sendmail and finger.
   "As a result of the worm," asked Guidoboni, "the sendmail program
was a little bit better, wasn't it?"
   "That is correct," Bostic responded.
   "And it certainly was more secure."
   "That is correct." By this time Bostic was smiling.
   "And the same thing for finger, is that not correct?"
   "That is correct," Bostic answered.
   As long as Guidoboni was on a roll, there was another matter he
wanted to clear up with Bostic. It was the question of authorized access.
Didn't programs such as finger and ftp extend to anyone who used them
a certain measure of access to all the computers on which the programs
ran? Bostic tilted his head as if this were the first time he had considered
the question. "Yes, it does," he answered. So might it be possible that,
in a sense, Robert Morris was authorized to use any computer that could
run those programs? Bostic agreed that he was.
   After each witness testified, reporters followed him outside for more
interrogation. Does Robert Morris deserve prison? "I certainly hope he
is convicted and sent to prison," said one angry computer manager.
"Otherwise, it will be open season on the whole network." The manager
from the Army research lab had a slightly different view: "It would be
bad to put someone like this in prison. He could learn things and be a
real threat."

Every day in court, Bob and Anne sat in the front row of the audience.
Anne kept a composition book propped on her lap and took occasional
notes. If anyone cut a conspicuous figure, it was the elder Morris. A
mischievous gleam in his eye was dampened by a trace of sadness. He
wore a gray three-piece suit and heavy black shoes bowed from wear. His
complexion was gray and he smoked constantly outside the courtroom
during recesses, enveloping whoever happened to be standing nearby in
a curtain of smoke.
   Over the three-day weekend, AT&T suffered a major failure in its
336   •   CYCfRPUNK
long-distance service. For much of Monday about half the long-distance,
international and toll-free 800 calls on the AT&T network didn't go
through. Rental car reservation clerks and other toll-free operators sat
idle. Television producers had trouble covering stories because they
couldn't reach their bureaus. The problem was traced to an AT&T
software glitch. A rumor circulated that a hacker in the New York area
had broken into AT&T computers there and caused the disruption as a
protest against Robert's prosecution. Mark Rasch even called colleagues
in Washington to see if it had indeed been the work of a hacker. In the
end, it turned out to be an error in the phone company's software. Yet
the incident again brought home the unexpected importance of comput-
ers in everyday life. Any member of the jury who had tried and failed to
make a long-distance call that day would have had a more vivid under-
standing of the Internet computer managers' frustration when faced with
Robert's worm. In the Morris defense camp there was concern that this
could make jurors less receptive to arguments in Robert's defense.
   By the second week of the trial, more supporters of the defendant
appeared in the courtroom. Peter Neumann, a security expert and former
Bell Labs researcher, came from California to testify for the defense.
Doug McIlroy was there as a character witness. Robert's friend Janet
stayed through most of the trial. Meredith and Ben had shown up in
time to see Robert testify. During the many long breaks, Janet, Robert,
Meredith and Ben sat hunched over the day's crossword puzzle in the
   That week, three longtime friends of the Morris family arrived-Ken
Thompson, Fred Grampp and Jim Reeds, all of whom also happened to
be pioneers in the field of computer security. They had flown from New
Jersey to Syracuse in a small private plane to do nothing more than sit
in the courtroom and show their support for the family in general and
the young defendant in particular. The defense already had more char-
acter witnesses than it could use, otherwise it would have called upon
these three. It was like a twenty-five-year reunion of computer security's
power hitters. Indeed, the scientists engaged in lively talks outside the
courtroom might have been mistaken for a group of technical conference
attendees on a coffee break. The Bell Labs scientists could recall the
Arpanet crash of October 1980, when a simple hardware failure on a
single computer brought the network to its knees. Of course, there were
only several hundred computers on the Arpanet then.
   When the Bell Labs trio arrived, Anne Morris began steering reporters
in their direction, letting the journalists know just what sort of eminent
                                                              RTH    T   337

scientists had taken the trouble to make the trip to Syracuse. In search
of fresh story angles after days of tedious technical testimony, the report-
ers clustered around Thompson, perhaps the best known among com-
puter scientists. His thinning brown hair brushing his shoulders,
Thompson looked the most like someone whose eccentricities were in-
dulged out of deference to his genius. Bob Morris stood by and watched
in amusement, twirling a cigarette in his fingers. When Thompson re-
marked that he thought the release of the worm was an irresponsible act,
Morris grinned and spoke up: "Have you ever done anything like that,
   It was peculiar that these of all people would be assembled in a court
of law in the first place. The group provided more than just a technical
reference point: they were the people behind the theoretical debates
over technological morality and intellectual property. Their presence
was a reminder of a long-standing tradition among computer security
experts that one earns one's stripes in the field by defeating a computer's
barriers. But that was in the old days, when the community was far
smaller, when computers weren't ubiquitous, when the stakes were much
lower, when breaching security was still considered to be all in the spirit
of science.
   Not the sort of people who allow idle time to slip by unproductively,
the Morris family came to court each day with books to keep them
occupied during recesses. The volumes they chose were light reading
only by their standards: Bob kept Suetonius's The Twelve Caesars at his
side, Robert had brought along a copy of Robert Graves's historical novel
Count Belisarius, and Meredith was reading Umberto Eco's lastest novel,
Foucault's Pendulum. Anne kept a careful watch on Bob, who was rather
enjoying the journalists' attention. She adopted a more studied approach
toward the assembled press: she spent many of the breaks quietly talking
to reporters of Robert's various virtues. His brilliance she illustrated with
a story about steamer trunks from Robert's childhood filled with back
issues of Scientific American. His sensitivity she certified by telling of the
fainting spell on the Washington Metro the day they first visited Gui-
    Paul Graham was put on the stand by the prosecution, and Andy
Sudduth was called by the defense. When Robert saw his two friends
enter the courtroom, he turned crimson, as if he were embarrassed for
all three of them. Testimony from the two Harvard students provided
some of the trial's few light moments. Paul was as animated as ever. He
recounted the conversations he had with Robert as the worm was in the
338   •   CygfRPUNK
planning stages, and the panicked call he got from his friend just after it
was clear that this would be no subtle experiment, but a monumental
    Paul and Andy brought to the trial a further air of elite institutions.
Andy told the court he had gone to Phillips Exeter Academy before
entering Harvard, and that he had helped Harvard win the Head of the
Charles rowing competition on the weekend just before the worm hit
when Robert was in Cambridge allegedly poring over UNIX source code.
Even the judge, a proud graduate of the University of Pennsylvania,
another Ivy League school, was beginning to get caught up in the club-
biness of it all. As if it mattered, he interrupted Paul's testimony to
clarify some rowing parlance. The judge insisted the sport was called
"crew," and it wasn't until Andy took the stand and called it "rowing"
that the matter was resolved. At the end of Andy's testimony, the judge
went out of his way to congratulate the witness on his 1984 Olympic
silver medal.
   The only element of mystery in the defense was whether Guidoboni
would put Robert on the stand. But it wasn't really much of a question.
As Guidoboni saw it, it could only help for the jury to hear the whole
story from Robert himself. Not only would this be the first time since
the worm hit that Robert made any public statement whatsoever, but
his sincerity and utter lack of defensiveness would help the jurors see
their way through the technical jargon to the core of the matter. Robert
Morris had made a terrible mistake and shouldn't be punished for it.
    But when he took the stand on the final day of testimony, Robert
unintentionally came across as somewhat aloof, less endearing than he
might have been. So intent was he on explaining the technical details
that, rather than steal the jurors' hearts, he seemed slightly superior.
   Guidoboni had a hard time getting Robert to express his feeling about
the worm incident without provoking an instant objection from the
prosecution. Guidoboni tried three times and each time Rasch stopped
him. Finally, Robert managed to blurt out, "It was a mistake and I'm
sorry for it," before the prosecutor could object to Guidoboni's fourth
    Even Robert's honesty backfired on him a bit. When Rasch cross-
examined him, Robert answered with a simple and direct "yes" to nearly
every question. Yes, he planned and wrote the worm deliberately and
consciously. Yes, he deliberately planned for the worm to break into
computers on which he did not have an account. And yes, he deliber-
ately made the worm's code difficult to understand.
                                                           ~TH    ...   339

   When both sides were finished with their questions, the judge had a
few of his own.
   "Let me ask you something, Mr. Morris," he began. "You designed
this program so that only once in seven times when the virus tried to
intrude on some computer where it was already there that it would
succeed again."
   "Did it work?"
   "I expect that it did work, yes."
   "So the mistake you made was that the access was too rapid, it ex-
ceeded your expectations. "
   "Yes, that is right."
   "So you misjudged the speed of the system."
   "Yes, I certainly made an error in judgment. "

After the testimony ended, Guidoboni argued for acquittal, claiming
that the government had not proven its case. At most, the defense
lawyer maintained, prosecutors proved the elements of a lesser charge, a
misdemeanor under the statute. The question of a lesser charge had
arisen before, and this time the judge appeared to be ready to consider it
seriously. But he could not give the jury the option of considering a
misdemeanor charge along with the felony unless the defense asked him
to. That meant the decision was up to Robert. It was a gamble. It meant
weighing the chance of an acquittal against the chance of a compromise
verdict. Having a misdemeanor on his record was certainly preferrable
to a felony. Then again, if the jury didn't have a lesser charge to fall
back on, it might choose the harsher conviction. When Munson ad-
journed the court until the following Monday, the matter was still un-
   The family had three tense days to kill in Syracuse. In search of
distraction, having already toured the nearby Corning glass factory,
everyone went bowling, then to the movies. Robert spent the weekend
sleeping late.
   Since the judge hadn't yet ruled on the prosecution's motion to disal-
low testimony related to Robert's intent, it was expected that his instruc-
tions to the jury would serve as his ruling. He took the weekend to think
about it. Both sets of attorneys presented themselves in the judge's
chambers first thing Monday morning and heard his decision on intent.
To Guidoboni's disappointment but not to his surprise, the judge had
340   •   CygERPUNK
not changed his mind on the irrelevance of Robert's intent to cause
damage. As the judge read the law, the question of intent pertained only
to whether the defendant intended to break in, not whether he intended
to cause damage.
   At that point, Guidoboni knew the case was over. His closing argu-
ments to the jury were listless; his demeanor smacked of defeat. Prose-
cutor Ellen Meltzer, on the other hand, made her closing arguments
with special vigor. She reminded the jury that they had heard Morris
himself admit to the crime, to its deliberate nature and to causing autho-
rized users to lose use of their computers. "Mr. Morris's computer worm
was not a juvenile prank," she intoned. "In no way was it a legitimate
Cornell research project gone awry."
   Indeed, she said, there is nothing to thank the defendant for. "You
do not thank a terrorist for increased airline security awareness and you
do not thank a drunk driver-"
   But before she could finish Guidoboni was on his feet, outraged by the
comparison. The judge agreed. The word terrorist had special meaning
in Syracuse. A year earlier, terrorists had blown upa Pan Am jumbo jet
over Lockerbie, Scotland; among the passengers were thirty-five students
returning home from a Syracuse University study-abroad program. Cer-
tainly in this city, of all places, Robert Morris could not be compared to
terrorists who claimed innocent lives. The judge sustained Guidoboni's
   When Judge Munson delivered his instructions to the jury, in an
implicit ruling on the prosecution's intent motion, he told the panel to
disregard the testimony concerning the defendant's intention to cause
damage. It was not relevant to the charge. He read the indictment to
the jurors and explained the elements of the crime. If, in its delibera-
tions, the panel should conclude that the government had failed to prove
each of the elements, then the jury must return with an acquittal. But
he said nothing about a lesser charge. Robert had decided to gamble on
an acquittal.
   During jury deliberations, Robert, Janet, Ben and Meredith started
playing cards at the defense table until the guard told them to stop.
Robert picked up his historical novel. Bob occupied himself with an old
eighty-five-cent Penguin Classic edition of Xenophon's History of Rome.
Anne mingled with some of the reporters. Five hours later, the jury filed
back into the courtroom and delivered its verdict. Robert watched in-
tently as each juror was polled, and as each said, "Guilty." Reporters
made a dash for the two pay telephones in the building. The judge set a
                                                          RTH    ~   341
date for posttrial motions and adjourned the court. Robert stood up and
smiled meekly at Janet. Anne rushed up to the defense table but Bob
hung back, staring straight ahead, the fingers of his left hand rubbing
together as if touching an invisible piece of fabric. Anne motioned for
him to join the rest of the family, which had formed a huddle at the
defense table. Anne wanted to talk to Guidoboni about their options.
The Morrises were people accustomed to having the privilege of options
and a felony conviction seemed to be no exception. In the face of
something essentially beyond her control, Anne was still considering
where to go from here. But Guidoboni held up his hands and said,
"Listen, folks, we don't have to go over this right now. There's plenty
of time to talk about it."
   In other circumstances, the computer scientist who worked for an
American intelligence agency so hidden that even its name was once
secret and his young son would quietly have lived out their lives as part
of the country's best and brightest. Now the father was caught in the
glare of publicity and the son was a felon. Someone who seemed destined
to become one of the country's software stars was facing jail.
   The lawyer gave Robert an apologetic handshake, snapped at the
throng of reporters on his way out of the courthouse, and returned to his
hotel. Flanked by family members, Robert left without saying a word. In
his son's absence, Bob Morris stood in the plaza in front of the court,
drawing reporters into a tight circle around him by responding to their
questions in a voice just above a whisper. He said he still believed that
his son didn't have a fraudulent or dishonest bone in his body.
      A s soon as they heard of Kevin Mitnick's arrest in December
1988, local law enforcement officials in Los Angeles rushed to see the
U. S. magistrate and urged that she deny bail. The suspect, they said,
had once fled to Israel and there was no telling where he might run this
time. Moreover, he was a known menace to the law-enforcement com-
munity. They claimed that he had tampered with one judge's credit
rating and disrupted the phone service of one of his probation officers.
The magistrate denied bail. " 'DARK SIDE' HACKER SEEN AS ELECTRONIC
   Assuming that there was no telling the havoc Mitnick might wreak
from a telephone alone, the judge also sharply restricted his telephone
access. He could call only those numbers that had been approved by the
   Alan Rubin, Mitnick's court-appointed lawyer, attempted to discredit
some of the myths that surrounded his client. Not only had Mitnick
never fled to Israel, he had never been out of the country. Rubin didn't
dispute the tampering with a probation officer's phone service, but he
did insist that Mitnick had never altered a judge's credit rating. After
Digital assessed the damage Mitnick had done at $160,000, the lawyer

                                                       £~~        "   343
worked out a plea bargain with the Justice Department: Mitnick would
plead guilty to two of the government's four felony counts and get a year
in federal prison and psychiatric counseling. The prosecutor on the case
supported the agreement because taking the case to trial would have
required granting immunity to Lenny DiCicco, whom the prosecutor
considered to be equally guilty. And for their own reasons, Digital offi-
cials were satisfied to see the matter resolved without a public trial.
    But the press had already painted such an unsavory a picture of this
defendant that the judge, perhaps influenced by what she read, rejected
the plea bargain. She pointed out that Kevin Mitnick had enjoyed leni-
ency in the past, and it had been ineffective. A year in prison wouldn't
be enough for this dangerous criminal, whose sophisticated stunts the
public found terrifying. "We won't know the damage Mr. Mitnick has
done until it's too late," the judge told the courtroom.
    In a show of the same willingness to turn on a friend that Lenny had
displayed, Kevin offered to cooperate with federal prosecutors against
Lenny. Then Kevin's attorney switched tactics. He convinced the judge
that his client's computer behavior was something over which his client
had little control, not unlike the compulsion to take drugs, drink alcohol
or shoplift. It was a persuasive argument. This time, the judge accepted
the one-year prison sentence. The prison sentence would be followed by
six months in a rehabilitation program. Now Kevin wasn't just a com-
puter hacker: he was a computer addict.
    Kevin spent the remaining months of his prison sentence at Lompoc,
a low-security federal prison in Southern California. He went from there
to a small residential program that emphasizes the twelve-step Alcoholics
Anonymous model for treatment. When he entered the program he
remained aloof from the others at first, asserting that his problem was
unique and no one else could understand it. He was cold and emotionally
remote. For the first several weeks, in what his counselor called a classic
example of denial, Kevin continued to maintain that he could stop any
time he wanted. Eventually, he came to accept the facts of his compul-
sion and his counselor said she believed he had changed his behavior.
    He also started working out nearly every day and dieting seriously.
While in the rehabilitation program, he lost one hundred pounds. The
conditions of Kevin's probation prohibited him from so much as touch-
ing a computer, but once he demonstrated that he could control his
behavior, he was allowed to search for computer work, although he was
still prohibited from using a modem.
    Kevin began attending meetings for codependents and adult children
344   •   CygERPUNK
of alcoholics. He was released from the program two months early and
got a job working as a programmer for a Los Angeles-area health-care
provider. Still hurt by the deception Kevin had carried out during his
many months of clandestine hacking, Bonnie Mitnick asked him for a
separation. Kevin moved in with his grandmother in Van Nuys, a sub-
urban town in the San Fernando Valley.
   Lenny eventually pled guilty to one felony count. He was sentenced
to five years' probation and 750 hours of community service. He started
out helping to set up a computer service for a homeless shelter, but when
the shelter closed down, he put off finding a new project and his proba-
tion officer didn't press him. He got a programming job at a small com-
pany in Orange County. Lenny was also ordered to pay $12,000 in
restitution to Digital. Lenny tried to get Kevin's old "X HACKER"
vanity license plate for his car, but the Department of Motor Vehicles
refused to give it to him. He settled for "VMS WIZ" instead.

                                  .. ...

Shortly before the trial in Celle, Pengo moved into a spacious apartment
on a quiet street in Kreuzberg and continued free-lance programming.
After his court testimony in January 1990, he awaited the official deci-
sion on his own fate. Three months later, after a series of difficult
discussions between Pengo's lawyer in Bayreuth and the German author-
ities, word came from Karlsruhe that all charges against Pengo had been
dropped. There was no probation and no fine.
    At the recommendation of the Digital Equipment Corporation lawyer
who attended the Celie trial, Digital considered filing a civil suit against
Pengo for his possession of the company's proprietary software. Pengo's
lawyer finally negotiated a settlement. Pengo signed a document prom-
ising never again to break into Digital's computers. He was not required
to pay restitution to the company.
    In late 1990, Emery Air Freight Corporation, whose network user
identification Pengo had used so generously, sent Pengo a letter demand-
ing $13,000. After receiving no response from Pengo, Emery's lawyers
filed a civil suit against him in Berlin. Once again, Pengo's lawyer inter-
vened, took up negotiations with Emery and settled out of court for
$2,300, just enough to cover Emery's attorneys' fees. Pengo has yet to
receive a bill from his lawyer.
                                                               f~~       T    345
          Markus Hess is still a programmer for a small publishing firm in Han-
          Cliff Stoll, now divorces from Martha Matthews, is busy writing a
    \   book on astronomy and lectures frequently on computer security.
          Laszlo Balogh is still living in Pittsburgh.

        While the final decision was being made in Pengo's case, Robert Morris
        was sentenced in Syracuse, New York. Reporters filled the courtroom.
        The entire Morris clan made the trip for the event. Bob was wearing
        new black shoes and a baseball cap bearing the name "Basil's Deli," a
        lunchtime hangout near NSA headquarters.
           Tom Guidoboni's partner, David O'Brien, made an impassioned plea.
        Taking care to refer to his client as "Robert," O'Brien called the twenty-
        four-year-old defendant "a decent kid" who had already taken sufficient
        responsibility for his actions. Guidoboni then took issue with the govern-
        ment's estimate of $160,000 in losses from the worm, and argued that
        Robert should be shown some leniency because he had taken responsi-
        bility for his actions. Guidoboni maintained that Robert deserved no jail
           When his turn came, Mark Rasch, the federal prosecutor, had little
        to say except that he believed some time in jail was appropriate. There
        had been a last-minute dispute within the Justice Department over the
        appropriate punishment. Unable to make up their minds for more than
        eight months after the incident about whether to prosecute, government
        officials now appeared to be having difficulty deciding how to punish
        Robert Morris. In an unusual move, the department decided not to file
        a sentencing recommendation in the case.
           Like a game-show host breaking for a commercial, Judge Munson
        called a five-minute recess and said he would come back with a sentence.
        When he returned, he asked if the defendant had anything to say.
        Robert replied that he didn't. The judge then delivered a short speech.
        He had been inundated with letters, he said. Some called for leniency,
        while others branded Robert Morris a grave threat to a free society.
        Munson complained that he could hardly step outside his front door
        without encountering someone who wanted to give an opinion. Even
        middle-aged women at his country club were buttonholing him to ex-
        press their sympathy for the young computer scientist. Prison, he said,
        would not be a suitable punishment for this crime. He sentenced Robert
        to three years' probation, a $10,000 fine, and four hundred hours of
346   ~   Cyg£~PUNK

community service. "I'd still like to hear about that trapdoor program,"
he added as he adjourned the court, referring to a technical term he had
heard during the court testimony. Had Robert been within reach, the
judge might have reached down and tousled his hair. A few days later,
Guidoboni announced that Robert planned to appeal the conviction.
   Like the act he had committed, Robert's sentence elicited deeply
divided opinions from both computer scientists and the larger commu-
nity. Many believed the decision not to include a jail term in the sen-
tence was fair. Robert had made a terrible mistake and had already paid
for it in accruing a debt of more than $150,000 in fines and legal fees.
Others reacted to the light sentence with particular rancor. Eugene
Spafford, the Purdue computer scientist turned ethicist, was quoted in
the Association for Computing Machinery's Journal of the ACM as call-
ing on the computer industry to boycott the products of any computer
company that would hire Robert Morris.
   For his own part, Robert kept a low profile. Beyond his statement on
the witness stand that he had written his program as a poorly conceived
experiment, he chose not to say what else had motivated him. After the
sentencing, he got a long hug from his mother and a firm handshake
from his father, but he still refused to talk with reporters. While his
family went downstairs to face the throng of reporters, Robert stayed in
the courtroom and discussed the terms of his probation with his attorney
and court officers. Later he left via a side door, unobserved.
   To fulfill his court-ordered community service, Robert worked at the
Boston Bar Association. Today he is working as a programmer at a
software engineering research firm in Cambridge, Massachusetts, and
learning ancient Greek as his father once did. He plans to apply for
admission to the Harvard University graduate program in computer sci-
ence. On March 7, 1991, the U.S. Court of Appeals for the second
circuit upheld Robert's conviction.

         The     history of phone phreaks that appears on page 18 ("When
    Susan and Roscoe met ... ") was drawn largely from personal inter-
    views, the chapter titled "Blue Boxes and Phone Phreaks" in The Biggest
    Company on Earth by Sonny Kleinfeld (Holt, Rinehart and Winston,
    1981) and from Fighting Computer Crime by Donn B. Parker (Charles
    Scribner's Sons, 1983).
       The description of the U.S. Leasing incident that begins on page 33
    ("The first opportunity for revenge ... ") was based on personal inter-
    views and on court transcripts.
       The history of Digital Equipment Corporation that begins on page 37
    ("This wasn't the reaction ... ") was based in part on the book The
    Ultimate Entrepreneur: The Story of Ken Olsen and Digital Equipment Cor-
    poration by Glenn Rifkin and George Harrar (Contemporary Books,
    1988) and on a speech titled "Digital Equipment Corporation: The First
    Twenty-five Years," delivered by Kenneth H. Olsen on September 21,
    1982, to the Newcomen Society in Boston.
       The description of the COSMOS incident that begins on page 48

348   & eyg6~PUNK

("It was at the Shakey's Pizza Parlor ... ") was based largely on tran-
scripts of a preliminary court hearing in the case.
   The description of the USC incident that begins on page 68 ("For all
Mark Brown knew ... ") was based on personal interviews and on offi-
cial police records.
   The description of the National GSC incident that begins on page 74
("Richard Cooper wanted to know ... ") was based on official police
   The description of the Santa Cruz Operation incident that begins on
page 86 ("Steph Marr had been around computers ... ") was based on
interviews with employees of the company and on reports filed by officers
of the Santa Cruz Police Department.
   The description of the events involving Pierce College and Kevin
Mitnick that begins on page 93 ("The first call from Pierce ... ") were
based on personal interviews and on reports of the Los Angeles Police
   The description of Kevin Mitnick's brush with Security Pacific Bank
that begins on page 100 ("For anyone with Kevin's record ... ") was
based on personal interviews and on reports of the Los Angeles Police
   The description of Digital Equipment Corporation's entanglement
with Kevin Mitnick and Lenny DiCicco that begins on page 105 ("Mark
Brown noticed immediately ... ") was based on personal interviews
with Digital employees and others, on electronic mail messages that
passed among Digital employees at the time and on FBI reports and court
   The description of Kevin Mitnick's final arrest in 1988 that begins on
page 128 ("It was the stunt ... ") was based on personal interviews and
FBI reports.
   Digital officials turned down our interview requests and declined to
comment on this and other sections of the book.

General Sources

Douglas Colligan, "The Intruder," Technology Illustrated, October/No-
       vember 1982.
"The Electronic Delinquents," transcript of the April 22, 1982, broad-
       cast of the ABC News program "20/20."
John Johnson, " 'Dark Side' Hacker Seen as Electronic Terrorist," Los
        Angeles Times, January 8, 1989.
Karen Kingsbury, "An Obsession with Computers," Los Angeles Daily
        News, December 19, 1988.
Karen E. Klein, "Magistrate Refuses Bail for Valley Computer Whiz,"
        Los Angeles Daily News, December 24, 1988.
Bill Lawren, "Computer Crime," Penthouse, July 1982.
Jeffrey Perlman and Debi A. Hastings, "Computer Raiders Hunt Se-
        crets," December 15, 1981.
Eddie Rivera, "The Phine Art of Phone Phreaking," L.A. Weekly, July
       18-24, 1980.
Ted Rohrlich, " 'Phone Phreak' Sentenced to ISO-Day Term," Los An-
      geles Times, June 11, 1982.
Ron Rosenbaum, "Secrets of the Little Blue Box," Esquire, October
John R. Wilke, "At Digital Equipment Slowdown Reflects Industry's Big
       Changes," The Wall Streetlournal, September 15, 1989.

In general, facts surrounding the espionage activities of the German
group were gathered from personal interviews, court testimony and the
final seventy-one-page verdict issued by the West German court.
   The message from CERN system manager Alan Silverman that ap-
pears on page 154 was reprinted in Die Hacker Bibel, Part 2.
   The log of the hacker's attempts to gain access to the White Sands
Missile Range that appears on page 180 was taken from The Cuckoo's
Egg by Cliff Stoll (Doubleday, 1989).
   The section on Laszlo Balogh of Pittsburgh was based on interviews
with Howard Hartmann, Linda Doebler, Cliff Stoll and Yale Gutnick.

General Sources

Thomas Ammann, Matthias Lehnhardt, Gerd Meissner and Stephan
     Stahl, Hacker fUr Moskau (Wunderlich, 1989).
350   •   Cyg£RPUNK
Stefan Aust, Der Baader-MeinhorKomplex (Hoffman und Campe Verlag,
Dieter Brehde, "Der Tod des Hackers," Stem, June 8, 1989.
Malcolm W. Browne, "World's Biggest Accelerator Surges to Life," The
        New York Times, August 8, 1989.
"Hacken fur den Weltfrieden," Suddeutsche Zeitung, January 12, 1990.
Jane Kramer, Europeans (Farrar, Straus & Giroux, 1988).
Steven Levy, Hackers: Heroes of the Computer Revolution (Anchor Press/
        Doubleday, 1984).
James M. Markham, "France's Minitel Seeks a Niche," The New York
        Times, November 8, 1988.
Robert Shea and Robert Anton Wilson, The Illuminatus! Trilogy (Dell,
Ulrich Sieber, The International Handbook on Computer Crime (John
        Wiley & Sons, 1986).
Mary Stolberg, "Informant Key in Computer Parts Theft Trial," The
        Pittsburgh Press, January 21, 1983.
Cliff Stoll, The Cuckoo's Egg (Doubleday, 1989).
Sherry Turkle, The Second Self: Computers and the Human Spirit (Simon
        and Schuster, 1984).
jurgen Voges, "Hacker wollten Weltfrieden sichern," Tageszeitung, Jan-
        uary 12, 1990.


In general, the events in this section draw on interviews with friends,
associates and relatives of Robert Morris, and with law-enforcement
      The description of events surrounding the paper on breaking the
M-209 encryption scheme that begins on page 267 ("In the mid-
1970s ... ") comes from an interview with two of the paper's authors.

General Sources

John Perry Barlow, "Crime and Puzzlement," Whole Earth Review, Fall
Harold L. Burstyn, "RTM and the Worm That Ate Internet," Harvard
        Magazine, May-June 1990.
Dorothy Denning, "Concerning Hackers Who Break Into Computer
        Systems," Digital Equipment Corporation, Systems Research
        Center. Paper presented at the Thirteenth National Computer
        Security Conference, Washington, D. c., October 1990.
Peter J. Denning, "The Science of Computing: The Internet Worm,"
        American Scientist, March-April 1989.
Ted Eisenberg, David Gries, Juris Hartmanis, Don Holcomb, M. Stuart
        Lynn and Thomas Santoro, "The Cornell Commission: On Mor-
        ris and the Worm," a report issued by Cornell University, Feb-
        ruary 6, 1989.
David Kahn, The Code breakers (Macmillan, 1967).
Gina Kolata, "When Criminals Tum to Computers, Is Anything Safe?"
        Smithsonian, August 1982.
Jonathan Littman, "The Shockwave Rider," PC/Computing, June 1990.
John Markoff, "Author of Computer 'Virus' Is Son ofN.S.A. Expert on
        Data Security," The New York Times, November 5, 1988.
John Markoff, "Cornell Suspends Computer Student," The New York
        Times, May 26, 1989.
John Markoff, "How a Need for Challenge Seduced Computer Expert,"
        The New York Times, November 6, 1988.
John S. Quarterman, The Matrix: Computer Networks and Conferencing
        Systems Worldwide (Digital Press, 1990).
Jon A. Rochlis and Mark W. Eichin, "With Microscope and Tweezers:
        The Worm from MIT's Perspective," Communications of the
        ACM, June 1989.
Donn Seeley, "A Tour of the Worm," University of Utah Department
        of Computer Science, Technical Report, November 1988.
John F. Shoch and Jon A. Hupp, "The 'Worm' Programs-Early Expe-
        rience with a Distributed Computation," Communications of the
        ACM, March 1982.
Eugene H. Spafford, "The Internet Worm: Crisis and Aftermath," Com-
        munications of the ACM, June 1989.
Ken Thompson, "Reflections on Trusting Trust," 1983 ACM Turing
        Award Lecture, Communications of the ACM, August 1984.
Michael Wines, "A Youth's Passion for Computers, Gone Sour," The
        New York Times, November 10, 1988.
      Hundreds of people agreed to be interviewed for this book. Many
sat patiently, explaining again and again some of the more technical
aspects of the subject. Many others helped us to reconstruct events,
hauling out their old calendars and notebooks and computer printouts.
We are especially grateful to Renate and Gottfried Hubner, Bob and
Anne Morris and Gil DiCicco for spending time with us to talk about
their sons. Thanks, too, to Bonnie Mitnick for speaking with us about
Kevin Mitnick who, despite repeated requests, refused to be interviewed.
   We are deeply indebted to Paulina Borsook, Mark Seiden and Debo-
rah Wise for their editorial body and fender work. The manuscript was
also read in its various stages of completion by Everett Hafner, Steven
Levy, Katherine Magraw, Annabelle Markoff, Andy Pollack, Peter
 Preuss, Debbie Yager and Susie Zacharias, all of whom made helpful
 comments. The manuscript benefited greatly from the keen eye of copy
 editor Stephen Messina.
    We received invaluable research help from Thomas Ammann, Keith
 Bostic, Dieter Brehde, Dave Buchwald, Evil Corley, Udo Flohr, Tom
 Guidoboni, John Johnson, Dan Kane, Ekkehard Kohlhass, Phil Lapsley,
 Jon Littman, Mike McAndrew, Doug McIlroy, Gerd Meissner, Peter
 Neumann, Dennis Ritchie, Eddie Rivera, Alan Rubin, Ulrich Sieber,
354   •   Cyg£RPUNK
Steve Steinberg, Cliff Stoll, Ken Thompson and Michael Wines. For
helping to piece together various incidents involving Kevin Mitnick, we
are grateful to Bob Ewen of the Los Angeles district attorney's office and
Detective Jim Black of the Los Angeles Police Department. For their
friendship and hospitality we'd like to thank John Kelley and Lisa Van
Dusen, Sara Charno, Silke Grossmann-Brehde, Keith Hammonds, An-
drea Klotz, Mara Liasson, David Olmos, Ralph and Sonya Raimi, Heiko
and Mechtild Rogge, Seth Rosenfeld, Marc Rotenberg and Gail Schares.
Thanks, too, to Paul Saffo and the folks at the Institute for the Future
in Menlo Park for office space and a Macintosh.
   Fortune smiled on us when we were introduced to Bob Bender, the
senior editor at Simon & Schuster who took this book on. He thought
this was just a swell idea from the start and he didn't change his mind.
His every suggestion was invaluable. And thanks to our agents John
Brockman and Katinka Matson for making the Bender connection.
   A final note: Roscoe and Susan, two of the characters in the Kevin
Mitnick section, cooperated with us in the understanding that their true
names would not be revealed. We respect their right to privacy.