by Steven Levy
Bob Marsh, hawking Processor Technology boards, often flew down to the packed SCCS meetings, and even sat on the SCCS board for a few months. He later described the difference between the two groups: “Homebrew was a place where people came together mysteriously, twice a month. It never was an organization. But SCCS was more organized. Those guys had megalomania. The politics were terrible, and ruined it.” Somehow, the particulars never became clear, a lot of money was misplaced in the buying scheme. The editor they hired to run the slick magazine felt justified in dropping the publication’s relationship with the club and going off on his own with the magazine (still publishing as Interface Age); a lawsuit resulted. The board meetings became incredibly tempestuous, and the bad feelings spread to the general membership meetings. Eventually the club faded away.
Though Lee’s plans were no less ambitious than those of the leaders of SCCS, he realized that this war must not be waged in a bureaucratic, follow-the-leader fashion. He was perfectly happy dealing with an army of Bob Marshes and Tom Pittmans, some changing the world by dint of useful products manufactured in the spirit of hackerism and others just going their way, being hackers. The eventual goal would be a mass distribution of the wonderment that Lee Felsenstein had experienced in his basement monastery. An environment conducive to the Hands-On Imperative. As Lee told a conference of the Institute of Electrical and Electronic Engineers in 1975, “The industrial approach is grim and doesn’t work: the design motto is ‘Design by Geniuses for Use by Idiots,’ and the watchword for dealing with the untrained and unwashed public is KEEP THEIR HANDS OFF! . . . The convivial approach I suggest would rely on the user’s ability to learn about and gain some control over the tool. The user will have to spend some amount of time probing around inside the equipment, and we will have to make this possible and not fatal to either the equipment or the person.”
The piece of equipment to which Felsenstein referred was his Tom Swift Terminal, which still had not been built in 1975. But it was getting close. Bob Marsh, eager to expand the scope of his booming Processor Technology company, offered Lee a deal he couldn’t refuse. “I’ll pay you to design the video portion of the Tom Swift Terminal,” he told him. That sounded all right to Lee, who had been doing work in documentation and schematics for Processor Technology all along. Bob Marsh, in the company’s first year of business, was adhering to the Hacker Ethic. The company distributed schematics and source code for software, free or at nominal cost. (In partial reaction to MIT’s high-priced BASIC, Processor Technology would develop its own and sell it, along with source code, for five dollars.) For a time, the company had a socialistic salary structure of $800 a month for all employees. “We didn’t pay attention to profits or management of almost any kind.”
Lee was not an employee, choosing to work on a contract basis. “I’d quote them a price,” Lee later recalled, and “they had to get the price up by a factor of ten, since I was such a small-time thinker. In terms of money.”
In less than three months, Lee had done a working prototype. Lee’s “video display module” (VDM) embodied a different philosophy than the other video board for Altair, Cromemco’s Dazzler. The Dazzler used color, and produced its flashy effects by constantly going back to the memory in the main chip of the Altair (or any of the other new computers that used a similar hardware bus). Steve Dompier liked to use his Dazzler while running BASIC: it threw up patterns on the screen that gave a Rorschach-like visual impression of the computer memory at a given time—a cryptic output which gave clues to program operation, much like the aural impression given of the TX-0’s memory by the speaker under the console.
Lee’s video display module, though, was a more stridently focused piece of equipment, designed with the eventual re-formation of Community Memory in mind. Its output was black and white, and instead of using dots it actually formed alphanumeric characters. (Lee considered adding another alternative—hexagrams, as found in the I Ching—but that idea got shelved somehow.) The cleverest thing about Lee’s VDM, though, was the way it used the speed of new microprocessor chips to allow the machine’s memory to be shared between computational duties and display duties. It worked like a mini-time-sharing system, where the two users were the video display and the computer itself. The VDM, along with an Altair and other expansion cards, made the promise of the TV Typewriter a reality, and was an instant success, even though it was, like almost every Processor Technology product, not ready till somewhat after the promised delivery date, in late 1975.
One person particularly impressed by the VDM was Les Solomon in New York. He was not content to bask in the reflected glory of launching Ed Roberts’ seminal machine. His magazine had followed up on the coup, he had delivered more computer-related cover stories, and now he was hoping to present a complete computer video display terminal—a self-contained item which would have the power of the computer as well as a display capability. It would be the next step beyond the Altair, a combination computer-teletype with video. No more goddamn bloody fingers from the flicking switches on the Altair front panel. In pursuit of the product, Solomon went to Phoenix to visit Don Lancaster, inventor of the TV Typewriter (the one Bob Marsh had tried to build in Berkeley), and convinced him to drive down to Albuquerque to meet Ed Roberts, maybe the two giants might combine on a terminal project. As Solomon later described it, the meeting was “bang, clash. A clash of egos. Don refused to change his design to match Ed’s computer because he said Ed’s was inefficient. Ed said, ‘No way, I can’t redesign it.’ They immediately decided to kill each other on the spot, and I separated them.”
So Solomon went to Bob Marsh, whose Processor Technology company already offered the VDM and memory boards and even a “motherboard” which could replace the basic circuitry of the Altair, and asked, “Why don’t you put them all together? Let’s make something we can look at.” If Marsh could deliver an “intelligent terminal” in thirty days, Solomon would put it on the cover.
Bob talked to Lee, who agreed to do most of the design, and as they discussed it they realized that what Les Solomon wanted was not merely a terminal, but a complete computer. In the year since the Altair had been announced, “hobbyist” computers, sold either in kits or assembled, had appeared, most notably one called the IMSAI, put out by a company whose employees had taken Werner Erhard’s est training. Almost all of these computers used the 100-pin Altair bus as their base. Almost all looked like the Altair, an oversized stereo receiver with lights and switches instead of an FM dial. All required some sort of terminal, usually a klunky teletype, for the user to do anything with it.
For that month, December of 1975, Lee and Bob worked on the design. Marsh wanted to use an 8080 chip, an idea which Lee at first still opposed for political reasons (why one centralized silicon dictator?) but came to accept as he realized that a truly “intelligent” terminal—one which gave you all the power of a computer—would need a brain. Lee figured he would use his junkyard-paranoid style to balance out the rest of the design, so that the brain would not be tempted to run amok. Marsh would often interrupt Lee’s design-in-progress to reveal his latest inspiration from the “feature creature.”
Lee later recounted this process in a magazine article: “When [Marsh] had little else with which to concern himself, he was continually turning up with new features and economies that he suddenly wanted incorporated in the design. He would explain the problem or opportunity and then preface his technical solution with an inevitable, ‘All’s ya gotta do is . . .’ Were the designer a prima donna, the relationship would terminate after the second such incident, with the designer fuming about ‘professionalism’ and ‘interference.’ Of course since my workshop was in the same room as his, I could not have gotten very far if I had wanted to stamp out in a rage.”
Marsh, like Lee, was thinking of the machine as a political tool as well as a good, fun product to design. “We wanted to make the microcomputer accessible to human beings,” he later said. “The public didn’t know it yet, but the c
omputer was the coming thing and every home would have one and people could use computers for useful things. We really weren’t sure what they were [but] we felt we were participating in a movement, in a way.”
Lee suggested that since they were putting the wisdom of Solomon into the machine, it should be called the Sol. (Les Solomon later commented: “If it works, they’ll say Sol means ‘sun’ in Spanish. If it don’t work, they’re gonna blame it on the Jewish guys.”)
Completing the Sol was a process that took six weeks of fourteen to seventeen-hour days, seven days a week. Lee, just about living on orange juice, spent endless hours staring at the Mylar spaghetti of the layout on the fluorescent light table. Meanwhile, one of Bob Marsh’s woodworker friends had managed to get a bargain on center-cut pieces of walnut, and it was determined that the sides of the Sol would be made of that classy material. The prototype boards were finally finished, only fifteen days after Les Solomon’s original deadline. Two weeks later, a day before the newly scheduled delivery date in late February 1976 in New York, they were racing to get all the workings to fit on an Altair-style bus, along with a kluged-up power supply, a keyboard, and even some preliminary software. The operating system was written by Processor Tech’s head of software development, Homebrewer Steve Dompier.
Ever frugal, Marsh had booked himself and Lee on a night flight. Finishing just in time, they had to race to a heliport in order to make the plane. They arrived at Kennedy around 6 A.M., frazzled, with the Computer of the Common Man distributed between two paper bags. Nothing was open at the airport, even for coffee, so Solomon invited them over to his home in Flushing for breakfast. By then Les Solomon’s home, particularly his basement workshop, was achieving legendary status as a proving ground for thrilling new breakthroughs. He would often entertain the young hardware hackers who designed these products, and his wife would always recognize them at a glance. “Because they all had the same thing,” Solomon would later explain. “That little burning inside the eyeball. She used to say there was an inside personality, and though they looked like disreputable bums, you looked them in the face, you looked in those eyes and you knew who they were. She’d look at them and what would come out was the brightness, the intenseness.”
The brightness dimmed on that cold February morning: Marsh and Felsenstein’s terminal didn’t work. After a quick day-trip to New Hampshire to meet the folks at the new hobbyist magazine Byte, Lee was able to get to a workbench and find the problem—a small wire had come loose. They went back to the offices of Popular Electronics and turned it on. “It looked like a house on fire,” Solomon later said. He had immediately grasped that he was looking at a complete computer.
The resulting Popular Electronics article spoke of an intelligent computer terminal. But it was clearly a computer, a computer that, when Processor Technology packaged it in its pretty blue case with walnut sides, looked more like a fancy typewriter without a platen. There were new schematics for the revised kit (under one thousand dollars), which of course were provided to anyone who wanted to see how the thing worked. Marsh later estimated that they got thirty to forty thousand requests for schematics. Orders for the kit kept pouring in. It looked like the Sol would be the machine that broke the computer out of the hobbyist market and brought hacking into the home.
The first public display of the Sol was at a show in Atlantic City called PC ’76. It was an odd affair, the first time the tradesmen of this hobbyist-computer business all got together to show their collective wares. The site was the Shelbourne Hotel, and in those pre-gambling days the hotel’s glory was visibly faded. There were holes in the walls, some of the doors to the rooms had no knobs, the air conditioning didn’t work. Some indignant elderly retirees living at the hotel almost attacked Steve Dompier in the elevator when they saw his long hair. Still, it was an exhilarating experience. Almost five thousand people attended, many of them traveling from other parts of the country (SCCS ran a large group excursion which many Bay Area people took advantage of). Homebrew-inspired companies like Processor Tech and Cromemco finally met similar souls from other parts of the country, and everybody stayed up far into the night, swapping technical hints and plotting the future.
The Sol got lots of notice. The hackers all seemed to agree that with its low profile, its typewriter-style built-in keyboard, and its video display, the Sol was the next step. Not long afterward, Processor Tech managed to get a Sol on television—on Tom Snyder’s “Tomorrow” show. The normally abrasive television personality came face to face with the newest manifestation of the hacker dream—a Sol computer running a game program written by Steve Dompier. The game was called Target, and it consisted of a little cannon on the bottom of the screen by which the user could shoot down a series of alien spaceships, made of alphanumeric characters, sailing across the top of the screen. It was a clever little hack, and Steve Dompier, as he later said, “basically gave it away.” After all, the point of writing those games was to see people have fun with the machine.
Target was perfect for showing Tom Snyder and a television audience a new way to look at those monsters shrouded in evil, computers. Imagine these grungy post-hippies being able to bring a computer over to a television studio, set it up, and have a total technical illiterate like Tom Snyder do something with it. Tom went along, and before you could say “commercial break” he was deeply involved—not in the least kidding—in shooting down aliens, which would zip across the screen in greater numbers as the game progressed, and even dispatch little parachutists loaded with grenades. It gave you a challenge you felt compelled to rise to. As you shot down the aliens, Tom Snyder was noticing, there was this feeling of . . . power. A feeling that gave you a small taste of what it must be like to use this machine to actually create. What mysteries lay within this typewriter-shaped machine? Even something as simple as Target could get someone thinking about that. “No one’s given it a definition yet,” Steve Dompier later said, “but I think there’s a piece of magic there.” In any case, as Dompier later recalled, “they had to drag Tom Snyder off the computer to have him finish the show.”
Chapter 12. Woz
Steve Wozniak did not sit near the front of the SLAC auditorium along with Lee Felsenstein during Homebrew meetings. His participation in the mapping sessions were infrequent. He had no great social scheme, did not incubate plans for a Community Memory-style assault on the foundations of the batch-processed society. Meeting after meeting, Steve Wozniak would be at the back of the room, along with a loose contingent of followers of his digital exploits—mostly high school-age computer nuts drawn by the sheer charisma of his hacking. He looked like a bum. His hair fell haphazardly on his shoulders, he had the kind of beard grown more to obviate the time-consuming act of shaving than to enhance appearance, and his clothes—jeans and sports shirts, with little variation—never seemed to fit quite right.
Still, it was Steve Wozniak, known to his friends as “Woz,” who would best exemplify the spirit and the synergy of the Homebrew Computer Club. It was Wozniak and the computer he’d design that would take the Hacker Ethic, at least in terms of hardware hacking, to its apogee. It would be the legacy of the Homebrew.
Stephen Wozniak did not reach his views of hackerism through personal struggle and political rumination as Lee Felsenstein did. He was more like Richard Greenblatt and Stew Nelson: a born hacker. He grew up in Cupertino, California, amidst the curving streets lined with small single-family homes and the one-story, sparsely windowed buildings that sowed the crop of silicon which would be so central to his existence. Even in grammar school, Wozniak could get so engrossed in mathematical ponderings that his mother had to rap on his head to bring him back to the real world. He won a science contest at thirteen for building a computer-like machine which could add and subtract. His friend Alan Baum later remembered him at Homestead High School: “I saw a guy scribbling these neat diagrams on a piece of paper. I said, ‘What’s that?’ He said, ‘I’m designing a computer.’ He had taught himself how to do it.
”
Baum was impressed enough to join this unusual classmate in a quest for computer access, and through contacts in the engineering-rich Silicon Valley they managed to get on various time-sharing computers. Every Wednesday they would leave school and have a friend sneak them into a computer room at the Sylvania company. They’d program the machine to do things like printing out all the powers of two and finding the primes. The two followed the computer industry with the serious passion with which fanatic sports fans might follow favorite teams. Every time they heard of a new minicomputer being released, they would write to the manufacturer, be it Digital or Control Data or whoever, and request the manual, a request often routinely fulfilled. When the manual came, they would devour it. They would instantly turn to the part which described the computer’s instruction set. They would note how many registers the machine had, how it added, how it did multiplication, division. They could discern from the instruction set the character of the machine, how easy it would be to use. Was this a machine to fantasize about? If it was, Woz later recalled, he would “spend hours in class writing code without ever being able to test it.” Once, after receiving a manual for a Data General Nova computer, he and Baum took it upon themselves to redesign it, even sending their new design to the company, in case Data General wanted to implement the suggestions of two high school kids.