Hackers

by Steven Levy

  No longer was it a struggle, a learning process, to make computers. So the pioneers of Homebrew, many of whom had switched from building computers to manufacturing computers, had not a common bond, but competition to maintain market share. It retarded Homebrew’s time-honored practice of sharing all techniques, of refusing to recognize secrets, and of keeping information going in an unencumbered flow. When it was Bill Gates’ Altair BASIC that was under consideration, it was easy to maintain the Hacker Ethic. Now, as major shareholders of companies supporting hundreds of employees, the hackers found things not so simple. All of a sudden, they had secrets to keep.

  “It was amazing to watch the anarchists put on a different shirt,” Dan Sokol later recalled. “People stopped coming. Homebrew [still moderated by Lee Felsenstein, who kept the hacker fire burning] was still anarchistic: people would ask you about the company, and you’d have to say, ‘I can’t tell you that.’ I solved that the way other people did—I didn’t go. I didn’t want to go and not tell people things. There would be no easy way out where you would feel good about that.”

  Homebrew still drew hundreds to its meetings, and its mailing list was over fifteen hundred—but there were many novices there, with problems that weren’t challenging to old hands who’d built machines when machines were nearly impossible to build. It no longer was essential to go to meetings. Many of the people involved in companies like Apple, Processor Tech, and Cromemco were too damned busy. And the companies themselves provided the communities around which to share information.

  Apple was a good example. Steve Wozniak and his two young friends, Espinosa and Wigginton, were too busy with the young firm to keep going to Homebrew. Chris Espinosa later explained: “[After the Computer Faire] our attendance at Homebrew started dropping off and ended completely by the end of the summer of 1977. We, in effect, created our own computer club [at Apple] that was more focused, more dedicated to producing things. When we started getting involved with Apple, we found what we wanted to work on and we wanted to spend all our time perfecting it, expanding it, doing more for it, and we wanted to go into one subject deeper rather than covering the field and finding out what everybody was doing. And that’s how you make a company.”

  In some senses, the “computer club” at Apple’s Cupertino headquarters reflected the same community feeling and sharing of Homebrew. The company’s formal goals were traditional—making money, growing, gaining market share—and some secretiveness was required even of Steve Wozniak, who considered openness the central principle of the Hacker Ethic he fervently subscribed to. But this meant that the people within the company could be even closer. They had to depend on each other to swap suggestions for floating-point BASIC or parallel printer cards. And sometimes, the community was loose enough to accept some old Homebrew friends. For instance, in mid-1977, John Draper appeared.

  The former “Captain Crunch” was in a bad way. Apparently certain authorities had objected to his willingness to share phone company secrets with anyone who bothered to ask; FBI agents trailed him and, according to his accounts of the incident, planted an informer who talked him into a blue-box escapade while agents waited to bust him. For this second conviction, he was sentenced to a brief jail term, and incarceration did not agree with the normally contentious Captain, a person taken to screaming like a six-foot-tall hyena if someone lit a cigarette twenty feet away from him. After his release, he needed legitimate work badly, and Woz got him hired as a consultant, designing a telephone interface board, something that would plug into one of the Apple’s expansion slots to allow you to connect the phone to your computer.

  Draper happily worked on the board. The people at Apple were amused by his programming style, which alternated bursts of brilliance with bizarre pedantic detours. Draper was a “defensive” programmer. Chris Espinosa, who had the unenviable task of trying to keep an eye on the unpredictable Captain, would later explain: “Say you’re writing a program and you discover you’ve done something wrong, like every time you try to use the program, a button pops up. Most programmers go in, analyze their program, find out what causes the button to pop up and cure it so it doesn’t do that. Draper would go in and code around the button so when the bug occurs, the program knows it’s made an error and fixes it, rather than avoiding the error in the first place. The joke is, if Draper were writing math routines for addition and he came up with the answer 2 + 2 = 5, he would put a clause in the program, if 2 + 2 = 5, then that answer is 4. That’s generally the way he writes programs.”

  But while the hackers at Apple were amused that the strange style of John Draper was turning out a featureful product, the people in charge of the business end of Apple got wind of the capabilities of Draper’s design. They did not like it. Apple was not a showcase for tricks; this was not Homebrew. And John Draper’s board could do some considerably neat tricks; not only did it interface with the phone, but it generated official phone company tones—it was a computer-driven blue box. What Stew Nelson had done with the PDP-1 over a decade ago could now be done in the home. The hacker instinct would have been to explore the capabilities of this hardware, which would enable you to explore systems all over the world. But though Apple felt it could benefit by the Hacker Ethic in distributing information about the innards of the machine and distributing its computers as complete systems to explore, it was not in the business of promoting pure hackerism. It was, after all, a business, with a line of credit and a truckload of venture capital provided by men in three-piece suits who did not relate to concepts like phone hacking. “When Mike Scott discovered what [Draper’s board] could do,” Espinosa later said, “he axed the project instantly. It was much too dangerous to put out in the world for anybody to have.”

  Killing that project was well in keeping with the propriety of the booming Apple Computer Company, which was selling computers like mad, and becoming respectable at a pace which had the Homebrew alumni dazzled. Randy Wigginton, for instance, realized by late summer in 1977 that this company had far eclipsed your normal growth story. That was when everyone went to Mike Markkula’s for a party to celebrate shipping a quarter-million dollars’ worth of equipment that month. It was only the beginning of a climb that turned Apple into a billion-dollar company within five years.

  During this period when everybody at Apple was celebrating the increasing revenues—piles of money that would make many of them so rich that they would be beyond millionaires, in the ozone of Croesus Mode, where wealth is counted in units of tens of millions—John Draper was at home, playing with his Apple. He set the completed board into his Apple II. He connected it to the telephone line. And he set it up so that it would “scan” entire telephone exchanges, looking for telltale tones which would inform him that a computer was on the other side of the line. A virgin computer that a hacker could enter and explore. He had hacked a program by which the computer could dial on its own. “It seemed like an innocent thing to do,” he later said. On its own, the computer began making a hundred and fifty calls an hour. Every time it discovered a computer at the other end of the line, the teletype printer attached to the machine would grind out the telephone number. After nine hours, John Draper would have a printout of every computer number in an entire three-digit exchange. “I just collected them,” he would later explain. The setup could also detect WATS Extenders service numbers, with which one could make free long-distance calls. (It was John Draper’s system which later would be the model for a young hacker’s computer break-in in the movie WarGames.)

  Unfortunately, the ever vigilant system that was the phone company had developed some new phone-hacking detection equipment. John Draper’s unprecedented output of over twenty thousand phone calls in under a week not only signaled that something was awry, but also exhausted the paper supply in the phone company printer which logged such irregularities. John Draper was confronted with another visit from the authorities. It was his third conviction, his first using a home computer. An inauspicious beginning for a new era of phone hac
king with personal computers.

  • • • • • • • •

  Some thought that the establishment of an industry of low-cost personal computers meant the war was won. They believed the widespread proliferation of computers and their innate lessons of openness and creative innovation would, in and of itself, spur the Hacker Ethic. But for Lee Felsenstein, the war was just beginning.

  His consuming passion was the resurrection of Community Memory. He still stuck to the dream whose glory he had glimpsed in the experiment at Leopold’s Records. It was perhaps exquisite irony that the development of the small computer industry had been aided in part by the introduction of the Pennywhistle modem, the VDM video board, and the Sol computer, all pieces of the mythic Tom Swift Terminal, a machine which could reach fruition only in the publicly accessed terminals of Community Memory branches. Irony, because a growing consensus among Lee’s peers held that the once bold Community Memory concept—and the Tom Swift Terminal itself—had been supplanted by the rapid acceptance of home computers. It was fine to desire a public terminal to be the heart of an information center that would be an “amalgam of branch libraries, game arcades, coffee houses, city parks, and post offices.” But why would people leave the house to go to a CM terminal when they could use an Apple Computer, along with a telephone interface right there at home, to communicate with any database in the world?

  The Tom Swift Terminal itself might have been shelved, but Lee still held to his goals. The science-fiction novel in which he was protagonist was taking bolder plot twists, confirming that it was a major work indeed. In the two “unforgettable years” since the triumphant Computer Faire, he had seen a company crumble. Processor Technology had suffered too much growth and too little sound management to survive. Through the whole year of 1977, orders for the Sol came in at a rate beyond the capacity of the company to fulfill them. In that fiscal year, Bob Marsh later estimated, the company did five and a half million dollars’ worth of sales, selling perhaps eight thousand machines. It moved into a clean, thirty-six-thousand-square-foot headquarters east of the Bay Area.

  But even as the future looked bright, with Bob Marsh and Gary Ingram figuring that if sales got up to fifteen or twenty million they’d sell out and get rich, the company was doomed by lack of planning and failure to address the competition of the new, cheaper, sleeker machines like the Apple, the PET, the TRS-80. Marsh later said that the company was thinking of going into that lower end of the market, but was intimidated by the power of the competing firms that had announced complete computers in the $1,000-and-under range. He figured that PT could sell the Sol as a more expensive, quality item, like MacIntosh amplifiers in the audio business. But the company missed the chance to extend its equipment effectively when its disk drive storage system proved to be unreliable. And it was unable to deliver software for its machines on time. There would be announcements of upcoming products in the PT newsletter, a spirited publication which mixed bug reports with cryptic quotations (“‘There are no Jewish midgets’—Lenny Bruce”). Months late, the products, either software programs or hardware peripherals, would still be unavailable. When PT had an offer to sell Sol computers through a new chain of computer stores called Computerland, Marsh and Ingram refused, suspicious because the owners of the chain were the same people who ran the company (also struggling, soon to be bankrupt) which made the IMSAI computer. Instead of Sols being sold as computer-terminals at Computerland, Apples were.

  “It’s embarrassing to think how Mickey Mouse we were sometimes,” Marsh admitted later. There was no business plan. Things would not get delivered on time, credit would not be extended to priority customers, and the constant PT errors in delivery and unprofessionalism with suppliers gave the company a reputation for arrogance and greediness.

  “We were just violating some of the basic laws of nature,” Marsh later said. When sales flattened, the money to run the company wasn’t there. For the first time they looked for investors. Adam Osborne, an already established gadfly of the young industry, introduced them to people who were willing to invest, but Marsh and Gary Ingram did not want to give up a substantial percentage of the company. “Greedy,” Osborne later said. Some months later, when the company was almost bankrupt, Marsh came back to accept the offer. It was no longer open.

  “We could have been Apple,” Bob Marsh said, years later. “A lot of people say that 1975 was the year of the Altair, ’76 was the IMSAI, and 1977 was the Sol. The dominant machines.” But by the end of those “unforgettable two years,” the engineer-managed companies that made those machines, machines available in kit form as well as assembled, machines which hardware hackers loved to play with . . . were gone. The dominant small computers in the market were Apples, PETs, TRS-80s, in which the act of hardware creation was essentially done for you. People bought these machines to hack software.

  Lee Felsenstein was perhaps the biggest financial beneficiary in Processor Technology’s short history. He had never been an official employee, and his royalties on the Sol eventually totaled over one hundred thousand dollars. He was never paid the last twelve thousand in royalties. Most of the money went toward the new incarnation of Community Memory, which had set up a headquarters in a large, two-level, barn-like loft structure in a West Berkeley industrial area. Efrem Lipkin and Jude Milhon of the original group were among the dedicated members of the new CM Collective, all of whom vowed to work for long hours and subsistence wages to establish permanently the thrilling experiment they’d worked on earlier in the decade. It required extensive work in developing a new system; the collective decided that funding could come, in part, by writing software products for these small computers.

  Meanwhile, Lee was broke. “The rational thing for me to do would have been to shut down [my engineering] business and get a job. But I didn’t,” he later said. Instead, he worked for almost nothing, designing a Swedish version of the Sol. His energies were divided between that, the hopelessly earnest Community Memory meetings, and monthly Homebrew meetings, which he still proudly moderated. The club was famous now that microcomputers were being acclaimed as the chief growth industry of the country. And the prime example of this was Apple Computer, which would gross $139 million in 1980, the year it went public, making Jobs and Wozniak worth a combined sum of well over $300 million. Croesus Mode.

  That was the year that Lee Felsenstein ran into Adam Osborne at the Computer Faire. Jim Warren’s show was now an annual event pulling in fifty thousand people in a weekend. Osborne was a trim, Bangkok-born Englishman in his forties with a thin brown mustache and an imperious vanity which propelled his column in trade magazines (entitled “From the Fountainhead”) to notoriety. A former engineer, he made a fortune publishing books on microcomputers when no one else was. He would sometimes bring boxes of them to Homebrew meetings and go home with empty boxes and wads of cash. His books eventually sold hundreds of thousands, McGraw-Hill bought his publishing house, and now, “with the money burning a hole in my pocket,” as he said, he was looking to go into the manufacture of computers.

  Osborne’s theory was that all the current products were too much oriented toward hackers. He did not believe that people cared to know about the magic that hackers found within computers. He had no sympathy for people who wanted to know how things worked, people who wanted to explore things, people who wanted to improve the systems they studied and dreamed about. In Adam Osborne’s view, there was nothing to be gained by spreading the Hacker Ethic; computers were for simple applications, like word processing or financial calculation. His idea was to provide a no-frills computer which would come with all you needed to get going—Osborne thought people were happiest when relieved of anxiety-producing choices, like which word-processing program to buy. It would be cheap, and small enough to carry on a plane. A portable Volkscomputer. He asked Lee Felsenstein to design it. Because the machine he wanted need only be “adequate,” designing it should not be too hard a task. “Five thousand people on the peninsula could have done it
,” Osborne later said. “I happened to know Lee.”

  So for twenty-five percent of this as yet unformed company, Lee Felsenstein designed the machine. He chose to interpret Osborne’s requirement that the machine be “adequate” to mean he could do his usual job of junkyard engineering, making sure that the design was solid enough to support well-tested components in an architecture that eschewed tricks and detours. “To be able to make a design that is good and adequate, works well, and is buildable and cheap and contains nothing fancy is an artistic problem,” he later said. “I had to be crazy enough and broke enough [to try it].” But Lee knew that he could fulfill the requirements. As usual, there was fear in the equation: Lee had an admittedly irrational fear of Adam Osborne; he guessed he identified Adam with the authority figures of his childhood. There was no way that these two could communicate deeply. Once Lee tried to explain Community Memory to him—his real career—and Osborne “didn’t get it,” lamented Lee. “He may be one of the last people to get what Community Memory is about when he sees it, uses it.” Yet Lee worked hard for Adam Osborne, working in a space in the Community Memory headquarters, and in six months he was done.

  He had fulfilled, he thought, the technical requirements as well as the artistic ones in building the machine which was known as the Osborne 1. Critics would later say that the plastic-cased machine had an uncomfortably small five-inch screen, and note other small problems, but when the computer first came out praise was plentiful and the Osborne Computer was soon a multimillion-dollar company. And, out of nowhere, Lee Felsenstein was worth over twenty million dollars. On paper.