CHAPTER 3
Lakeside was the most prestigious private school in Seattle, and I wanted nothing to do with it. My Ravenna friends were moving on to seventh grade at Eckstein Junior High, the nearby public school, and I’d assumed I’d be with them. Worse yet, Lakeside was all boys, a grim prospect for a twelve-year-old.
But when my parents heard that I’d spent most of sixth grade reading on my own in the back of the room, they decided that I needed more of a challenge. They would have to sacrifice to pay the Lakeside tuition—$1,335, a lot for a middle-class family in those days. But they wanted me to have opportunities they’d missed out on in Oklahoma.
“Why do I have to go to private school?” I kept asking.
“Because you’ll learn more,” my mother replied. “And there will be a lot of other smart kids there. It’ll be good for you.”
Lakeside’s entrance test was famously difficult. I decided to fail on purpose, and that would be that. It was a foolproof plan until I sat down with the exam: multiple choice, with lots of object rotations and pattern matching, a variation on a standard IQ test. This is kind of interesting, I thought. Let’s see how hard these questions are. I decided to solve the first set, just to see if I could, and then compensate at the end with a bunch of wrong answers.
The next thing I knew, time was called: “Pencils down!” It was one of those tests that no one finished completely, and I hadn’t gotten around to filling in those mistakes. But I was sure I wouldn’t be admitted, anyway, since the odds were so slim.
I got in. And my parents were right. It was really good for me.
MODELED AFTER A New England prep school, Lakeside was a collection of old brick buildings on thirty acres near the Jackson Park Golf Course in north Seattle. I was thrown into a forty-eight-member class of the city’s elite: the sons of bankers and businessmen, lawyers and UW professors. With scattered exceptions, they were preppy kids who knew each other from private grammar schools or the Seattle Tennis Club.
Just about everybody was smart at Lakeside, and they had skills and study habits that I lacked. The teachers were dynamic and demanding, prone to answering questions with questions. (The anomaly was Mr. Dunn, my volatile French teacher, who responded to careless conjugations with volleys of chalk and erasers.) For a while, I was tentative about raising my hand. I’d listen to the discussion and think my own thoughts, and then I’d chime in if nobody else did.
It took me most of seventh grade to get my bearings. Finally I clicked with Mr. Spock, my English teacher and the brother of Benjamin Spock, the world-famous pediatrician. “Paul has continued to be the most perceptive and thoughtful boy in my class,” he wrote in my spring report card. Gradually I got used to being challenged. I’d grow more intellectually in my six years at Lakeside than in any other phase of my life.
IN EIGHTH GRADE, two events stood out. For a pregame football rally, I rigged up an oil heater transformer under a chair that held an effigy in the opposing team’s colors. When the moment was right, the transformer set off a bunch of firecrackers stuffed in the dummy’s arms. It looked like an electrocution, just as I’d planned.
My second big moment came when I was chosen to deliver the graduation address for Lakeside’s lower school. It was my first speech, and I slaved over it. As I rose before classmates, faculty, parents, and honored guests, I felt a strange sensation in my legs. My knees were knocking, just like a cartoon.
It was 1967, and artificial intelligence was the hot theme in science fiction. I’d read Isaac Asimov’s I, Robot, with its First Law of Robotics (“A robot may not injure a human being or, through inaction, allow a human being to come to harm”), and Colossus, a 1966 British novel about a malevolent megacomputer that wound up ruling the world. Newspapers of the day were filled with headlines like “Computers Are Taking Over,” or “Automated Government Is Here.”
I began by hailing “the age of the computer” and a future that “holds for us the bright prospect of even more remarkable things to come.” After acknowledging the specter of computers someday replacing human workers on assembly lines, I paid my respects to the machines’ “amazing capabilities” in mathematics and their uses in banking, medicine, and the military. I pointed out that U.S. moon probes were in fact computer-run robots. But I was equally interested in what computers couldn’t do: “They cannot have an original idea. They are unable to go beyond the limitations of their programming. …”
Were we on the threshold of a thinking robot? I closed with a prediction: “In fifty years, a robot with a fairly large brain cell capacity will be within reach.” Today it appears that I was highly optimistic. With 2017 now around the corner, we’re still not close to matching the abilities of the incalculably complex human brain.
When I recently reread that speech, it brought back the image of a boy who was fascinated by computers but had little practical knowledge beyond the flip-flop circuit. All I knew came secondhand from things I’d read. When I was growing up, few people outside major universities or big corporations had ever seen a real computer. It would have been hard to imagine that I’d ever lay my hands on one.
* * *
WHILE LAKESIDE SEEMED conservative on the surface, it was educationally progressive. We had few rules and lots of opportunities, and all my schoolmates seemed passionate about something. But the school was also cliquish. There were golfers and tennis players, who carried their rackets wherever they went, and in the winter most everyone went skiing. I’d never done any of these things, and my friends were the boys who didn’t fit into the established groups. Then, in the fall of my tenth-grade year, my passion found me.
My honors geometry teacher was Bill Dougall, the head of Lakeside’s science and math departments. A navy pilot in World War II, Mr. Dougall had an advanced degree in aeronautical engineering and another in French literature from the Sorbonne. In our school’s best tradition, he believed that book study wasn’t enough without real-world experience. He also realized that we’d need to know something about computers when we got to college. A few high schools were beginning to train students on traditional mainframes, but Mr. Dougall wanted something more engaging for us. In 1968 he approached the Lakeside Mothers Club, which agreed to use the proceeds from its annual rummage sale to lease a teleprinter terminal for computer time-sharing, a brand-new business at the time.
On my way to math class in McAllister Hall, I stopped by for a look. As I approached the small room, the faint clacking got louder. I opened the door and found three boys squeezed inside. There was a bookcase and a worktable with piles of manuals, scraps from notebooks, and rolled-up fragments of yellow paper tape. The students were clustered around an overgrown electric typewriter, mounted on an aluminum-footed pedestal base: a Teletype Model ASR-33 (for Automatic Send and Receive). It was linked to a GE-635, a General Electric mainframe computer in a distant, unknown office.
One senior hunched over the machine and its khaki-colored keyboard, while another looked on and made an occasional cryptic comment. To the keyboard’s right was an embedded rotary dial, for the modem; to its left sat the punch, which spewed a continuous stream of inch-wide, eight-column paper tape. Each character was defined by the configuration of holes punched out among the eight channels. (An inch length of tape held ten characters; a small program might run two or three feet.) In front of the punch, a paper-tape reader translated your programs and sent them to the GE computer.
The Teletype made a terrific racket, a mix of low humming, the Gatling gun of the paper-tape punch, and the ka-chacko-whack of the printer keys. The room’s walls and ceiling had to be lined with white corkboard for soundproofing. But though it was noisy and slow, a dumb remote terminal with no display screen or lowercase letters, the ASR-33 was also state-of-the-art. I was transfixed. I sensed that you could do things with this machine.
That year would be a watershed in matters digital. In March 1968, Hewlett-Packard introduced the first programmable desktop calculator. In June, Robert Dennard won a patent for a one-transistor cell of dynamic random-access memory, or DRAM, a new and cheaper method of temporary data storage. In July, Robert Noyce and Gordon Moore cofounded Intel Corporation. In December, at the legendary “Mother of All Demos” in San Francisco, the Stanford Research Institute’s Douglas Engelbart showed off his original versions of a mouse, a word processor, e-mail, and hypertext. Of all the epochal changes in store over the next two decades, a remarkable number were seeded over those ten months: cheap and reliable memory; a graphical user interface; a “killer” application, and more. Had anyone connected the dots, they might have foreseen the transformation of computers and how they would soon be used.
THE CLASSIC MAINFRAMES of my youth were the size of tractor-trailers and wildly expensive. Those early IBMs and UNIVACs had no more computing power than today’s pocket calculators, but they took up entire rooms and threw off tremendous heat, even after transistors replaced vacuum tubes. They were overseen by trained operators who kept them running around the clock while the customers stayed outside, looking in. To gain access to computing, programmers used a keypunch machine to convert handwritten code into a deck of punch cards, one card per line. They’d snap a rubber band around the deck and bring it to an operator to have the cards read in.
Then the programmers returned to their offices to wait, because the work went on the operators’ schedule. Depending on their job’s priority, they’d pick up a printout hours or sometimes days later. If one card was bent or out of sequence, or a single comma in the wrong place, they’d get an error message and not much else. They’d have to deduce their mistake and start again.
“Batch processing,” as this system was called, worked fine for large-scale information management tasks, like corporate payrolls. But it became so frustrating for programmers that they mounted a guerrilla movement for greater interactivity. In 1957, the visionary John McCarthy demonstrated a radical software prototype: a “Compatible Time-sharing System,” as McCarthy called it, “that permits each user of a computer to behave as though he were in sole control.” Instead of passively waiting for punch cards to be processed, users communicated with the computer through their terminal keyboards. You could “talk” to a mainframe, receive a prompt reply, then make your corrections. Programming became more like a conversation.
Time-sharing made computer time affordable by spreading costs among hundreds of users. Dozens of people could engage one computer simultaneously, with the central processing unit shifting from one person’s work to the next in a fraction of a second. The new back-and-forth rhythm wasn’t merely more efficient. It was a leap that made card decks superfluous and computer users far more productive. In 1965, General Electric packaged a refined version of McCarthy’s system with the original Dartmouth BASIC and launched a commercial service. Three years after that, Bill Dougall and the Mothers Club brought it to Lakeside.
I was lucky to come of age in a time of fundamental change in the computer industry. Computing power, once the sole province of government and the wealthiest corporations and universities, could now be parceled out at an hourly rate. New technology delivered that power to scattered offices or schools. As usual, timing was crucial. If I’d been born five years earlier, I might have lacked the patience as a teenager to put up with batch-processing computers. Had I come around five years later, after time-sharing became institutionalized, I would have missed the opportunities that come from trying something new.
RATHER THAN MAKE programming a formal part of the math curriculum, Lakeside offered it as an independent study option. We were lightly supervised by Fred Wright, a young math teacher who’d taken a summer course in punch card programming at Stanford. Mr. Wright gave us a BASIC manual and a few starter problems to whet our appetites, and then he let us loose. Because we didn’t know the “correct” way of doing things, we devised our own techniques. We became resourceful of necessity.
Only the most cursory documentation was furnished to help us. The BASIC manual was fifty-odd pages long, and I consumed it in a day or two. I memorized the twenty or so main keywords and how certain keys functioned on the Teletype. The language felt foreign for the first hour or two, and then it was—Oh yeah, I get it. BASIC was a lot easier than French: consistently logical, no irregular verbs, compact vocabulary. When I got stumped, I’d ask one of the seniors for help: How do you make that work? How do you print that? They were a month or so ahead of me and happy to show off what they knew.
In one of my first programs, borrowed from a manual, I graphed a sine wave. I watched the teleprinter’s carriage swing back and forth to print a perfect pattern of asterisks, as though moved by an unseen, mesmerizing hand. Within days Fred Wright had little left to teach us. Now and then he’d pop his head in, smile, and say, “How are you guys doing?” Some of the stodgier teachers grumbled that we had too much freedom, but Mr. Wright loved riding that fine line between control and chaos, unleashing our enthusiasm.
It’s hard to convey my excitement when I sat down at the Teletype. With my program written out on notebook paper, I’d type it in on the keyboard with the paper tape punch turned on. Then I’d dial into the GE computer, wait for a beep, log on with the school’s password, and hit the start button to feed the paper tape through the reader, which took several minutes.
At last came the big moment. I’d type “RUN,” and soon my results printed out at ten characters per second—a glacial pace next to today’s laser printers, but exhilarating at the time. It would soon be apparent whether my program worked; if not, I’d get an error message. In either case, I’d quickly log off to save money. Then I’d fix any mistakes by advancing the paper tape to the error and correcting it on the keyboard while simultaneously punching a new tape—a delicate maneuver nowadays handled by a simple click of a mouse and a keystroke. When I achieved a working program, I’d secure it with a rubber band and stow it on a shelf until the next session.
For young people today, this process might seem hopelessly laborious, like cracking a walnut with a Rube Goldberg machine. But for high school students in the late 1960s, it was astounding to get “instant” feedback from a computer, even if you had to wait several seconds for the machine’s next move in a game of Yahtzee. In a sense, that time-sharing terminal marked my start in personal computing years before personal computers. Programming resonated with my drive to figure out whether things worked or not and then to fix them. I’d long marveled at the innards of things, from transistors and integrated circuits back to that young-reader’s book on road equipment. But crafting my own computer code felt more creative than anything I’d tried before. I sensed that there would always be more to learn, layer upon layer of knowledge and techniques.
Soon I was spending every lunchtime and free period around the Teletype with my fellow aficionados. Others might have found us eccentric, but I didn’t care. I had discovered my calling. I was a programmer.
TWENTY OR SO students dropped into the computer room from time to time, but only half a dozen made it the hub of their universe. Although programming at its heart is a solitary venture, we became a nascent brotherhood. With no teachers to guide us, we traded commands and tricks of the trade. While a few of the acolytes were older students like Robert McCaw and Harvey Motulsky, I was one of four younger ones who formed the core. Ric Weiland, the son of a Boeing engineer, resembled Spock in Star Trek without the pointy ears: quiet, kind, meticulous. Ric built his own tic-tac-toe relay computer in the ninth grade, but never sought attention; he was happier in the background. Kent Evans, a minister’s son two years younger than Ric and I, had frizzy hair, an intricate set of braces, and unflagging intensity. He was game for anything.
One day early that fall, I saw a gangly, freckle-faced eighth-grader edging his way into the crowd around the Teletype, all arms and legs and nervous energy. He had a scruffy-preppy look: pullover sweater, tan slacks, enormous saddle shoes. His blond hair went all over the place. You could tell three things about Bill Gates pretty quickly. He was really smart. He was really competitive; he wanted to show you how smart he was. And he was really, really persistent. After that first time, he kept coming back. Many times he and I would be the only ones there.
Bill came from a family that was prominent even by Lakeside standards; his father later served as president of the state bar association. I remember the first time I went to Bill’s big house a block or so above Lake Washington, feeling a little awed. His parents subscribed to Fortune and Bill read it religiously. One day he showed me the magazine’s special annual issue and asked me, “What do you think it’s like to run a Fortune 500 company?” I said I had no idea. And Bill said, “Maybe we’ll have our own company someday.” He was thirteen years old and already a budding entrepreneur.
Where I was curious to study everything in sight, Bill would focus on one task at a time with total discipline. You could see it when he programmed—he’d sit with a marker clenched in his mouth, tapping his feet and rocking, impervious to distraction. He had a unique way of typing, sort of a six-finger, sideways scrabble. There’s a famous photograph of Bill and me in the computer room not long after we first met. I’m seated in a hardback chair at the teleprinter in my dapper green corduroy jacket and turtleneck. Bill is standing to my side in a plaid shirt, his head cocked attentively, eyes trained on the printer as I typed. He looks even younger than he actually was. I look like an older brother, which was something Bill didn’t have.
LIKE ALL TEENAGE boys, we loved games. Harvey Motulsky created a text-based version of Monopoly, with the computer’s random number generator “rolling the dice.” Bob McCaw put together a virtual casino program (including craps, blackjack, and roulette) that involved three hundred lines of code. We proudly mounted the printout up one wall, across the ceiling, and down the other.
Within a month, we’d run through the Mothers Club’s budget for computer time for the year, so they allocated a little bit more. In early November, as computer blackjack began to pall, I got news from Harvey. A time-sharing company had opened in Seattle’s University District. It needed people for acceptance testing of its new-model leased computer, a Digital Equipment Corporation PDP-10.
The next night I asked my father to take me to the Computer Center Corporation, a ten-minute drive from our home. I peered through the plate glass, into a room that never went dark, at the mysterious puppy in the window: a black mainframe with cabinet after cabinet and panels of blinking lights. The CPU alone was about five feet wide. It was the first time that I’d seen an actual computer in the flesh, and it seemed not quite real that such a thing could exist just forty blocks from where I lived. All I wanted to do at that moment was log on, connect, and have at it.
Today’s average laptop is thirty thousand times faster than the machine I was lusting after, with ten thousand times more memory. But in its day, the PDP-10 was the most advanced species of an evolutionary alternative to the batch-processing establishment. Founded by Ken Olsen and Harlan Anderson, DEC made its first splash in 1960 with the PDP-1, the first truly interactive, “conversational” computer. Less than a decade later, the PDP-10 became the mainstay for the Defense Department’s ARPANET (the original Internet) and a time-sharing workhorse. It ran faster than GE’s system at Lakeside and had a broader software repertoire, including FORTRAN and other languages, plus a rich array of online utilities.
Fortunately for me and my fellow Lakesiders, this wonderful hardware all relied on a new operating system, TOPS-10, that was apt to crash whenever it served too many users at a time. Computer Center Corporation (which we’d call C-Cubed) had taken delivery of its leased PDP-10 in October 1968, with a plan to start selling time in the New Year. In the meantime, their TOPS-10 needed to be debugged before the paying customers arrived. As an added incentive for C-Cubed, its lease payments would be deferred until the software functioned reliably. The company needed somebody to push the system to its limit, which was where we came in.
One C-Cubed partner was a Lakeside mother who’d heard about our little tech fraternity. A few days after my sneak preview, Fred Wright ushered us into the building to make introductions. A resident guru laid out the deal: We could have unlimited free time on their terminals, off-hours, as long as we abided by their ground rules. “You can try to crash the computer,” he said, “but if it crashes from something you do, you’ve got to tell us what you did. And you can’t do it again until we tell you to try.”
The following Saturday, we met in the C-Cubed terminal room, a space three times the size of our cubbyhole at Lakeside. We were delighted to find a bank of a half-dozen ASR-33s: no more waiting to get on. Through another door lay the sanctum sanctorum, the computer room. Manned seven days a week by three shifts of operators, it was big and square and fluorescent-bright, with a shiny raised floor to keep the fat power and data cables out of harm’s way. Whenever a bulky disk drive was installed, industrial-size suction cups were used to lift the floor and run new cables. Between the air conditioning and the hulking computer’s fans, the place was so noisy that some operators wore hearing protectors, like workers on a factory floor.
For us, shifting from the GE-635 to the PDP-10 was like trading in a Corolla for a Ferrari. Saturdays were not nearly enough. We’d bus down to C-Cubed after school, cutting gym class to get there earlier, our junior briefcases in hand. (I doted on mine, which was brown leather and popped open at the lightest touch of my thumbs.) We were on the road to becoming hackers, in the original, nonfelonious sense of the term: fanatical programmers who stretched themselves to the limit. As author Steven Levy has noted, hacker culture was a meritocracy. Your status didn’t hinge on your age or what your father did for a living. All that counted was ingenuity and your hunger to learn more about coding.
Every neophyte needs a master, and C-Cubed had three of them. They were world-class programmers all, with a nerdish élan and a tinge of the exotic. Unlike the business-side executives, they didn’t treat us like nuisances; I suspect they may have seen in us their younger selves. At times it felt as though I’d jumped from high school into a postgraduate seminar in advanced systems programming.
Steve “Slug” Russell, the company’s hardware chief, was short and round, with a wry sense of humor. Then thirty-one, he’d followed John McCarthy from Dartmouth to MIT. There Russell had created Spacewar, the first truly interactive computer game, on a PDP-1.
Bill Weiher, slim and bespectacled, never said much. Known for developing SOS (an acronym for Son of STOPGAP, one of the first great text editors), he looked like a scribe from the Middle Ages. I’d see him crunching away tirelessly at his terminal, building elaborate structures of intricate code.
Dick Gruen, an ex-DEC consultant who’d met Russell and Weiher at Stanford, was the most gregarious of the lot, a junk-food addict and Falstaffian jokester with a mop of curly hair. According to Gruen, the operating system had yet to be born that he could not crash, and he was clever enough that I believed him.
To them we were “the Lakeside kids” or “the testers.” On occasion they’d have us simultaneously run a bunch of copies of a chess program to place an extra-heavy load on the system. Our assignment played to a teenager’s impulse to wreck things just for fun, while channeling it into something positive. As I later told a Seattle journalist, “The most effective way to learn was going hands-on with what was the top machine of the time, learning about how it worked, what it took to ‘make it or break it.’ ”
Another approach was to stress-test a piece of software until it failed, when we’d scribble down what happened on a piece of paper and move on. The ultimate coup was to crash the whole operating system, which would be apparent when the teleprinter froze and buzzed as you tried to type. Later Russell and Gruen would determine the source of the snag, happy as clams, knowing that their lease payment to DEC had been once again forestalled. We were happy, too. As long as we kept finding bugs, we’d extend our Camelot of free time.
When one of our mentors came by, I was almost too intimidated to speak. We adopted their jargon; a kludge, for example, was a baling-wire-and-gum sort of coding fix. They put up with our pestering, and every now and then threw us a bone from something they’d been working on. We were in awe of how efficiently they coded, a critical skill in an era of limited computer memory.
Mostly we were free to bang away on our own small projects. Bill worked on a war game; Ric grappled with FORTRAN. I wrote code for a matchmaking program. In the evening, we usually had the teleprinter room to ourselves. When we needed to pick up our listings, we’d knock on the computer-room door, say hello to the night operator, collect our printout, and return to our Teletypes. We might steal a glance at the PDP-10, but that was as close as we got.
THE KEY TO commercial time-sharing was permanent, high-speed data storage, a way to gain easy access to your work. C-Cubed limped along for months with old-generation disk drives that limited most customers to a couple dozen files of modest length. So there was great anticipation when Russell took delivery of a box about eight feet long by four feet high: a new moving-head disk drive from Bryant Computer Products in Walled Lake, Michigan. One of the company’s field service representatives, a thick-accented Southerner, called it the Giant Bryant. The name stuck.
The drive was built to heroic dimensions. A massive electric motor at the center ran a thick shaft that supported a dozen oxide-coated steel disks, each more than three feet in diameter. They spun in unison while a set of hydraulic arms with magnetic heads, floating on thin cushions of air, moved across their surfaces to read the data. The drive could store around 100 million characters, an order of magnitude beyond anything else available. (Today’s laptop drives typically store six hundred times as much data in 0.002 percent of the volume.)
Unfortunately, the Giant Bryant was flaky to a fault. Every so often, with as little provocation as a nearby footstep, a head would touch a disk and strip off the oxide: the ominous head crash, with data irretrievably lost and the disk damaged beyond repair.
For archival storage, C-Cubed used a less imposing device called DECtape. It came in four-inch canisters—small enough to slip into a pocket, large enough to hold a million characters. One 260-foot roll had the capacity of 2,500 feet of paper tape, or slightly more than the eight-inch floppy disks to be introduced by IBM five years later. Even within the limits of its motorized, reel-to-reel drive, DECtape was faster than paper tape and much sturdier, with dual redundancy and two layers of Mylar protecting the oxide. In demonstrations, DEC salesmen would punch a quarter-inch hole in the tape and then show that its data was intact.
Best of all, DECtape featured a directory structure, just like the Giant Bryant or the floppy drives to come. Traditional magnetic tapes were like sequential streams where stored information couldn’t be safely updated; if you wrote something new in the middle of a tape, subsequent data would be lost. But DECtape was organized in discrete blocks of data, and one block could be rewritten without affecting any other. Now we could store half a dozen or more programs on a single tape, find all of them by name, and edit them independently or write over them. Up until I bought my own home terminal, my DECtapes were the first piece of computing technology that really belonged to me. We all wanted more of them—they were status symbols. Those little canisters made my work feel less ephemeral, more substantive, as though it had real and lasting value.