Dealers of Lightning

by Michael Hiltzik

  While Mott was back in Boston studying the human factors, Tesler worked on the visual representation of the interface. His ambition was to build it around icons and menus—thumbnail-sized illustrations that would perform discrete functions when clicked with the mouse, and lists of commands that could be executed at any given time. For a while he almost started sympathizing with CSLs view of how heavy graphics bur­dened the underpowered Altos: His first graphical interfaces worked so slowly that to demonstrate his scheme he had to record it on videotape at one-ninth normal speed so it would appear natural when played back in real time.

  In 1975, after a year of work, Gypsy was ready for launch. Mott brought a couple of Altos and a high-speed Dover laser printer back to Ginn and wired them to a phototypesetter that would output camera-ready text. For the first time on such a large scale, professional editors manipulated text on a screen and stored it on magnetic disks rather than cutting, past­ing, and marking a typed manuscript with progressively illegible changes.

  "Initially the reaction to the concept was, 'You're going to have to drag me kicking and screaming,'" Mott recalled. "But everyone who sat in front of that system and used it, to a person, was a convert within an hour."

  In every way possible, Gypsy mimicked Ginn's customary routines. The system retained multiple versions and drafts of every file and displayed them as a list. An editor could use the mouse to scroll down the list and click on the desired version to open it. (This was the first time the mouse was used as it is today, to execute point-and-click operations; Engelbart's system and Bravo both used it simply to position the cursor within a block of text.)

  Mott s diligence in drawing the Ginn editors into the design phase paid off. Instead of an editing process "so laborious that there was a point at which you threw up your hands and said, 'I just don't want to do this any­more,'" he recalled, the Ginn staff "found the ability to edit on the screen and always have a clean copy improved the quality of the editing itself. They could do a lot more of it before it became frustrating."

  Within PARC, Bravo and Gypsy decisively tipped the balance in favor of the Alto over POLOS. Simonyi, Tesler, and Mott had shown that the Alto could support an interactive office system that worked fast enough to enhance—call it "augment"—the professional office worker's intelli­gence. Since POLOS was slipping even further behind schedule, the suc­cess of the Ginn experiment sealed its doom. "The only real question," remarked Ted Kaehler, one of Kay's engineers, "was whether POLOS would be obsolete before it was even operational." In the end, it was.

  CHAPTER 15

  On the Lunatic Fringe

  F

  or all their coolness as killer apps, Bravo and Gypsy only scratched the surface of the Alto's vast capabilities. Although it was not the first machine small enough to be used by an individual—the LINC had been there before—the Alto was the first one deliberately designed as a general-purpose "personal" appliance: individualistic and infinitely customizable. The computer was no longer a machine to which man had to adapt, but one endlessly adaptable to every users needs.

  The Alto's mystique worked potently on its new owners, who anthro­pomorphized their machines like drivers of Volkswagen Beetles, painting them in bright colors and christening them with considerable ceremony. Kay named his first so-called interim Dynabook "Bilbo," after J. R. R. Tolkien's heroic Hobbit. John Ellenby, a Briton who was placed in charge of readying the Alto design for large-scale production, called his the "Gzunda" ("because it 'gzunda' the desk"). The graphics researcher Dick Shoup connected his to a color video terminal, creating the first color computer monitor, and Taylor got his rigged to beep out the opening bars of "The Eyes of Texas Are Upon You" whenever he received an e-mail message.

  Linked by Ethernet to each other, to printers, and to a host of other devices such as video displays and organ keyboards, the Altos lit PARC's creative fuse. Thacker had designed the first custom application, a pro­gram called "SIL" (for "Simple Illustrator") that automated the process of laying out computer circuits and allowed schematics to be translated directly into printed boards. But scores more were right behind.

  There were "Draw" by Patrick Beaudelaire and "Markup" by William Newman, which picked up where Ivan Sutherland's "Sketchpad" left off by giving users the power to place freehand drawings directly onto the bitmapped screen. (Of course, where Sutherland's program worked only on the lone TX-2 at MIT, these would run on any Alto.) There were pro­grams to compose music and animation, to format text documents, and to assist the writing of more programs, and dozens of programs in Smalltalk.

  Meanwhile the all-pervading Ethernet encouraged the spread of e-mail and consequently the development of "Laurel," a program to simplify composing, reading, and filing e-mail messages. (A later ver­sion was dubbed "Hardy.") This was a huge leap toward Taylor's grail of the computer as communications device. "Computers didn't com­municate with each other then," Thacker recalled. "Except at PARC."

  And there they communicated almost nonstop, a digital chatter that would be the envy of today's Internet junkies. On the Alto network Xerox employees started the first on-line clubs, played the first networked com­puter games, even completed the first joint research projects widiout ever meeting their partners face to face. "At PARC I received my first electronic junk mail, my first electronic job acceptance, and first elec­tronic obituary," recalled one lab supervisor. Warren Teitelman once returned from a week out of town to find his electronic mailbox crammed with 600 messages.

  All this activity plainly pointed to a future in which the computer's functions would no longer be dictated by hardware, but software: stan­dardized, commercialized, and boundlessly adaptable. Butler Lampson was one of the first to divine the possibilities. In 1972, even before the Alto took form and more than ten years before the advent of the IBM Personal Computer and the software and hardware industries it spawned, he set forth in a professional publication his "rather phantas­magoric" vision of a Utopian computer-enhanced destiny:

  Millions of people will write non-trivial programs, and hundreds of thousands will try to sell them. Of course, the market will be much larger and very much more diverse than it is now, just as paper is more widespread and is used in many more ways than are adding machines. Almost everyone who uses a pencil will use a computer, and although most people will not do any serious programming, almost everyone will be a potential customer for serious programs of some kind. . . . Such a mass market will require mass distribution. Analogues of bookstores, newsstands and magazine subscriptions seem plausible, as well as the kind of mail-order and home improve­ment marketing patterns we have now.

  Raised on time-sharing, most of PARC's computer scientists had trou­ble getting acclimated to the experience of having computer cycles at their personal disposal. A few even felt a twinge of guilt the first time they turned their backs on an idle Alto, as though they were leaving food on the table while others starved. But for the first time in the history of com­puting, resources were abundant enough to waste.

  Few people at PARC were as devoted to the machine as Alan Kay. It was as diough this was what he had been waiting for all his fife—indeed, it had virtually been built to his personal specifications. And when it appeared, he was fully prepared to take advantage of it.

  Among the Tom Sawyers Kay was known for wielding the most beguil­ing paintbrush in the building. No one evangelized more convincingly on behalf of ideas he found compelling, whether they were his own or belonged to others. Kay proselytized out of necessity. The experience of emerging from grad school with a four-hundred-page thesis describing a machine that could not be physically realized had sent him into a psy­chological tailspin. An old tendency toward depression, spurred by his inability to execute, reasserted itself.

  "Right about that first year at PARC, under psychotherapy, I discov­ered I was confusing my talent with my temperament," he said. "I didn't have the temperament of a programmer. I realized I needed a group."

  This epiphany resembled
that of a poet suddenly finding his voice. Like all the self-educated, having once grasped an idea Kay was impatient to move onto the next. He was a man of bifurcated nature, simultaneously a peerless formulator of theory and an instinctive craftsman with a short attention span. Having spent decades as an intellectual lone wolf, Kay redirected his gift for communicating enthusiasm toward the goal of attracting followers, often at the university lectures for which he was much in demand. It helped that he could size up a potential disciple in a split second, and that technical aptitude was not a prerequisite.

  "After I would give a talk there would always be a fair number of people who would come up at the end with special stars in their eyes. At that stage nobody really knew how to do this stuff anyway, so I tended to hire people who could buy into the romance of the whole thing, because you could go a really good distance on romance."

  Inside the building it was impossible to pass within a few yards of Kay's door without sensing a gravitational tug. Perhaps his most important recruit was swept into his orbit that way, never to leave. Dan Ingalls had come to PARC on a temporary contract to help George White set up the SDS Sigma 3 he had acquired for his work in speech recognition.

  "My office ended up across the hall from Alans," Ingalls said. "I kept noticing that I was more interested in what I was hearing across the hall than in the speech work I was hired to do. These conversations I was eavesdropping on were all about open-ended computer science stuff, which I was interested in. One day I walked over and said, 'Hey, what are you up to?' And that led to his talking about his whole picture of personal computing and how one might make a simple job of a lot of the impor­tant things through some new language."

  Alan Kay on a tear through Ideaspace was a very formidable force. Ingalls hastened to finish his job for White, then cadged himself a full­time appointment across the hall. Kay had not been daydreaming when he told Ingalls about his plans for a new computer language. What he had in mind would become perhaps the first project in his life he would see through to fruition. Dan Ingalls, it turned out, was the person he needed to make it happen.

  By mid-1973 Kay's so-called "Learning Research Group" numbered eight. They were so miscellaneous in their skills and credentials that Bob Taylor took to calling them, not entirely facetiously, "the lunatic fringe."

  Ingalls brought along his friend Ted Kaehler, who had also come to help George White with his speech recognition project but found Idea- space more interesting. Diana Merry's route was even more random. A transplanted Iowan with little programming training, she got a job as sec­retary to Gerald Lucovsky and John Urbach, managers in the General Sciences Lab.

  From the first, however, Merry was entranced by Taylor's work and the other mysterious goings-on "down the hall among the computer folks" where she spent most of her free hours, as she recalled later. Eventually Taylor managed to get her transferred to a job as Jerry Elkind’s assistant.

  One of the office machines there was an elaborate electric typewriter that could do minimal text formatting through the application of a com­plicated sequence of keystrokes. No one ever utilized the beastly device's capability except Merry, who was caught at it one day by Alan Kay.

  "You're a programmer!" he exclaimed.

  "No kidding," she said.

  Impressed by the natural skills of this secretary smuggled over from the physics lab, Kay gave her a few hours of rudimentary training. After that it was a relatively simple matter to get her assigned to his group.

  Then there was Adele Goldberg, an educational technology specialist from the universities of Michigan and Chicago with fiery red hair and a turbocharged thought process ("Adele we described as speaking at nine- tenths of a Lampson," Merry recalled). Kay filled out the team with, among others, people like Tesler, whom he shared on a roughly fifty-fifty basis with Bill English until after Gypsy was finished, when he joined LRG full-time; Chris Jeffers, the childhood friend and "chief of staff"; and interns and summer students who strayed into his orbit with that tell­tale shimmer in their eyes.

  Kay reveled in his people's eclectic backgrounds, which did not always include work toward a doctorate. "A doctoral thesis is anything you can get three faculty members to sign," he would say in their defense (quot­ing Ivan Sutherland, who had been a signer of his). Or: "Point of view is worth eighty IQ points." He did not care if his recruits had doctorates— although he certainly employed a high percentage of academically gifted scientists and engineers—but he monitored their points of view meticu­lously.

  Like Taylor, he believed strongly that a lab's success depended on a shared vision. But he was determined to avoid Taylors tendency toward militaristic discipline. The Learning Research Groups dogma sprung from Alan Kay’s mind as surely as CSL's did from Taylors or Lampson's, but the lab's ambiance was less like an Army barracks than a bohemian party where the guests all happened to concur in their host's choice of wine.

  No corner of PARC generated anything like the Kay group's free­wheeling mania. "It was an amazingly seductive environment," recalled Merry. "I was there late at night all the time. People were so full of ideas and excitement, and of course everybody knew more than anybody else about how the world was supposed to be."

  Socially the Learning Research Group was also PARC's most cohesive unit. "There was just a wonderful, personal feel to the group that spoke really of caring about one another and supporting one another," Merry said. Kay strived to build collegiality by sponsoring annual team "offsites" at a favorite retreat, the seaside resort of Pajaro Dunes located south of Santa Cruz, a couple of hours by car from Palo Alto. Here they could spend three or four days together on unstructured tours of Ideaspace, the pressures of work carefully relegated to the background. Later in time and inspired by Kay and Jeffers, the more musically inclined group mem­bers took to engaging in ragged lunchtime jam sessions ("We played poorly but with great zest," recalled Goldberg, the house clarinetist)— including Ingalls on flute and Merry trying to keep up on a trumpet she had rarely touched since high school. "There used to be an old joke that we really didn't care whether a new recruit could do any computer sci­ence, what we really needed was a bass fiddle," she said.

  Even back home, Kay recalled, the group spent much of the daytime "outside of PARC, playing tennis, bike-riding, drinking beer, eating Chinese food, and constantly talking about the Dynabook and its potential to amplify human reach and bring new ways of thinking to a faltering civilization that desperately needed it (that land of goal was common in California in the aftermath of the Sixties)."

  Loose as the groups structure was, everyone understood where the intellectual power lay. It was in the combination of Alan Kay and Dan Ingalls. They were yet another PARC partnership, like Metcalfe/Boggs and Lampson/Deutsch, that was unlikely, implausible, and uncannily powerful. Ingalls filled a role in Kay’s professional life no one else had seemed willing or able to undertake. He was Alan s reality filter. It was no secret that Kay’s ideas tended to embrace about 200 percent of what was technically practical. But there were few people around willing to sift through the whole sandbox to identify the pragmatic parts. As Lampson recalled, "Alan would come around and say, I want to do x and y. You would ask him four questions about how this is actually going to work and you'd discover that it isn't actually going to work. So at CSL we'd say, 'No, we don't want to build that.'"

  But Ingalls was an instinctive master at picking out the subset of Ideaspace that was actually doable, and doing it. Even before the first Altos were designed and built, Ingalls had started working on one such subset. By the time the machines were finished, his efforts had yielded the masterpiece of computer science called Smalltalk.

  Smalltalk would make Kay's reputation more than Ingalls s, but Kay never forgot who transformed it from idea to reality.

  "Nobody would ever have heard of me," he said later, "if it wasn't for Dan Ingalls."

  Kay always claimed to get his best ideas in the shower. Conveniently, Building 34 had a shower in the basement
. More expediently, it was never in use during his most productive time of the day—from four in the morning, when he typically came to work, until about eight, when the rest of his team would start drifting in.

  Emerging from the basement one morning, he came upon Kaehler and Ingalls in a hallway bull session about "how large a programming lan­guage would have to be to have great power," as he recalled the scene. In a flash Kay posed a dare almost as audacious as the one Thacker had accepted from Bill Vitek.

  "With as much panache as I could muster, I asserted that you could define the most powerful language in the world in a page of code. They said, 'Put up or shut up.'"

  Kays challenge was grounded in his convictions about what a program­ming language should accomplish. Languages are much more than mere programs: They are blueprints for the thought process itself, software for the computer and its programmer. Almost from the moment he had encountered his first computer, Kay understood that the hallmark of a great system must be its simplicity. Only then can one be certain it has been fully distilled down to its essentials.