The Soul of a New Machine

by Tracy Kidder

  Although not a bad student, Rasala had the misfortune of following his brother through school, and from the elementary grades on, his teachers never stopped reminiscing about his brother. "From grade school on, it was always my rap — I never lived up to my potential," he said. He didn't sound bitter, he seemed to believe it was a matter of fact.

  Rasala went to Rensselaer Polytechnic Institute and studied electrical-engineering, including logic design. But this was during the early sixties, and he never laid hands on a computer until he went to work at Raytheon, right after college. Rasala never had the feeling, which others in the Eclipse Group remembered, of having his true calling suddenly revealed to him the moment he touched a computer. And it took him seven years, he said, to recognize "the opportunities and pleasures of working.

  "To me, it was more important to play Softball, take karate lessons, and play hearts at lunch." At Raytheon he carved out for himself an undistinguished, comfortable niche, doing routine assignments. When he finally did get an assignment that he could not fulfill by looking up answers in books, he worked hard and succeeded, somewhat to his own surprise. That project whet his appetite. But he knew he had established a reputation for being "an average guy," and he did not think other interesting assignments would come his way at Raytheon. To shuck that reputation, he decided he had to start over at another company. In a trade paper, he saw an advertisement for Data General — not one of the early, brazen ads, just an ordinary listing: "Engineers Wanted." On a whim, he drove out to the company's headquarters and said, in essence, "Here I am." Engineers and executives, including Carl Carman, interviewed him on the spot, and Rasala was impressed. "At Raytheon, I never even saw someone at Carman's level." Data General offered him a job that very day and he accepted their terms without so much as an argument. "I wasn't very good at bargaining, I guess." And he went to work, not on the job he thought he had been hired to do, but on a piece of the hardware for the first Eclipse. "I was had," he recalled, and he smiled.

  When Rasala arrived, all that remained to be done on his part of that Eclipse hardware were finishing touches. West, who was just another engineer then, noticed Rasala, however. He liked the dogged, straightforward way in which Rasala went about finishing up. As Rasala himself observed much later on: "With Tom, it's the last two percent that counts. What I now call 'the ability to ship product' — to get it out the door." Rasala looked at me squarely. "And I may not be the smartest designer in the world, a CPU giant, but I'm dumb enough to stick with it to the end."

  West called himself "a mechanic" — one who could take the ideas of engineers technically brighter than he and make those ideas work; he thought he had spotted that quality in Rasala, too, and he had promoted it. West groomed Rasala as a leader of hardware teams. Eventually, he put Rasala in charge of the hardware for the biggest of the 16-bit Eclipses, the M/600. As often happened, the pace of that project became frenetic. It took a year of increasingly intense labor from Rasala, and when it ended he felt tired. He was looking around for something easy to do when West asked him to manage the development of Eagle's hardware. Rasala declined.

  In his office, the door closed, West schemed. "How am I gonna get Rasala to sign up?" he asked Alsing. Unlike Wallach, Rasala wasn't a student of advanced architectures. Rasala wasn't worried about Eagle's supposed inelegance; he was just tired. Offering him a chance for vengeance wouldn't work, because Rasala had been pounding out the M/600 during the EGO wars and he didn't have anything to feel vengeful about. Rasala was a hard one, West said. Bringing him to the signed-up condition was mainly a matter of persistence. West presented Eagle to him as a test of strength. Could they get this machine out the door on time? Again and again, he told Rasala that the company needed this machine desperately. Rasala never actually said, "I'll do it." One day, it seemed to him, he was just doing it.

  For months afterward, he would come home at night and his wife would ask him, "How was your day?"

  "It was terrible," Rasala would tell her. But as he went on describing the day's events, his wife noticed, he became increasingly excited.

  "Maybe it's masochism," Rasala said. "But I guess the reason I do it fundamentally is that there's a certain satisfaction in building a machine like this, which is important to the company, which is on its way to becoming a billion-dollar company. There aren't that many opportunities in this world to be where the action is, making an impact." It struck him as paradoxical, all this energy and passion, both his own and that of the engineers around him, being expended for a decidedly commercial purpose. But that purpose wasn't his own. He had enjoyed his years at Raytheon; life had been pleasant there and he had been an easygoing fellow. Now, in the Eclipse Group, for several years in a row he had been ¦ working overtime without extra pay in an atmosphere that was decidedly not easygoing. Why had he made the switch to Data General and now signed up to work on Eagle?

  Rasala said, "I was looking for" — he ticked the items off on his fingers — "opportunity, responsibility, visibility."

  What did those words mean to him, though?

  Rasala shrugged his shoulders. "I wanted to see what I was worth," he explained.

  * * *

  More than anyone else in the Eclipse Group, Rasala borrowed expressions from West: fundamentally; basically; the win; some notion of; quick-and-dirty; the ubiquitous canard; the long, drawnout and, as in, "So and so says such and such, ahhhhhnd, it's a canard." But Rasala tended to qualify and expand, as if he feared ambiguities, whereas West was inclined, in his daily commerce, to the short, forceful sentence. West's words and phrases just didn't sound the same on Rasala's tongue.

  The team was having problems with one of the support groups. They believed that this group was slowing Eagle's debugging; this group wasn't trying hard enough — which was to say, they were working just normal hours on Eagle's behalf. They believe Eagle will fail, Alsing thought; he could see it in their eyes, he said. West stood in a corridor discussing the problem with Rasala. "I'm gonna detonate those guys," said West, in a flat, calm voice, as if he were planning a business trip. Then he promised to "waste 'em." He added, "We'll string them up by their toes."

  Months later, looking back, Rasala decided that when West took on people in other groups and "beat them up," as this sort of activity was called, he did so in a purposeful way, one that was usually "not so much mean as intimidating." In one fairly typical case, a support group missed a promised deadline. At a meeting, West questioned the group's leader. The leader said that he hadn't received a certain piece of equipment yet. West would not let the matter rest there. He wanted to know why it hadn't been shipped and what was being done about it. West was "always pushing," Rasala noticed. But the act, Rasala would come to believe, was never as dramatic as the language West used to describe either what he was going to do or what he had actually done. For a long time, however, Rasala took West's descriptions as a fairly literal model of how to contend, and when Rasala confronted a troublesome support group, he beat his fist on the table and shouted and made open threats. I saw him after one such encounter, returning to his cubicle. He was red. "I don't want to beat people up," he said, throwing his hands down. "I don't want to be a bad guy. I just want to get something done." He took some deep breaths, then proceeded to explain that it really wasn't the support group's fault. Eagle wasn't their machine, they couldn't be expected to work as diligently on it as the Eclipse Group.

  Bob Beauchamp remembered, by way of drawing the contrast between West and Rasala, receiving a set of orders from West. "Do this," West said to him, and in West's tone of voice, Beauchamp heard, "No ifs, ands or buts, no arguments, no commenting, no nothin' —just figure out how to do it and get it done."

  "Did it work?"

  "Yeah, it worked," Beauchamp said. "When Tom said do something, people did jump." He went on, "Rasala was practicin' that, I think. He could bark too."

  Beauchamp grinned. He and Rasala were good friends now. Before they were, early in the project, Beauchamp was standin
g around with one of the Hardy Boys one evening, talking casually, when Rasala came out of his cubicle and, looking them up and down, said, "Don't you guys have work to do?"

  "What the hell?" Beauchamp thought to himself. "It's after working hours." So he said to Rasala, "No."

  To Beauchamp's surprise, Rasala didn't say any more. He just walked back to his cubicle.

  "Then I knew," Beauchamp said.

  Rasala participated in West's campaign to create the illusion of a machine before one existed, and on the whole Rasala endorsed the strategy. But he worried. Clearly, he could not make West's approach his own. "We're getting the rest of the company geared up now, and we could fall flat on our face. This whole thing scares me, because we're pushing." At West's insistence he had made up a debugging plan that would bring Eagle in by April. Then he made up one that would bring the machine home by May. In mid-March, when Rasala knew that they wouldn't come close to that second mark either, he went to a meeting with West and heard his boss insist to a number of people in others groups that they would have all the bugs out by May. "Now May one is the target and we're not gonna do it," said Rasala when he came back. "It's sort of a poker game. Everyone's bluffing, but everybody seems to know the rules of the game, and I don't know those rules yet." He continued, "I strongly believe in supporting my boss, because I know he'll support me. So I try to be consistent with what Tom is saying, one, and two, I try to be as honest as possible."

  One day Rasala bought a self-help manual that promised on its cover to hold the secret to understanding one's boss.

  Just a few minutes after I first met him, Rasala told me, "I'm not a bright guy." I wrote that down in my notebook, and for the rest of the evening, he kept saying: "But I'm a dumb guy. I must be, you wrote it down." Rasala was far from dumb, though. He was just somewhat puzzled and, at the same time, bent on self- improvement.

  When Rasala said something earnestly — if they failed, it would be his fault, no one else's, he told me, for instance — you never felt that he meant something else. Somehow he found the time, after working what usually amounted to a ten- or a twelve- hour day, to drive into Boston one night a week and take a course in a new programming language. He was going to fill in the gaps in his technical education, and he would not skip a class, not for a snowstorm or even a ball game.

  Most engineers, I think, consider themselves to be professionals, like doctors or lawyers, and though some of it clearly serves only the interests of corporations, engineers do have a professional code. Among its tenets is the general idea that the engineer's right environment is a highly structured one, in which only right and wrong answers exist. It's a binary world; the computer might be its paradigm. And many engineers seem to aspire to be binary people within it. No wonder. The prospect is alluring. It doesn't matter if you're ugly or graceless or even half crazy; if you produce right results in this world, your colleagues must accept you. It's an ex citing environment to contemplate; you can change the way people think if you can provide the right reason, and you can predict the way in which others may change you. Since there are only right or wrong answers to questions, technical disputes among engineers must always have resolutions. It follows that no enmity should proceed from a dispute among engineers.

  Clearly, West believed in these precepts — designs were right or wrong; engineers were winners or losers and those judgments had nothing to do with whether he liked them personally or not. But there were many reasons to doubt the reality of this binary world. What had the EGO wars proven but that talented engineers can dispute narrow technical issues and never come to agreement? Amazed at the alacrity with which West sometimes consigned engineers around him to the status of winners and losers, and convinced that West was usually right, Alsing nevertheless always wondered to what extent West made his own judgments come true. Rasala had always believed that a technical argument between two engineers couldn't lead to hard feelings. But he had recently lost his oldest professional friend because of what he had imagined to be a technical disagreement.

  An engineer was supposed to be eager to advance in his company's hierarchy, and Rasala was, but he had other dreams. Some years before, he had taken a trip to Jackson Hole, Wyoming, and ever since, he had wanted to go back there and do something simple: he thought he'd like to open a grocery store in Jackson Hole someday. To Rasala, many of the young Hardy Boys seemed hip and cultivated in ways that were foreign to all his previous experience of engineers. But he did not for that (or any other) reason disapprove of them. And he didn't seem to envy them or want to change his ways and be just like them. He talked about them with plain curiosity, as if he were a traveler in their country.

  One evening, over beers, Rasala complained about some insipid movie recently shown on TV. He compared this stinker to The Prime of Miss Jean Brodie, an example of a film he'd liked. Actually, he had admired the protagonist most of all. "I don't know why," he said. "She's romantic, foolish, unrealistic — everything an engineer's not supposed to be. But I like her." He gave a short nod, as if to say, "This decision is final." Then he said it again. "I like her."

  Rasala wasn't West and he wasn't dumb either. He seemed to be in transit.

  When the hardware team started to design Eagle, Rasala opened a sort of diary. It reads like the journal of a frontiersman's travails, except that instead of wolves, hostile Indians and busted wagon wheels, the reader encounters the indecisiveness of logic designers, an affliction that causes a designer to "spin" from one possible approach to another; the regular breakdown of the computer they were using to work on their computer; and the tendency for all bright young engineers to redesign and redesign in search of perfect unattainable solutions. In the diary, one delay follows another. Schedules always slip. The last entry, recorded several months before they stopped designing, reads, "Overall, things look lousy."

  This was the biggest job Rasala ever had, and he was just finding his way around in it when he wrote that journal. I got the impression that a few engineers doubted his suitability, on the grounds that others were more brilliant than he. Rasala himself said: "I'm an implementer. I'm not gonna go out and invent anything. But making it work is fun. It's something I think I do reasonably well. I don't have Wallach's knowledge, I'm not on top of the architecture per se, but I'm a good designer, I think, and I'm a better debugger."

  The sheer physical complexity of the machines demanded that the debuggers plod, and plodding did not always come easily to the quick, bright Hardy Boys. Were signals just barely making it to their destinations between ticks of the computer's clock? If so, they had better go back and work on problems they thought they had solved. Was there a lot of noise inside the machines? Noise, as Rasala explained it, is what makes your TV go haywire when you turn on your blender — stray voltages that are propagated through the machine; and too much noise fouls performance. Noise is not the most interesting of problems, but coping with it requires experience and imagination and a certain doggedness. On one occasion, the Hardy Boys found that while one board was failing, another one, identically wired, worked right. The problem, they discovered, lay in a single chip. They wanted to throw that chip away, put in another, and go on. But Rasala would not allow it. The suspect chip might not be defective; it might just be working a little more slowly than most of its kind. In that case, they would have a real problem to fix. It might never show up again in the lab, but when the time came to mass-produce this machine, it would almost certainly come back to haunt them. Some slow chips were inevitable, and they had to provide for them.

  In the main, Rasala's technical role was to play the brake. When the Hardy Boys complained loud and long about updating boards every day, he bent the rule, but insisted that they perform this crucial, unpleasant task every Saturday. When he put them on two shifts, he made a point of working the better part of both. He was a good West Pointer, he led by example.

  Rasala took a week off from the debugging that summer and spent the time building a porch on his house. I came by one d
ay to give him a hand and got a dose of his style of management.

  He lived in a fairly new house, of quasi-Colonial design, located in one of the many residential neighborhoods that had sprung up in the area more or less in rhythm with the growth of the numbers on the bottom line of Data General's annual reports. The numbers had grown faster than trees do. His house lay so close to headquarters that he could have ridden a bike to work, although, given the traffic nearby, it would have been a dangerous commute.

  While working with Rasala, I made a careless reading of a carpenter's level and we consequently made the frame of the porch just a shade cockeyed. Rasala stood back and stared at the frame for a while. Then he said, "Ship it."

  He teased me about the mistake, in a friendly way. He also prodded. When I arrived, he said he expected to finish the entire job that day. Around three in the afternoon, I said I was tired. "Tired?" said Rasala, reaching for a tenor's high note. 'Tired? You're not tired. How can you be tired? You mean to say you're tired already?"

  Somehow, I felt I had to keep working. An hour and a half later we still weren't finished, and then I really felt tired and said so. Rasala let go on the instant, brought out some beer, and declared the day a success.

  For most of their time together, most of the Hardy Boys got along well with Rasala. He teased them and he prodded them, and they did the same, both to him and to each other. One evening just before going home, Holberger made a minor, careless mistake. Discovering it, Rasala told the rest of the crew, "I hope you're gonna leave Holberger a nasty note. He'd leave you one." Rasala liked a contentious atmosphere, a vigorous, virile give-and-take among himself and his crew. "Smart, opinionated and nonsensitive, that's a Hardy Boy," he declared.