Chapter 1

CODE WARRIOR

Dave Cutler was reared on adversity. He learned at a young age to care for himself, to keep his own counsel, to find a way around or through the obstacles in his path.

He was born on March 13, 1942, in Lansing, the state capital of Michigan. Lansing was auto country, home to a slew of car and car-parts makers. His father, Neil, worked in Lansing's Oldsmobile plant for nearly his entire life, first in the plant's shipping department and then as a janitor.

Neil Cutler was an intelligent and exacting man, but he was quiet and lacked ambition. He had been stricken with rheumatic fever as a boy, and it had left him too frail to play sports. Poor eyesight made it difficult for him to enjoy the outdoors. A certain bitterness crept into him. He was not sociable; he struck some as almost a hermit. At home, he could be unpredictable, angry and gruff. He drank.

Arleta, Neil's wife, raised her son Dave and his older sister, Bonnie, in an apartment above the home of Neil's parents in DeWitt, a town of some one thousand people about eight miles north of Lansing. DeWitt was surrounded by farmland and consisted mainly of retired farmers who had moved off their farms and into the town. When Dave was eight the Cutlers moved out of town to a forty-acre spread. The land wasn't suitable for farming and did not contain a dwelling. Neil built a small home, one side of which was literally carved out of the earth. By then Arleta had given birth to two more children. The family seemed to spend all its time together in one large room. Arleta kept a large garden, and the family planted pine trees on the land. In time thousands of trees took root and grew.

From the age of ten David Cutler earned money whenever he could. He mainly worked during the summertime for the many farmers in the area, building barns or doing odd jobs. One summer, he worked in a fertilizer plant. Another year he collected old newspapers with a friend, filling an entire trailer for sale to a recycler.

As a teenager, Cutler was drawn to sports. With a graduating class of thirty-four students, his tiny high school pressed him into service. He ran track and played baseball, basketball and football. He was co-captain of the basketball team and the quarterback of the football team. In one game, he ran for two touchdowns, running almost the entire length of the field for one score. He was very fast.

The local newspaper treated Cutler as a star, chronicling his exploits. Neil skipped nearly all of his son's games; it took a personal invitation from the football coach for him even to consider attending the one game during his son's senior year in which every player's father was introduced. Neil (who went) said he disliked sports, but Arleta suspected that jealousy had kept her husband from the sidelines.

Father and son were distant. While still in high school, Cutler moved out of his parents' home for a time, living first with the family of a baseball coach and then with Bonnie. At school, meanwhile, Cutler did well enough without studying hard. Graduating in June 1960, Cutler seemed serene about his prospects. Somewhere inside him sprang a confidence bordering on arrogance and a belief that he could be the best at anything he tackled. Others shared Cutler's buoyant sense of himself. In his high school yearbook, classmates captured his specialness in a line beneath his photograph:

"None but himself could be his parallel."

A small Michigan college called Olivet cobbled together several athletic and academic scholarships, offering them as a package to Cutler. He signed on. His freshman year, he started at quarterback, calling and directing his own plays just like a pro. He threw the ball well and ran one hundred yards fast -- in less than eleven seconds. He was about five feet nine inches tall, weighed about 175 pounds and had thick, strong legs. His coach, Stu Parsell, called him "a one-in-a-million player" and marveled at his elusiveness. Cutler was a wily player who confessed he "loved to run over people."

In the huddle, Cutler smartly dished out assignments in between plays. He brooked no dissent, berating teammates for their lapses and telling them: "This huddle is my territory. When you're in it, shut up." When players "screwed up," he said, "I'd really ride them, telling them what to do...to get out there and do their job."

Coach Parsell realized that Cutler relied on more than athletic skills. "He was smart enough to know he couldn't win alone," Parsell said. "He brought the other players up with him. They rose to him." The team responded to his brash assertions because Cutler led by example and "knew what he wanted."

Cutler's game peaked in his sophomore season. The long-suffering Olivet Comets, who had lost twenty-one games in a row in the late fifties, suddenly went white-hot in the fall of 1961. With Cutler at the helm, the team won its first eight games. Then, in its final game, disaster struck. Midway through the game, Cutler took the snap from the center and rolled right, preparing one of his quarterback rushes. He had already scored that season on just such a gambit. This time, he was in the clear, running full tilt along the sideline, right along his team's bench, so close that Coach Parsell could have grabbed him. Then a defender charged toward him, hurling his body in Cutler's way. Cutler tried to jump over him, but the defender smacked him squarely. He crumpled to the ground, his leg broken, his season over.

Cutler tried to return the next season, but on the eve of the opening game a doctor told him he risked permanent injuries if he played on. Cutler reluctantly withdrew.

With the end of his football days, Cutler concentrated on his studies. He excelled in math, dabbled in the sciences but finally decided to pursue engineering. When he graduated in January 1965, he was offered a job programming computers for General Motors. Along with other big companies, GM had begun shifting its business records from paper to computer in the 1950s. But Cutler was not eager to join GM. He knew nothing about computers, which seemed vaguely threatening -- even sinister -- to him. In the mid-1960s, many people shared this dystopian view of computers. These machines, which were designed to crunch numbers, were treated with skepticism and sometimes hostility because they symbolized regimentation. Computers seemed to bend humans to their will, forcing men and women to do little more than tend smart machines.

This gave a bad reputation to computers and the task of writing programs for them. Hardly anyone wished to call himself a programmer, and people who did were considered odd. Just a few years before Cutler graduated from Olivet, the top programmer in the Netherlands, an erstwhile physicist, described himself as a programmer on his marriage license. To his dismay, authorities rejected the license on the grounds that there was no such job.

Alert to signs of esteem and status, Cutler held a "very stereotyped view of programmers." To a young man, raised in relative penury and intent on making his way up the economic ladder without kowtowing to authority, programming "seemed a very uncreative job"; and those who did it followers of "this fixed bunch of rules," not leaders who called their own plays.

He wanted no part of software and turned General Motors down fiat. Instead he took a post with DuPont. He adapted easily to the conservative and prosperous chemical giant. He kept his hair short and maintained a military bearing. He thought first of earning his keep; he had married a woman he'd met in college, and had already fathered a child.

DuPont assigned Cutler to a unit that helped customers find uses for its materials. One of his first jobs was to model a new way that Scott Paper intended to make foam insulation for use in jackets and other garments. The model was so complicated that it required a computer to create. Off Cutler went to a school run by IBM, where he learned how to program an IBM computer.

Cutler spent a week at the school. He felt humbled. Programming "was just the most bizarre situation, because you're used to doing something and thinking you've done it right," he later said. "But it isn't right. You just don't notice it isn't right. On a computer there is no consolation in discovering you're almost right. Almost means you're still just wrong."

Even veteran programmers often found their jobs excruciatingly tedious. In those days, no one had his own computer, of course. Dozens of programmers would share a single mainframe computer. The mainframe, large enough to fill a room, handled many jobs at once in batches. In batch jobs, a programmer punched instructions onto perforated cards, added a stack to the queue and waited for results. Since the mainframe was so expensive, the batch schedule was strict. It often took hours or longer to learn the fate of a program. If it failed, a programmer could spend an entire day just correcting keypunch errors.

Cutler returned to DuPont determined to excel at programming. The activity intrigued him because, in a program, he was master of his environment. He also found he had a rare ability to hold in his mind at once the various and far-flung pieces of a program. He began to crave programming. Impatient with the long lines in DuPont's computer facility, he worked in the middle of the night, when computer time was much cheaper and he could assemble and revise his cards in peace. "There was hardly anybody there," he recalled. "I could make twice as many mistakes, and get on and off the computer when I wanted to."

Foam making, by contrast, did not keep Cutler awake at night. In less than a year, he had succumbed to the attraction of computing. Having found in the computer the ideal means to answer a question, he promptly lost interest in the question and fell blindly in love with the tool. Indeed, Cutler had found a calling in life. "What I really wanted to do was work on computers, not apply them to problems."

So Cutler, looking for a new job that involved programming, found one with another DuPont division that needed help in maintaining its central computer, which was made by Univac. In the 1950s Univac made the best computers for data processing, but by the late 1960s the company was in decline. DuPont asked Cutler to improve the reliability of its aging Univac, which meant fiddling with the machine's operating system. Until then Cutler had never even thought about operating systems. But the company's computer experts seemed not to know much either, and he jumped in. Computer programs fall into two rough classes. Applications, or "apps" for short, are the visible world of software. They comprise the programs used by ordinary people. Applications track orders or inventory, retrieve names and phone numbers, prepare a document for printing or control the design of a newsletter.

Operating systems, on the other hand, are part of the invisible world of software. They are the computer's heartbeat, throbbing in the background. Applications may appear to do all the work, but in reality operating systems open and close files, create directories of the stored information and direct the traffic to and from the computer's input, output, storage and networking devices.

In the formative years of digital computing, following World War II, both the operating system and applications were considered afterthoughts by designers. The "hardware" of electronics, as distinct from the "software" of programs, was so difficult that engineers could hardly see past it. The most important type of hardware was the circuitry or processors that actually carried out the instructions given the computer. A second set of devices made it possible to get data into and out of a computer. A third class stored information. A fourth class allowed one computer to send information to another, over special cable or telephone lines.

The question of software generally arose only after the hardware pieces fell into place. Computers were not designed with specific software in mind; rather, the programmer worked with what the hardware gave him. E. W. Dijkstra, a leading theorist of programming, once summarized the prevalent attitudes toward code writing in the formative period of computing. He declared:

What about the poor programmer? Well, to tell the honest truth, he was hardly noticed. For one thing, the first machines were so bulky that you could hardly move them and besides that, they required such extensive maintenance that it was quite natural that the place where people tried to use the machine was the same laboratory where the machine had been developed. Secondly, the programmer's somewhat invisible work was without any glamour: You could show the machine to visitors and that was several orders of magnitude more spectacular than some sheets of coding. But most important of all, the programmer himself had a very modest view of his own work: his work derived all its significance from the existence of that wonderful machine. Because the machine was unique, he knew his programs had only local significance. And since the machine would live for a short time...he knew that little or none of his code held lasting value.

The essence of programming seemed deceptively simple. A person wrote a request to a computer. This request was stated in a way the computer could understand. It also was stated, ultimately, in terms that only a specific computer could understand. The same request, written exactly the same way, fed to a computer with a different design or circuitry, would be unintelligible.

Besides being handmaidens of a specific computer, the first computer programs were crude. Before World War II, when computers were largely mechanical, a program often amounted to nothing more than a person flipping switches, rerouting wires or shifting gears. In the 1930s, it took many days to prepare the Differential Analyzer, the decade's most powerful mechanical computer, to attack a fresh problem. A decade later, it still could take a few days to set up an early digital computer to answer a tough question.

More flexible machines read requests contained on punch cards or paper tape, but these were fed by hand into a machine. The primitive state of programming cried out for an advance.

The breakthrough came in 1944 when John von Neumann, a Hungarian-born mathematician living in the United States, advanced the concept of a stored program. The idea was familiar to others in the field, but yon Neumann saw its significance most clearly. With a stored program, the instructions for the computer could be kept in the machine's own memory and treated in the same way as data. This would make it far quicker to launch a program and easier to modify it or switch from one program to another.

As the stored-program concept spread throughout the nascent computer culture, programming exploded, quickly gaining adherents. It was hard going. Digital computers have either two states, on or off, and so respond only to binary messages, which consist of ones (on) and zeros (off). Every term in a program ultimately must be expressed through these two numbers, ensuring that ordinary mathematical statements quickly grow dizzyingly complex. In the late 1940s, programming a computer was, as one observer put it, "maddeningly difficult."

Before long programmers found ways to produce binary strings more easily. They first devised special typewriters that automatically spit out the desired binary code. Then they shifted to more expansive "assembly" languages, in which letters and symbols stood for ones and zeros. Writing in assembly was an advance, but it still required fidelity to a computer's rigid instruction set. The programmer had to know the instruction set cold in order to write assembly code effectively. Moreover, the instruction set differed from computer model to computer model, depending on its microprocessor design. This meant that a programmer's knowledge of an assembly language, so painfully acquired, could be rendered worthless whenever a certain computer fell out of use.

By the early 1950s, organizations that relied heavily on computers, most notably the three branches of the U.S. military, began to realize that software was a headache, and an expensive one. Leading-edge programmers searched for ways to make it easier to write effective programs. In 1951, Grace Murray Hopper, a mathematician with the U.S. Navy's Bureau of Ordnance Naval Reserve, conceived of a program called a compiler, which translated a programmer's instructions into the strings of ones and zeroes, or machine language, that ultimately controlled the computer. In principle, compilers seemed just the thing to free programmers from the tyranny of hardware and the mind-numbing binary code.

Hopper's insight spawned countless efforts at simplifying code writing. Probably the most important came from IBM which built a compiler called Formula Translation, or Fortran. It contained thirty-two instructions, such as PUNCH, READ DRUM and IF DIVIDE CHECK, which referred to the precise binary terms required by the computer. By the late 1950s, Fortran was hugely influential. "Now anyone with a logical mind and the desire could learn to program a computer," one historian of computing has written. "You didn't have to be a specialist, familiar with the inner workings of a computer and its demanding assembly language. By using Fortran's simple repertoire of commands, you could make a computer do your bidding, and the compiler would automatically translate your instructions into efficient machine code."

While Fortran enabled a programmer to use the same set of instructions to program any number of computers, the resulting programs often required changes to run on different machines. Moreover, Fortran was geared toward scientific and engineering problems. Other languages, such as Common Business-Oriented Language (Cobol, for short), sprang up for other purposes. Before long, programmers confronted a jungle of computer languages, and the course of their careers was often decided by the one they chose to learn best.

Hopper was convinced that overcoming the difficulties posed by proliferating computer languages would rank among the greatest technical challenges of the future. "To me programming is more than an important practical art," she said in a lecture in 1961. "It is also a gigantic undertaking in the foundations of knowledge." Ironically, she fretted that the greatest barrier to progress might come from programmers themselves. Like converts to a new religion, they often displayed a destructive closed-mindedness bordering on zealotry. "Programmers are a very curious group," she observed.

They arose very quickly, became a profession very rapidly, and were all too soon infected with a certain amount of resistance to change. The very programmers whom I have heard almost castigate a customer because he would not change his system of doing business are the same people who at times walk into my office and say, "But we have always done it this way." It is for this reason that I now have a counterclockwise clock hanging in my office.

In the early 1960s, IBM attempted to revolutionize the field of software by making it possible to run the same program on any number of computers. IBM proposed a family of machines, covering most of the market and controlled by one operating system. The System/360 line, developed at a cost of five hundred million dollars, was a huge success. But its birth was painful, mainly because the cost and difficulty of creating the software was grossly underestimated. It wasn't until five years after the first 360 hardware was introduced in 1964 that all of its software ran well. By then, IBM had spent nearly as much writing the software as designing the hardware. This astonished the company's managers and vividly highlighted "the greatest impediment to advances in computer technology," the problem of managing large software projects.

At DuPont, Dave Cutler found himself on the front lines of the effort by big business to tame the computer. The growing standardization of operating systems greatly improved the usefulness of computers. But much of the burden for writing applications still fell to the purchasers of computers themselves. This had the unintended effect of turning large companies into breeding grounds for programmers. DuPont asked Cutler to create a program to analyze experimental data from its labs. The job was tricky because it required two computers working in tandem. One would receive data, put it in a file and then ship it to the second computer, which would analyze the data and send results out to researchers.

The desired program was called a "real-time" system, which enabled a computer to receive information and respond immediately rather than within hours or days. The first real-time system had been created for the Whirlwind computer the Air Force and Navy used to track enemy aircraft and direct U.S. attack planes to their targets.

Real-time programs were a valuable innovation. To a company such as DuPont, trying to find new materials and uses for them, flesh answers were far more helpful than delayed ones. DuPont wanted Cutler's programs to run on small, fast computers from Digital Equipment Corporation. Formed a decade earlier by a Whirlwind engineer, Digital was a rising star among minicomputer suppliers, who broke sharply with tradition. In the past, computer designers had promoted large mainframes that shared their power between many jobs. Minicomputers, often priced well below a hundred thousand dollars, made it practical for the first time to dedicate a computer to a single job, such as keeping track of parts, the data for an experiment or the operation of a machine tool.

Though useful, minicomputers such as Digital's PDPs typically arrived with little software. Buyers usually had to create their own. Over the next few years, Cutler wrote his real-time program and other PDP software, becoming an expert in Digital code.

But Cutler grew tired of DuPont. He wanted to work for a computer company, and he chose Digital. The company, by exploiting a hole in IBM's product line, was not only growing rapidly but eschewing conventional business practices. It avoided acquisitions and housed Digital's employees in an abandoned mill outside Boston, in Maynard, Massachusetts. Rather than enforce the top-down style of management embraced by IBM, Digital allowed engineers to run with their own ideas, even at the risk of duplication.

Cutler was eager to write software for Digital's computers -- so eager that a casual meeting with a Digital salesman led to a job interview with the company. Digital, its hardware proliferating, needed strong code writers. Cutler fitted the bill. In 1971, at the age of twenty-nine, he took a job with Digital and moved to Massachusetts.

Cutler took Digital by storm. Before long, he was ranked among the company's software stars and given his own team of code writers. He was completely absorbed in his work. By then, his first marriage had broken up and his second was headed toward the shoals. He was, by his own admission, no family man. When he finally split up with his second wife, he vowed never to marry again. "Marriage is a mistake you only make twice," he said.

Cutler was the classic programmer. He was single-minded, obsessive and competitive. He had great stamina and confidence. He paid tremendous attention to details. And he kept getting better at designing and writing code. "Most people learn one neat thing and spend the rest of their life doing it," a colleague observed. "Not Cutler. He learns from his successes. He does the next one better. And every time he gets to another level. That's just astonishing, because so many people who succeed in technology never repeat it." Moreover, his very single-mindedness -- beyond his software assignment at hand, he displayed no other intellectual passions -- was a boon: "He not only ignores anyone and anything that might interfere, he denigrates them."

Cutler turned insults into an art form. Volatile and stubborn, he could be all brass and bravado, unconcerned with even the most mundane courtesies. He lost his temper, issuing a stream of profanities at the slightest provocation. Each new outburst burnished his legend. Strangers suffered most. A woman engineer once first met Cutler in a room full of printers, where he was fumbling with crumpled and twisted sheets of paper. She looked at him expectantly as he glanced her way. Then Cutler shouted: "Are you the fuckhead that screwed up the printer?"

Though he impressed no one with his manners, Cutler won praise for his role in building a series of real-time operating systems for Digital's PDP-11 computer. He excelled at the tricky task of squeezing a program into a smaller space. This was crucial, since the smaller the program, the faster it would run and the less computer memory it would consume. Cutler made plain his desire for compact code by keeping a rubber stamp on his desk that bore the motto: Size Is The Goal. When he thought a programmer wanted to add a feature that too greatly increased the demands on memory, he would use the stamp on his rejection memo. The stamp made programmers more careful about adding to their code and provided a source of jokes. Soon, the phrase "Thighs Is The Goal" was prominently displayed in the men's room.

To his crew, Cutler was a hero, a superman exempt from the ordinary rules. "He makes you feel you're one of his partners in greatness and that you will never have a better, more trustworthy ally," one said. Cutler expected everyone to perform flawlessly. To convey an urgent message, he'd call a huddle. He enjoyed teasing his mates but went to any lengths to improve the quality of their code. Sometimes, he even retrieved code thrown in the trash, corrected it with his red pen and returned the code to its author.

Not only could Cutler be stern and unfeeling, he expected absolute candor from his team. "If you try to snow him, if you try to deceive him, he'll be ruthless," one colleague said. "He won't just let you hang yourself, he'll attack. And he doesn't forget."

He was tight with praise too. Roger Heinen, one of Cutler's favorite programmers, often ached for a kind word from his mentor. "When you're out there and drop the ball, he is quick with criticism," Heinen said. After weathering such an attack, Heinen wondered if Cutler still valued him. Too afraid to ask, he would reach inside his desk for a note that Cutler had written him sometime before. The note said simply: "Roger, you really did a good job. Thanks, Dave."

A powerful patron made it possible for Cutler to flourish at Digital. Early on, his careful coding and driven leadership caught the notice of Gordon Bell, Digital's top engineering executive and the designer of its most important computers. A rare combination of technical genius and strategic savvy, Bell joined Digital in 1960 at the age of twenty-six after studying at MIT. He spent the next fifteen years "guiding, arguing, fidgeting and creating the computer strategies that [turned Digital into] IBM's strongest challenger."

Engineers were drawn to Bell, a warm and animated man. His short attention span was both endearing and aggravating. His mind would wander so much that those who really needed to speak with him tried to isolate him from distractions. "To talk to Gordon you had to take him in the car, drive and not let him turn on the radio," one said.

By early 1975, Bell saw Digital losing steam, despite its impressive financial results. The PDP-11 computer was then five years old, and IBM was on the verge of introducing its first true minicomputer. Bell felt Digital needed a much-improved machine to retain its lead over IBM. Over the winter, he conceived of a new line of computers that he felt would keep Digital ahead of its rivals in minicomputers while at the same time satisfying those customers who had invested heavily in the PDP-11. The new computer line, called the Vax, would run a new operating system, known as VMS, which would make it possible to run both new programs and old ones written for the PDP-11. "Backward compatibility" meant that many existing programs would work on the Vax computer from its inception, rather than waiting for programmers to write new applications as often happened in the past. The Vax design, meanwhile, was "scalable," which meant that the same software would run on Vaxes of varying sizes. Customers could switch to a larger Vax while retaining their familiar software.

All this suggested that the Vax marked a watershed in the history of computing. In assembling a team, Bell had his pick of Digital's brightest engineers. He asked Cutler, who he felt was Digital's top programmer, to lead the effort to create the VMS operating system.

Calling together Cutler and four other senior engineers, Bell launched the Vax project on April 1, 1975. For months the group hashed out design concepts in fiery meetings. "They started out quietly, with fifteen minute overviews of the agenda, but inevitably dissolved into shouting matches of apparent chaos and animosity. Somehow they always ended on a high note, engineers streaming out of the conference room smiling."

Bell helped Cutler by protecting him from meddlesome outsiders. "I wouldn't tolerate anyone getting in Dave's way on anything," he said. Even so, the VMS project put Cutler under great pressure. It showed. When he got out of bed some mornings, he was too dizzy to stand straight. An examination showed him to have sky-high blood pressure. He immediately went on medication to lower it. As a precaution Digital examined the other managers on the VMS team. The nurse who performed the tests said Cutler's blood pressure was the highest she'd ever seen.

Colleagues, often bloodied by Cutler's verbal jousts, couldn't resist taunting him about his health problem. During a heated debate one coworker jokingly said the suspense was killing him. Couldn't Cutler have his heart attack now, so they could stop worrying about it? "Can't you just get it over with?" he pleaded.

It was clear by then that Cutler needed help creating the VMS software. The team was small -- ten people at its height -- and Cutler was the senior programmer. But he wouldn't ask for help. He didn't do that kind of thing. He had a favorite saying: "When all is said and done, much more is said than done."

While Cutler was busy doing, the project's general manager, a man Cutler jokingly called "the boss," planned to hire a senior programmer to help Cutler whip the VMS software into shape. When Cutler left for a week's vacation, the boss interviewed his prime candidate and hired him, after Cutler returned. Cutler didn't think he needed a codesigner, but the newcomer proved to be his technical equal. If Cutler was grateful for the help, he did not say.

The first Vax computer was produced in October 1977. Its wild success vaulted Cutler into the front ranks of software designers. Bell, for one, didn't hesitate to call him the best writer of operating programs in the world. Wealth and recognition -- the hallmarks of achievement -- came Cutler's way. None of this altered his temperament. Digital's burgeoning bureaucracy annoyed him. "I'm a doer, and things became harder and harder to do," he said. To his distress, he could no longer direct the VMS project with his customary authority. The program was now so important to the company's future that a bevy of managers wished to improve it. Cutler fumed: "This was an opportunity for every Tom, Dick and Harry to come and poke holes in whatever you wanted to do, and hold up the project. Sure, there has to be a certain amount of review. That's healthy. But people who had no business with anything were popping up saying, why are you doing this, why are you doing that?"

Seeking less interference, Cutler left the VMS team. He wanted out of Digital's stultifying atmosphere. One day, he threatened to quit, telling Bell he wished to form his own company. Bell countered with an offer Cutler couldn't refuse: "Take anybody you want. Go anywhere you want to go. Do anything you want to do. And Digital will pay for it. Tell me how much money you need, and we'll fund you."

Cutler jumped at the chance. In the spring of 1981 he gathered his faithful flock and, wishing to move as far from Digital's Massachusetts office as possible, toured the West Coast in search of a new home. After spending a day in Seattle, Cutler made his decision. He was swayed by its proximity to mountains (he enjoyed skiing), the wilds (he still hunted) and the Pacific Ocean. A fine seafood dinner at a restaurant overlooking Shilshole Bay burnished his mood. The lack of a direct flight from Boston to Seattle, making it more time consuming for his corporate overseers to reach him, added to the location's appeal.

Cutler was happy at the helm of his own lab. "We knew what we wanted to build, and we could make all the decisions right there," he said. Bell stood ready to promote his ideas. At the lab, Cutler dressed casually, wearing his familiar sweat suits and sneakers, only now he didn't even don what one cohort called the "higher-quality T-shirts" he used to wear back east. A couple of nights a week, he invited everyone to join him at a pub, where he wasn't bashful about drinking large amounts of Red Hook, a local brew that he treated as a divine fluid. During the winter, he led his gang on ski trips, often outskiing people ten years younger. "He's not fun to ski with," said one coworker. "I just eat his snow."

Cutler was ambitious, but his plans met with mixed reactions. Along with software, he wished to design entire computers. Digital agreed to bring one of his machines to market, but his support withered after Bell, his patron, left Digital in 1983. "When Gordon left, our support network [at Digital] started to disappear," said Heinen. Suddenly, Cutler lacked clout. "It was clear the move to Seattle was a mistake," one colleague said. "From so far away, he lost his ability to champion his ideas effectively."

But he still had his freedom. On his business card he unabashedly identified himself as "Supreme Commander." In a lab "yearbook," compiled for the amusement of his crew, Cutler mocked the button-down etiquette of corporate life by posing for a photograph attired in a sleeveless white undershirt with a cheap striped tie hanging sloppily around his neck. Another photo, purportedly of his desktop, revealed three prominent items: an aerosol spray can inscribed with the words Bull Shit Reper was ambitious, but his plans met with mixed reactions. Along with software, he wished to design entire computers. Digital agreed to bring one of his machines to market, but his support withered after Bell, his patron, left Digital in 1983. "When Gordon left, our support network [at Digital] started to disappear," said Heinen. Suddenly, Cutler lacked clout. "It was clear the move to Seattle was a mistake," one colleague said. "From so far away, he lost his ability to champion his ideas effectively."

But he still had his freedom. On his business card he unabashedly identified himself as "Supreme Commander." In a lab "yearbook," compiled for the amusement of his crew, Cutler mocked the button-down etiquette of corporate life by posing for a photograph attired in a sleeveless white undershirt with a cheap striped tie hanging sloppily around his neck. Another photo, purportedly of his desktop, revealed three prominent items: an aerosol spray can inscribed with the words Bull Shit Repellent; a sign, issued by the "British Chicken Authority," that asked: Is your cock plump enough? and a drinking glass bearing the slogan: Joy Is A Tight Pussy.

Cutler balanced his ribaldry by espousing a stark and appealing philosophy of excellence. Mediocrity was for the weak, and greatness could be achieved only through a willingness to break irrational or outmoded conventions (and damn the consequences). "First of all, quality is something that has to be believed in by everybody," Cutler wrote in the yearbook. "This means management on down to the lowest peon. There is no room for a wishy washy manager that always leans the way the wind is blowing in the front office nor is there room for an engineer that tries to make it look like he did more work than he really did by short changing the quality aspects. It takes balls to stand up for quality at times and everyone on my projects gets that license. If some jerk somewhere wants to ship before we're ready, we tell him to pound sand."

In 1985, Cutler narrowly won approval from Digital to design a new family of computers called Prism. The hardware was based on a chip created in-house. The operating program, called Mica, would be a flesh design, though it would be able to run Vax applications. Cutler assembled nearly two hundred people for the project, which had all the trappings of a stand-alone company. He designed the first Prism computer for scientists who wanted specialized calculations known as vectors performed at blinding speeds. These calculations helped scientists simulate everything from the weather to a nuclear explosion to the composition of the ocean floor. Typically, so-called supercomputers, costing millions of dollars, were needed for such mathematically intensive simulations. Because of its innovative hardware and software, the first Prism computer could be priced at a tenth of the cost of a supercomputer.

Even as Cutler's team pushed ahead with Prism, rivals elsewhere at Digital skewered his plans. By June 1988, Cutler had been asked to revise or discard so many projects that he wasn't shocked when he was flown to Digital's headquarters and told of Prism's cancellation. Digital wished to disband Cutler's hardware group, forcing almost one hundred engineers to find work elsewhere at the company. This, Cutler felt, was "the last straw." On the plane back to Seattle he drank heavily. The following morning, June 18, 1988, he still seemed tipsy when he entered a room packed with his staff. He was furious at Digital for letting his people down. He could afford to walk away; he had a two-hundred-thousand-dollar annual salary and a two-million-dollar home on Puget Sound. But he worried about the future of his people.

He looked at them before speaking. Some people started to cry; others went numb. A few veterans couldn't even face a defeated Cutler and rode out the meeting in a bar.

"You all look so sad," Cutler said.

Then he told them: Prism was dead. There was nothing he could do to stop it. He cried.

Gathering himself, Cutler gave everyone at the lab a month off -- with pay.

Copyright © 1994 by G. Pascal Zachary