Hackers

by Steven Levy

  The idea was not unappealing to Ken, who had announced to Softalk as early as March 1981 that he was “firing himself from the On-Line staff in hopes that [he’d] be able to get some programming done.” And surely it was clear that something had to be done about the managerial mess that was thickening as the company sold more software, took on more deals, tried to get hold of more programmers, and shuffled more paper, even if a lot of the paper was in the form of data handled in Apple computers.

  The problem came from Ken’s hacking On-Line as if it were a computer system, tweaking a marketing plan here, debugging the accounting there. Like his computer hacking, which was characterized by explosive bursts of innovation and inattention to detail, his business style was punctuated by flashes of insight and failures to follow through on ideas. He was among the first to recognize the value of a low-cost word processing package for the Apple (a culmination of the idea MIT’s Model Railroad Club hackers had when they wrote “Expensive Typewriter” on the TX-0), and had the patience to support the program through innumerable revisions—the program, eventually called “Screenwriter II”, would gross over a million dollars in sales. But his friendly competitors would laugh at his habit of writing huge royalty checks for programmers on the same checkbook he used for his supermarket accounts. He would help develop a program called “The Dictionary,” which corrects an Apple user’s spelling, but then would place a magazine advertisement for the product which contained ten spelling errors, including a misspelling of the word “misspell.”

  Ken’s new office was just about buried in junk. One new employee later reported that on first seeing the room, he assumed that someone had neglected to take out a huge, grungy pile of trash. Then he saw Ken at work, and understood. The twenty-eight-year-old executive, wearing his usual faded blue Apple Computer T-shirt and weather-beaten jeans with a hole in the knee, would sit behind the desk and carry on a conversation with employees or people on the phone while going through papers. The T-shirt would ride over Ken’s protruding belly, which was experiencing growth almost as dramatic as his company’s sales figures. Proceeding at lightning pace, he would glance at important contracts and casually throw them in the pile. Authors and suppliers would be on the phone constantly, wondering what had happened to their contracts. Major projects were in motion at On-Line for which contracts hadn’t been signed at all. No one seemed to know which programmer was doing what; in one case two programmers in different parts of the country were working on identical game conversions. Master disks, some without backups, some of them top secret IBM disks, were piled on the floor of Ken’s house, where one of his kids might pick it up or his dog piss on it. No, Ken Williams was not a detail person.

  He knew it, too. Ken Williams came to believe that his company had grown so big it had to be run in a more traditional manner by someone without hacker tendencies. Finally, he came up with a candidate. His former boss, Dick Sunderland.

  Ken knew Dick Sunderland as a representative of the vague qualities that a respectable business should have, qualities that On-Line conspicuously lacked: predictability, order, control, careful planning, uniform outlook, decorum, adherence to guidelines, and a structured hierarchy. It was no accident that these missing qualities were things that hackers loathed. If Ken had set out to find someone who best represented the antithesis of the Hacker Ethic, he would have been hard-pressed to top his former boss. The act was akin to someone acknowledging that he was sick, and perversely choosing the worst-tasting medicine as a restorative.

  There was something more insidious in the choice as well. One reason why Ken had left Informatics several years earlier was that Dick Sunderland had told him, “Ken, you have no management potential.” The idea of being Dick Sunderland’s boss, therefore, appealed greatly to Ken’s affection for toppling the established order.

  For Dick Sunderland, the prospect of working for Ken Williams initially struck him as absurd. “Come up and run my company!” Ken had chirped to him over the phone from this mountain complex near Yosemite. This was no way to recruit executives, thought Dick. There is no way, he told himself, I am going to get mixed up in a deal like this. Dick was completing an MBA program, a move which he felt would put him in line for the very top positions at Informatics. But by the time Ken called him a second time, Sunderland had been worrying about his future at Informatics, and had been thinking of the booming microcomputer field. In early June, Dick drove up, and had lunch at the Broken Bit with the motley crew of Oakhurst retreads and college dropouts that made up Ken’s upper management. He looked at the venture capital deal and was impressed. Eventually, he came to think that On-Line, as he later put it, “had a hell of a potential, something I could work with. I could bring what was missing-cohesive leadership to make things jell.” Dick realized the home software industry was “new, like clay . . . you could mold it and make it happen, make a winner . . . BOOM! It was the opportunity of a lifetime for me.”

  On the other hand, he would be working for Ken Williams. For over a month, Dick and his wife April would spend hours sitting in the backyard of the Los Angeles house they had carefully decorated over the years, kicking around this fantasy that would mandate their evacuation from the house, and it would be clear that the number one risk was the personality of this wild programmer-turned-software-czar. Dick consulted professionals to discuss what it would be like, a careful manager working for this reckless entrepreneur; he spoke to management experts, even a psychiatrist. Sunderland became convinced he could handle the Ken Problem.

  On September 1, 1982, Dick Sunderland began as the president of On-Line Systems, which coincidentally was also changing its name. Reflecting the proximity of Yosemite, the company would now be called Sierra On-Line, and the new logo had a drawing of Half-Dome Mountain in a circle. A change to accommodate the new age.

  A week before Dick arrived, Ken was feeling expansive. It was the day that he drifted over to give his blessing to the hacker who had “auditioned” with his Wall Wars game. After that encounter, he talked to a visitor about the potential stardom of his charges. He admitted that some of his authors had become brand names, almost like rock stars. “If I release a game and put the John Harris name on it, it will sell a ton more copies than if I don’t,” he said. “John Harris is a household name in [Atari] households. Among Atari computer owners, probably a higher percentage have heard of John Harris than [of] most rock stars.”

  But now that Dick Sunderland’s approach was imminent, Ken was hoping that the programmers’ power would be lessened. He was now a hacker who was convinced that hackers should be stifled. He was counting on Dick to get the standard programmer royalty down from thirty percent to twenty. “I don’t think you need a genius of programming” to make a hit game, said Ken. “The days of needing an A-student programmer to write an acceptable game aren’t over, but within a year of being over. Programmers, they’re not a dime a dozen, but they’re 50K a dozen. Moving the spaceship [on the video display] isn’t a problem anymore, What’s needed is to guess what the marketplace wants, access to the distribution channels, money, gimmicks, marketing promotion.”

  Sitting in his office that day, speaking in his startlingly candid what-the-hell tone, guessing that his company would “either be $200 million in sales by 1985 or bankrupt,” adding “I’m not real hung up on which,” Ken Williams promised to retreat to the mountain, like some high-tech pilgrim, and contemplate the next step in bringing about the computer millennium.

  But to the surprise of almost no one, Ken Williams did not keep his promise to “fire himself.” It would have been as out of character as a hacker abandoning a hot game program before all the proper features were written into it. Ken Williams had presented the company to Dick as if his goal—getting a company to a point where it was big enough to be left to a manager—were accomplished. But like a hacker, Ken Williams did not see things in terms of goals. He was still enamored of the process of running On-Line, and the clash of cultures between hacker informalism and bureau
cratic rigidity threw the company into turmoil.

  It was almost as if a fight were being waged for the soul of the industry. Among the first things Dick Sunderland tried to impose at Sierra On-Line was a rigid corporate structure, a hierarchy in which employees and authors would only be permitted to take up problems with immediate superiors. Dick requested the secretaries to distribute copies of the organization chart, with a box at the top for Ken, one underneath for Dick, and a series of boxes underneath, all connected by lines which represented the only authorized channels of communication. That this approach was antithetical to hackerism did not disturb Dick, who felt that hacker attitudes had almost brought the company to bankruptcy and ruin.

  Dick particularly wanted an end to Summer Camp. He had heard stories about the rowdy goings-on, the drugs, the impromptu parties, the pranks during working hours . . . he’d even heard from the janitorial staff that there’d been actual fucking in the office at night! Those kinds of things had to stop. He particularly wanted Ken to maintain a more executive-like relationship with his employees and to promote more orderly, rational lines of communication. How can you maintain a hierarchical structure when the chief executive gets in his hot tub with low-level employees?

  To Dick’s mind, the flow of information should be channeled with discretion, with an unambiguous interpretation controlled by the people at the top. People who don’t have the broad view of things should not be upset by getting dribs and drabs of information. What Dick had to contend with at On-Line, though, was an incredible rumor mill, fed by the unfettered flow of information that company had been accustomed to. And Ken Williams, Dick said, “nurtures [the rumor mill] rather than quells it. He has no sense of discretion!” Everything was public record with Ken, from his personal life to his bank account.

  Dick was convinced, though, that Ken knew On-Line needed responsible management, or it would die. But Ken was so reluctant to step back. Sunderland could settle the personnel situation, bring in carefully considered candidates, keep the payroll under control . . . and then Ken would tell him, bang, that he just hired somebody to be his administrative assistant, a job opening that did not exist until that very minute. “And who did he hire?” Dick would say. “Some guy driving a Pepsi truck in L.A.”

  “This is casebook stuff.” Dick said. He recalled reading about it in business school: entrepreneur who gets going on a brilliant idea, but can’t handle it when the business gets big. It all came from the hacker origins of the company. Ken was saying that the time for hackers was over; he wanted to limit programmers’ power in the company. But he wasn’t making it easy for Dick.

  It was particularly tough trying to negotiate the royalty down from thirty to twenty percent when the programmers had the impression that the company was rolling in money. It really wasn’t, but no one believed that when they saw green just about falling from the windows. Everyone knew about the house Ken was building outside of town. It would be four hundred feet long. A party room that would be the biggest in the area. A crew of over a dozen were working full-time on it . . . they had constructed an entire office on the work site, with phone hookups and everything. The house was not even half finished, and already Ken had invited the whole company to come to the site on weekends to play in the built-in racquetball court. It was not the best way to convince programmers to opt for austerity.

  Ken Williams’ point of view was somewhat different. He had hired Dick, and would often defend him. But he thought it necessary to keep his hand in. Ken felt responsible to the people he had hired and to the vision of the company itself. He knew the industry as well as anyone; Dick was a newcomer to the family. Also, Ken Williams was having too much fun: leaving now would be like walking away from a crap table when you’re on the hottest roll of your life. Or, more to the point, it was like telling a hacker that he could no longer play with the machine. Those words did not register with hackers. Once you had control, the godlike power that comes from programming mastery, you did not want to let go of it.

  Roberta Williams would agree. Just as Ken treated On-Line like a complex computer program to be hacked, Roberta thought of the company as a creative project which should be lovingly embellished and elegantly structured, like an adventure game. Like authors of an adventure game, she and Ken had enjoyed ultimate control over the company; turning it over was difficult. She compared the situation to hiring a governess: “You would think, wouldn’t it be great to have someone come in and watch the kids every day while I’m doing this thing I want to do. I can design adventure games. But then she starts telling the kids everything they can do—‘Oh yes, you can have a peanut butter and jelly sandwich.’ And I may not believe in peanut butter and jelly. I might prefer them to have beef. That person says, ‘Peanut butter is good, there’s a lot of protein in peanut butter. You hired me, let me do my job.’ That’s what we’re running into with Dick. He said, ‘You gave me the power to do this, you wanted to go off and program.’ Now we are saying, ‘Yeah, that’s what we thought we wanted, but it turns out we don’t want to give up control.’”

  • • • • • • • •

  While the management of Sierra On-Line struggled to find itself, the Third-Generation hackers there were glum over the changes in their company. They would talk over frozen dinners at the Hexagon House before playing Dungeons and Dragons. Or they would discuss the deteriorating moral state of the company over pizza and Cokes at Danny’s, a bleak roadhouse on Route 41 with picnic-style tables covered with plastic checkered tablecloths. Most of the customers were local families who didn’t seem to like the On-Line people much, but it was almost the only place in town where you could get a pizza and play videogames, which the hackers played compulsively, with no visible sense of involvement or even interest, while they waited for their food.

  They were proud of their positions and almost puzzled at their good fortune in getting paid for work they loved. In the early 1980s, hacking games was about the only form of commercially viable artistry where, with almost no capital, you could truly be an auteur: single-handedly you could conceive, script, direct, execute, and polish a work, completing an objet d’art which was every bit as good as the bestselling game on the market. This Third Generation found itself in an artistically privileged position. The fact that publishers competed for their wares made things pleasant on the one hand but confusing on the other. There were no rules for this kind of thing. It was a rare twenty-year-old hacker who had the business acumen and intestinal fortitude to cope with a negotiator as forceful as Ken Williams or as formally intimidating as Dick Sunderland. Since money wasn’t the main issue for the hackers, they’d agree to almost anything if they thought it was fair. Business wasn’t as much fun as hacking was.

  Still, in the fall of 1982, it was the most creative programmers who drove the industry. Brøderbund was riding high on Choplifter, written by a twenty-eight-year-old former artificial intelligence hacker named Dan Gorlin. The game was based on the Iran hostage crisis: a chopper crossed enemy lines and tried to rescue sixty-four hostages—little animated figures who waved when they saw the helicopter. It was the big game of the year, and consistent with the Carlstons’ classy approach to the business. They loved their hackers. They talked all the time about what great artists their “game designers” were.

  Sirius had been developing its own superstars, but Gebelli, the designer who had done almost all their games in the first year of Sirius’ existence, was not one of them. According to Jerry Jewell, Gebelli thought that Sirius was not the best agency for display and sale of his artworks—this after receiving a quarter of a million dollars in his first year, noted Jewell incredulously—and, along with a defecting Sirius executive, began his own company, modestly named Gebelli Software. It did not join the top ranks of the industry.

  Sirius survived the loss by importing teen-age hackers from other parts of the country, and they delivered some hit games called Beer Run, Twerps, and The Earth Dies Screaming. Jerry Jewell acted as a sometimes rowdy big b
rother to his young programmers. What Jewell really lusted after was the mass VCS market, and after signing a major deal to develop games for Twentieth-Century Fox’s new videogame division he was afire with visions of his products as household words, not just in the Apple or Atari world, but everywhere. He figured that some of his programmers might make as much as a million dollars a year.

  At On-Line, where the VCS had been a mere flirtation, Ken Williams and Dick Sunderland were not talking about a million dollars a year for their programmers. They were trying to cut the royalty down from thirty to twenty percent. And when On-Line hackers gathered at places like Danny’s, they would compare notes and find that they were in agreement: thirty percent was fair, and twenty percent was not. Brøderbund and Sirius were still offering higher royalties. Some of the hackers had been approached by an exciting new company called Electronic Arts. It consisted of ex-Apple people who promised to treat hackers as culture heroes, like rock stars.

  Ken and Dick had tried to convince them that twenty percent was a fair figure in light of the drastically increased costs of promoting and testing and distributing a game in this new, more professional stage of the industry. On-Line was increasing its advertising, hiring more support people, boosting its promotional staff. But the programmers saw Sunderland and his regime as bureaucracy, to which, as hackers, they had a generic allergy. They missed the days of Summer Camp and handshakes for contracts. John Harris, for instance, chafed at the idea of paying a lawyer to help him negotiate a six-figure contract (“They charge one hundred dollars just to read it!” he howled). Harris and the other On-Line hackers would see all these managers and support people being hired, just to do the same thing that the company did before—release the games that the hackers wrote. From their point of view, it seemed to indicate another hacker sin—inefficiency. Along with an emphasis on the sizzle of marketing rather than the substance of hacking.