by Steven Levy
There were two Altair programs that could take advantage of the Dazzler then. One was called Kaleidoscope, and it shimmered and changed shape. It was a great moment for Solomon, seeing the computer he had helped bring to the world making a color television set run beautiful patterns.
Then they tried another program: LIFE. The game-that-is-more-than-a-game, created by mathematician John Conway. The game that MIT wizard Bill Gosper had hacked so intently, to the point where he saw it as potentially generating life itself. The Altair version ran much more slowly than the PDP-6 program, of course, and with none of those elegantly hacked utilities, but it followed the same rules. And it did it while sitting on the kitchen table. Garland put in a few patterns, and Les Solomon, not fully knowing the rules of the game and certainly not aware of the deep philosophical and mathematical implications, watched the little blue, red, or green stars (that was the way the Dazzler made the cells look) eat the other little stars, or make more stars. What a waste of time, he thought. Who cares?
Then he began idly playing with the machine, sketching out a pattern to run. He happened, in his inebriation, to put up something resembling the Star of David. He later recalled: “I ran the program and watched the way it ate itself up. It took about ten minutes and finally it died. I thought, ‘Gee that’s interesting—does that mean the Jewish religion is about to go after two hundred forty-seven generations?’ So I drew a crucifix. That went for one hundred twenty-one generations. Does this somehow mean that Judaism will outlast Christianity?” Soon he was putting up crescents and stars and symbols of different meanings, and the three of them—four of them, including the Altair—were exploring the mysteries of the world’s religions and nationalities. “Who the hell needs philosophy at three o’clock in the morning, drinking?” Solomon later said. "It was a computer. It was there.”
But Les Solomon had more magic to transmit. One of the stories he would tell, stories so outrageous that only a penny-pincher of the imagination would complain of their improbability, was of the time he was exploring in pursuit of one of his “hobbies,” pre-Colombian archeology. This required much time in jungles, “running around with Indians, digging, pitching around in the dirt . . . you know, finding things.” It was from those Indians, Les Solomon insisted, that he learned the vital principle of vril, a power that allows you to move huge objects with very little force. Solomon believed that it was the power of vril which enabled the Egyptians to build the pyramids. (Perhaps vril was the power that Ed Roberts was talking about when he realized that his Altair would give people the power of ten thousand pyramid-building Egyptians.) According to his story, Solomon met a venerable Indian brujo and asked if he might learn this power. Could the brujo teach him? And the brujo complied. Now, after the drunken evening with the LIFE program, Solomon attended a Homebrew meeting at SLAC where he was accorded the respect of an honored guest—the midwife of Ed Roberts’ Altair. And after the meeting, Solomon was telling the hardware hackers about vril. There was some skepticism.
Outside of SLAC were huge orange picnic tables with concrete bases. Solomon had the Homebrew people touch their hands on one of the tables, and he touched it, too. They simply had to think it would rise.
Lee Felsenstein later described the scene: “He’d said, ‘Hey, let me show you . . .’ We were hanging on his every word, we’d do anything. So about six people surrounded the table, put their hands on. He put his hand on top, squinted his eyes and said, ‘Let’s go.’ And the table raised about a foot. It rose like a harmonic motion, [as elegantly as] a sine wave. It didn’t feel heavy. It just happened.”
Afterward, even the participants, save Solomon, were not sure that it had really happened. But Lee Felsenstein, seeing another chapter close in that earth-shattering science-fiction novel that was his life, understood the mythic impact of this event. They, the soldiers of the Homebrew Computer Club, had taken their talents and applied the Hacker Ethic to work for the common good. It was the act of working together in unison, hands-on, without the doubts caused by holding back, which made extraordinary things occur. Even impossible things occur. The MIT hackers had discovered that when their desire to hack led them to persist so single-mindedly that the barriers of security, exhaustion, and mental limits seemed to shrink away. Now, in the movement to wipe away generations of centralized, antihacker control of the computer industry, to change the world’s disapproving view of computers and computer people, the combined energy of hardware hackers working together could do anything. If they did not hold back, not retreat within themselves, not yield to the force of greed, they could make the ideals of hackerism ripple through society as if a pearl were dropped in a silver basin.
Homebrew Club was sitting atop the power of vril.
Chapter 11. Tiny BASIC
While the hunger to build and expand the Altair was as insatiable in the hardware hackers of the seventies as the desire to hack PDP-1s and 6s was to the MIT hackers of the sixties, a conflict was developing around the Homebrew Computer Club which had the potential to slow the idealistic, bootstrapping process and stem the rising tide lifting them all. At the heart of the problem was one of the central tenets of the Hacker Ethic: the free flow of information, particularly information that helped fellow hackers understand, explore, and build systems. Previously, there had not been much of a problem in getting that information from others. The “mapping section” time at Homebrew was a good example of that—secrets that big institutional companies considered proprietary were often revealed. And by 1976 there were more publications plugging into what was becoming a national pipeline of hardware hackers—besides PCC and the Homebrew newsletter there was now Byte magazine in New Hampshire—you could always find interesting assembly-language programs, hardware hints, and technical gossip. New hacker-formed companies would give out schematics of their products at Homebrew, not worrying about whether competitors might see them; and after the meetings at The Oasis, the young, blue-jeaned officers of the different companies would freely discuss how many boards they shipped, and what new products they were considering.
Then came the outcry over Altair BASIC. It would give the hardware hackers a hint of the new fragility of the Hacker Ethic. And indicate that—as computer power did come to the people—other, less altruistic philosophies might prevail.
It all started out as a typical hacker caper. Among the products that MITS had announced, but not yet sent out to those who had ordered it, was a version of the BASIC computer language. Among the tools an Altair owner could have, this was to be one of the most highly coveted: because, once you had a BASIC on your Altair, the machine’s power to implement systems, to move mental pyramids, would improve “by orders of magnitude,” as the expression went. Instead of having to laboriously type in machine language programs onto paper tape and then have to retranslate the signals back (by then many Altair owners had installed I/O cards which would enable them to link the machines to teletypes and paper-tape readers), you would have a way to write quick, useful programs. While software hackers (and certainly such ancient assembly-language zealots as Gosper and Greenblatt) disdained BASIC as a fascist language, to hardware hackers trying to extend their systems it was an incredibly valuable tool.
The problem was that at first you couldn’t get a BASIC. It was particularly maddening because MITS supposedly had one, though no one at Homebrew had seen it run.
Indeed, MITS did have a BASIC. It had had the language running since early spring 1975. Not long before MITS began shipping Altairs to computer-starved Popular Electronics readers, Ed Roberts had gotten a phone call from two college students named Paul Allen and Bill Gates.
The two teenagers hailed from Seattle. Since high school the two of them had been hacking computers; large firms paid them to do lucrative contract programming. By the time Gates, a slim, blond genius who looked even younger than his tender years, had gone off to Harvard, the two had discovered there was some money to be made in making interpreters for computer languages like BASIC for new comp
uters.
The Altair article, while not impressing them technically, was exciting to them: it was clear microcomputers were the next big thing, and they could get involved in all the action by writing BASIC for this thing. They had a manual explaining the instruction set for the 8080 chip, and they had the Popular Electronics article with the Altair schematics, so they got to work writing something that would fit in 4K of memory. Actually, they had to write the interpreter in less than that amount of code, since the memory would not only be holding their program to interpret BASIC into machine language, but would need space for the program that the user would be writing. It was not easy, but Gates in particular was a master at bumming code, and with a lot of squeezing and some innovative use of the elaborate 8080 instruction set, they thought they’d done it. When they called Roberts, they did not mention they were placing the call from Bill Gates’ college dorm room. Roberts was cordial, but warned them that others were thinking of an Altair BASIC; they were welcome to try, though. “We’ll buy from the first guy who shows up with one,” Roberts told them.
Not long afterward, Paul Allen was on a plane to Albuquerque with a paper tape containing what he and his friend hoped would run BASIC on the machine. He found MITS a madhouse. “People would work all day long, rush home, eat their dinner and come back,” MITS executive Eddie Currie later recalled. “You could go in there any hour of the day or night and there would be twenty or thirty people, a third to half the staff [excluding manufacturing]. And this went on seven days a week. People were really caught up in this because they were giving computers to people who were so appreciative, and who wanted them so badly. It was a grand and glorious crusade.”
Only one machine at MITS then had 4K of memory, and that barely worked. When Paul Allen stuck the tape in the teletype reader and read the tape in, no one was sure what would happen. What happened was that the teletype it was connected to said, READY. Ready to program! “They got very excited,” Bill Gates later said. “Nobody had ever seen the machine do anything.”
The BASIC was far from a working version, but it was close enough to completion and its routines were sufficiently clever to impress Ed Roberts. He hired Allen and arranged to have Gates work from Harvard to help get the thing working. When, not long afterward, Gates finally took off from school (he would never return) to go to Albuquerque, he felt like Picasso stumbling upon a sea of blank canvases—here was a neat computer without utilities. “They had nothing!” he said later, awed years after the fact. “I mean, the place was not sophisticated, as far as software went. We rewrote the assembler, we rewrote the loader . . . we put together a software library. It was pretty trashy stuff, but people could have fun using the thing.”
The difference between the Gates-Allen software library and the software library in the drawer by the PDP-6 or the Homebrew Club library was that the former was for sale only. Neither Bill Gates nor Ed Roberts believed that software was any kind of sanctified material, meant to be passed around as if it were too holy to pay for. It represented work, just as hardware did, and Altair BASIC was listed in the MITS catalog like anything else it sold.
Meanwhile, the hunger at Homebrew for an Altair BASIC was getting unbearable. As it turned out, Homebrew members were perfectly capable of writing BASIC interpreters, and some of them would do just that. Others, though, had ordered Altair BASIC and were impatiently awaiting delivery, just as they had impatiently awaited delivery of other MITS products. Patience with MITS was getting thin, especially since the debacle with the dynamic memory boards which Ed Roberts insisted should work and never did. People who had been burned by buying memory boards began to snort and pout when they spoke of Ed Roberts’ company, especially since Roberts himself, who had attained legendary status as a reclusive genius who never left Albuquerque, was spoken of as a greedy, power-mad foe of the Hacker Ethic. It was even rumored that he wished ill on his competitors. The proper hacker response to competitors was to give them your business plan and technical information, so they might make better products and the world in general might improve. Not to act as Ed Roberts did at the First World Altair Convention, held at Albuquerque a year after the machines were introduced, when the strong-willed MITS president refused to rent display booths to competitors, and, according to some, raged with fury when he heard that companies like Bob Marsh’s Processor Technology had rented suites at the convention hotel and were entertaining prospective customers.
So when the MITS Caravan came to the Rickeys Hyatt House in Palo Alto in June of 1975, the stage was set for what some would call a crime and others would call liberation. The “Caravan” was a MITS marketing innovation. Some of the MITS engineers would travel in a motor home, dubbed the MITS-mobile, from city to city, setting up Altairs in motel seminar rooms and inviting people to see the amazing low-cost computers at work. The turnout would largely be people who ordered Altairs and had questions on when they could expect delivery. People who owned them would want to know where they went wrong in assembling the monster. People who owned MITS memory boards would want to know why they didn’t work. And people who’d ordered Altair BASIC would complain that they hadn’t gotten it.
The Homebrew Computer Club crowd was out in force when the Caravan met at the Rickeys Hyatt on El Camino Real in Palo Alto in early June, and were amazed when they found that the Altair on display was running BASIC. It was connected to a teletype which had a paper-tape reader, and once it was loaded anyone could type in commands and get responses instantly. It looked like a godsend to those hackers who had already sent in several hundred dollars to MITS and were impatiently waiting for BASIC. There is nothing more frustrating to a hacker than to see an extension to a system and not be able to keep hands-on. The thought of going home to an Altair without the capability of that machine running in the pseudo-plush confines of the Rickeys Hyatt must have been like a prison sentence to those hackers. But hands-on prevailed. Years later, Steve Dompier tactfully described what happened next: “Somebody, I don’t think anyone figured out who, borrowed one of their paper tapes lying on the floor.” The paper tape in question held the current version of Altair BASIC written by Bill Gates and Paul Allen.
Dan Sokol later recalled that vague “someone” coming up to him and, noting that Sokol worked for one of the semiconductor firms, asking if he had any way of duplicating paper tapes. Sokol said yes, there was a tape-copying machine available to him. He was handed the tape.
Sokol had all sorts of reasons for accepting the assignment to copy the tapes. He felt that MITS’ price for the BASIC was excessive. He thought that MITS was greedy. He had heard a rumor that Gates and Allen had written the interpreter on a big computer system belonging to an institution funded in part by the government, and therefore felt the program belonged to all taxpayers.
He knew that many people had paid MITS for the product already, and their getting an early copy wouldn’t hurt MITS financially. But most of all, it seemed right to copy it. Why should there be a barrier of ownership standing between a hacker and a tool to explore, improve, and build systems?
Armed with this philosophical rationale, Sokol took the tape to his employer’s, sat down at a PDP-11, and threaded in the tape. He ran it all night, churning out tapes, and at the next Homebrew Computer Club meeting he came with a box of tapes. Sokol charged what in hacker terms was the proper price for software: nothing. The only stipulation was that if you took a tape, you should make copies and come to the next meeting with two tapes. And give them away. People snapped up the tapes, and not only brought copies to the next meeting but sent them to other computer clubs as well. So that first version of Altair BASIC was in free-flowing circulation even before its official release.
There were two hackers, however, who were far from delighted at this demonstration of sharing and cooperation—Paul Allen and Bill Gates. They had sold their BASIC to MITS on a basis that earned them royalties for every copy sold, and the idea of the hacker community blithely churning out copies of their program and giving t
hem away did not seem particularly Utopian. It seemed like stealing. Bill Gates was also upset because the version that people were exchanging was loaded with bugs that he was in the process of fixing. At first he figured that people would just buy the debugged version. But even after MITS did release the debugged BASIC, it became clear that Altair users were not buying as many copies as they would if they hadn’t had a “pirated” BASIC already running. Apparently, they were either putting up with the bugs or, more likely, having a grand old hacker time debugging it themselves. Gates was becoming very upset, and when David Bunnell (who was then editing the newly begun Altair Users’ Newsletter for MITS) asked him what he wanted to do about it, Gates, then nineteen and imbued with a cockiness that comes from technical virtuosity and not necessarily social tact, said maybe he should write a letter. Bunnell promised him he could get the letter out to the troublemakers.
So Gates wrote his letter, and Bunnell not only printed it in the Altair newsletter but sent it to other publications, including the Homebrew Computer Club newsletter. Entitled an "Open Letter to Hobbyists,” it explained that while he and Allen had received lots of good feedback about the interpreter, most of the people praising it hadn’t bought it. The letter got to the heart of the matter quickly:
Why is this? As the majority of hobbyists must be aware, most of you steal your software. Hardware must be paid for, but software is something to share. Who cares if the people who worked on it get paid?
Gates went on to explain that this “theft” of software was holding back talented programmers from writing for machines like the Altair. “Who can afford to do professional work for nothing? What hobbyist can put three man-years into programming, finding all the bugs, documenting his product and distributing for free?”