by Steven Levy
Saunders, being an upperclassman, had come to the TX-0 later in his college career than Kotok and Samson: he had used the breathing space to actually lay the foundation for a social life, which included courtship of and eventual marriage to Marge French, who had done some nonhacking computer work for a research project. Still, the TX-0 was the center of his college career, and he shared the common hacker experience of seeing his grades suffer from missed classes. It didn’t bother him much, because he knew that his real education was occurring in Room 240 of Building 26, behind the Tixo console. Years later he would describe himself and the others as “an elite group. Other people were off studying, spending their days up on four-floor buildings making obnoxious vapors or off in the physics lab throwing particles at things or whatever it is they do. And we were simply not paying attention to what other folks were doing because we had no interest in it. They were studying what they were studying and we were studying what we were studying. And the fact that much of it was not on the officially approved curriculum was by and large immaterial.”
The hackers came out at night. It was the only way to take full advantage of the crucial “off-hours” of the TX-0. During the day, Saunders would usually manage to make an appearance in a class or two. Then some time spent performing “basic maintenance” things like eating and going to the bathroom. He might see Marge for a while. But eventually he would filter over to Building 26. He would go over some of the programs of the night before, printed on the nine-and-a-half-inch-wide paper that the Flexowriter used. He would annotate and modify the listing to update the code to whatever he considered the next stage of operation. Maybe then he would move over to the Model Railroad Club, and he’d swap his program with someone, checking simultaneously for good ideas and potential bugs. Then back to Building 26, to the Kluge Room next to the TX-0, to find an offline Flexowriter on which to update his code. All the while, he’d be checking to see if someone had canceled a one-hour session on the machine; his own session was scheduled at something like two or three in the morning. He’d wait in the Kluge Room, or play some bridge back at the Railroad Club, until the time came.
Sitting at the console, facing the metal racks that held the computer’s transistors, each transistor representing a location that either held or did not hold a bit of memory, Saunders would set up the Flexowriter, which would greet him with the word “WALRUS.” This was something Samson had hacked, in honor of Lewis Carroll’s poem with the line “The time has come, the Walrus said . . .” Saunders might chuckle at that as he went into the drawer for the paper tape that held the assembler program and fed that into the tape reader. Now the computer would be ready to assemble his program, so he’d take the Flexowriter tape he’d been working on and send that into the computer. He’d watch the lights go on as the computer switched his code from “source” (the symbolic assembly language) to “object” code (binary), which the computer would punch out into another paper tape. Since that tape was in the object code that the TX-0 understood, he’d feed it in, hoping that the program would run magnificently.
There would most probably be a few fellow hackers kibitzing behind him, laughing and joking and drinking Cokes and eating some junk food they’d extracted from the machine downstairs. Saunders preferred the lemon jelly wedges that the others called “lemon gunkies.” But at four in the morning, anything tasted good. They would all watch as the program began to run, the lights going on, the whine from the speaker humming in high or low register depending on what was in Bit 14 in the accumulator, and the first thing he’d see on the CRT display after the program had been assembled and run was that the program had crashed. So he’d reach into the drawer for the tape with the FLIT debugger and feed that into the computer. The computer would then be a debugging machine, and he’d send the program back in. Now he could start trying to find out where things had gone wrong, and maybe if he was lucky he’d find out and change things by putting in some commands by flicking some of the switches on the console in precise order, or hammering in some code on the Flexowriter. Once things got running—and it was always incredibly satisfying when something worked, when he’d made that roomful of transistors and wires and metal and electricity all meld together to create a precise output that he’d devised—he’d try to add the next advance to it. When the hour was over—someone already itching to get on the machine after him—Saunders would be ready to spend the next few hours figuring out what the heck had made the program go belly-up.
The peak hour itself was tremendously intense, but during the hours before, and even during the hours afterward, a hacker attained a state of pure concentration. When you programmed a computer, you had to be aware of where all the thousands of bits of information were going from one instruction to the next, and be able to predict—and exploit—the effect of all that movement. When you had all that information glued to your cerebral being, it was almost as if your own mind had merged into the environment or the computer. Sometimes it took hours to build up to the point where your thoughts could contain that total picture, and when you did get to that point, it was such a shame to waste it that you tried to sustain it by marathon bursts, alternately working on the computer or poring over the code that you wrote on one of the offline Flexowriters in the Kluge Room. You would sustain that concentration by “wrapping around” to the next day.
Inevitably, that frame of mind spilled over to what random shards of existence the hackers had outside of computing. The knife-and-paintbrush contingent at TMRC was not pleased at all by the infiltration of Tixo-mania into the club: they saw it as a sort of Trojan horse for a switch in the club focus, from railroading to computing. And if you attended one of the club meetings held every Tuesday at 5:15 P.M., you could see the concern: the hackers would exploit every possible thread of parliamentary procedure to create a meeting as convoluted as the programs they were hacking on the TX-0. Motions were made to make motions to make motions, and objections ruled out of order as if they were so many computer errors. A note in the minutes of the meeting on November 24, 1959, suggests that “we frown on certain members who would do the club a lot more good by doing more S&P-ing and less reading Robert’s Rules of Order.” Samson was one of the worst offenders, and at one point an exasperated TMRC member made a motion “to purchase a cork for Samson’s oral diarrhea.”
Hacking parliamentary procedure was one thing, but the logical mind-frame required for programming spilled over into more commonplace activities. You could ask a hacker a question and sense his mental accumulator processing bits until he came up with a precise answer to the question you asked. Marge Saunders would drive to the Safeway every Saturday morning in the Volkswagen and upon her return ask her husband, “Would you like to help me bring in the groceries?” Bob Saunders would reply, “No.” Stunned, Marge would drag in the groceries herself. After the same thing occurred a few times, she exploded, hurling curses at him and demanding to know why he said no to her question.
“That’s a stupid question to ask,” he said. “Of course I won’t like to help you bring in the groceries. If you ask me if I’ll help you bring them in, that’s another matter.”
It was as if Marge had submitted a program into the TX-0, and the program, as programs do when the syntax is improper, had crashed. It was not until she debugged her question that Bob Saunders would allow it to run successfully on his own mental computer.
Chapter 2. The Hacker Ethic
Something new was coalescing around the TX-0: a new way of life with a philosophy, an ethic, and a dream.
There was no one moment when it started to dawn on the TX-0 hackers that by devoting their technical abilities to computing with a devotion rarely seen outside of monasteries, they were the vanguard of a daring symbiosis between man and machine. With a fervor like that of young hot-rodders fixated on souping up engines, they came to take their almost unique surroundings for granted. Even as the elements of a culture were forming, as legends began to accrue, as their mastery of programming started to sur
pass any previous recorded levels of skill, the dozen or so hackers were reluctant to acknowledge that their tiny society, on intimate terms with the TX-0, had been slowly and implicitly piecing together a body of concepts, beliefs, and mores.
The precepts of this revolutionary Hacker Ethic were not so much debated and discussed as silently agreed upon. No manifestos were issued. No missionaries tried to gather converts. The computer did the converting, and those who seemed to follow the Hacker Ethic most faithfully were people like Samson, Saunders, and Kotok, whose lives before MIT seemed to be mere preludes to that moment when they fulfilled themselves behind the console of the TX-0. Later there would come hackers who took the implicit Ethic even more seriously than the TX-0 hackers did, hackers like the legendary Greenblatt or Gosper, though it would be some years yet before the tenets of hackerism would be explicitly delineated.
Still, even in the days of the TX-0, the planks of the platform were in place. The Hacker Ethic:
Access to computers—and anything that might teach you something about the way the world works—should be unlimited and total. Always yield to the Hands-On Imperative!
Hackers believe that essential lessons can be learned about the systems—about the world—from taking things apart, seeing how they work, and using this knowledge to create new and even more interesting things. They resent any person, physical barrier, or law that tries to keep them from doing this.
This is especially true when a hacker wants to fix something that (from his point of view) is broken or needs improvement. Imperfect systems infuriate hackers, whose primal instinct is to debug them. This is one reason why hackers generally hate driving cars—the system of randomly programmed red lights and oddly laid out one-way streets cause delays that are so goddamned unnecessary that the impulse is to rearrange signs, open up traffic-light control boxes . . . redesign the entire system.
In a perfect hacker world, anyone pissed off enough to open up a control box near a traffic light and take it apart to make it work better should be perfectly welcome to make the attempt. Rules that prevent you from taking matters like that into your own hands are too ridiculous to even consider abiding by. This attitude helped the Model Railroad Club start, on an extremely informal basis, something called the Midnight Requisitioning Committee. When TMRC needed a set of diodes or some extra relays to build some new feature into The System, a few S&P people would wait until dark and find their way into the places where those things were kept. None of the hackers, who were as a rule scrupulously honest in other matters, seemed to equate this with “stealing.” A willful blindness.
All information should be free.
If you don’t have access to the information you need to improve things, how can you fix them? A free exchange of information, particularly when the information was in the form of a computer program, allowed for greater overall creativity. When you were working on a machine like the TX-0, which came with almost no software, everyone would furiously write systems programs to make programming easier—Tools to Make Tools, kept in the drawer by the console for easy access by anyone using the machine. This prevented the dreaded, time-wasting ritual of reinventing the wheel: instead of everybody writing his own version of the same program, the best version would be available to everyone, and everyone would be free to delve into the code and improve on that. A world studded with feature-full programs, bummed to the minimum, debugged to perfection.
The belief, sometimes taken unconditionally, that information should be free was a direct tribute to the way a splendid computer, or computer program, works—the binary bits moving in the most straightforward, logical path necessary to do their complex job. What was a computer but something that benefited from a free flow of information? If, say, the accumulator found itself unable to get information from the input/output (I/O) devices like the tape reader or the switches, the whole system would collapse. In the hacker viewpoint, any system could benefit from that easy flow of information.
Mistrust Authority—Promote Decentralization.
The best way to promote this free exchange of information is to have an open system, something that presents no boundaries between a hacker and a piece of information or an item of equipment that he needs in his quest for knowledge, improvement, and time online. The last thing you need is a bureaucracy. Bureaucracies, whether corporate, government, or university, are flawed systems, dangerous in that they cannot accommodate the exploratory impulse of true hackers. Bureaucrats hide behind arbitrary rules (as opposed to the logical algorithms by which machines and computer programs operate): they invoke those rules to consolidate power, and perceive the constructive impulse of hackers as a threat.
The epitome of the bureaucratic world was to be found at a very large company called International Business Machines—IBM. The reason its computers were batch-processed Hulking Giants was only partially because of vacuum tube technology. The real reason was that IBM was a clumsy, hulking company that did not understand the hacking impulse. If IBM had its way (so the TMRC hackers thought), the world would be batch processed, laid out on those annoying little punch cards, and only the most privileged of priests would be permitted to actually interact with the computer.
All you had to do was look at someone in the IBM world and note the button-down white shirt, the neatly pinned black tie, the hair carefully held in place, and the tray of punch cards in hand. You could wander into the Computation Center, where the 704, the 709, and later the 7090 were stored—the best IBM had to offer—and see the stifling orderliness, down to the roped-off areas beyond which unauthorized people could not venture. And you could compare that to the extremely informal atmosphere around the TX-0, where grungy clothes were the norm and almost anyone could wander in.
Now, IBM had done and would continue to do many things to advance computing. By its sheer size and mighty influence, it had made computers a permanent part of life in America. To many people, the words “IBM” and “computer” were virtually synonymous. IBM’s machines were reliable workhorses, worthy of the trust that businessmen and scientists invested in them. This was due in part to IBM’s conservative approach: it would not make the most technologically advanced machines, but would rely on proven concepts and careful, aggressive marketing. As IBM’s dominance of the computer field was established, the company became an empire unto itself, secretive and smug.
What really drove the hackers crazy was the attitude of the IBM priests and sub-priests, who seemed to think that IBM had the only “real” computers, and the rest were all trash. You couldn’t talk to those people—they were beyond convincing. They were batch-processed people, and it showed not only in their preference of machines, but in their ideas about the way a computation center, and a world, should be run. Those people could never understand the obvious superiority of a decentralized system, with no one giving orders—a system where people could follow their interests, and if along the way they discovered a flaw in the system, they could embark on ambitious surgery. No need to get a requisition form. Just a need to get something done.
This antibureaucratic bent coincided neatly with the personalities of many of the hackers, who since childhood had grown accustomed to building science projects while the rest of their classmates were banging their heads together and learning social skills on the field of sport. These young adults who were once outcasts found the computer a fantastic equalizer, experiencing a feeling, according to Peter Samson, “like you opened the door and walked through this grand new universe . . .” Once they passed through that door and sat behind the console of a million-dollar computer, hackers had power. So it was natural to distrust any force that might try to limit the extent of that power.
Hackers should be judged by their hacking, not bogus criteria such as degrees, age, race, or position.
The ready acceptance of twelve-year-old Peter Deutsch in the TX-0 community (though not by nonhacker graduate students) was a good example. Likewise, people who trotted in with seemingly impressive credentials wer
e not taken seriously until they proved themselves at the console of a computer. This meritocratic trait was not necessarily rooted in the inherent goodness of hacker hearts—it was mainly that hackers cared less about someone’s superficial characteristics than they did about his potential to advance the general state of hacking, to create new programs to admire, to talk about that new feature in the system.
You can create art and beauty on a computer.
Samson’s music program was an example. But to hackers, the art of the program did not reside in the pleasing sounds emanating from the online speaker. The code of the program held a beauty of its own. (Samson, though, was particularly obscure in refusing to add comments to his source code explaining what he was doing at a given time. One well-distributed program Samson wrote went on for hundreds of assembly-language instructions, with only one comment beside an instruction that contained the number 1750. The comment was RIPJSB, and people racked their brains about its meaning until someone figured out that 1750 was the year Bach died, and that Samson had written an abbreviation for Rest In Peace Johann Sebastian Bach.)
A certain esthetic of programming style had emerged. Because of the limited memory space of the TX-0 (a handicap that extended to all computers of that era), hackers came to deeply appreciate innovative techniques that allowed programs to do complicated tasks with very few instructions. The shorter a program was, the more space you had left for other programs, and the faster a program ran. Sometimes when you didn’t need much speed or space, and you weren’t thinking about art and beauty, you’d hack together an ugly program, attacking the problem with “brute force” methods. “Well, we can do this by adding twenty numbers,” Samson might say to himself, “and it’s quicker to write instructions to do that than to think out a loop in the beginning and the end to do the same job in seven or eight instructions.” But the latter program might be admired by fellow hackers, and some programs were bummed to the fewest lines so artfully that the author’s peers would look at it and almost melt with awe.