The Soul of a New Machine

by Tracy Kidder

  Rasala observed the battle with mounting anxiety and then with resignation. Most people, at one time or another in their careers as adults, reenact the molting of their adolescences. Rasala liked to say that he spent his professional childhood at Raytheon, and that at Data General, under West, he grew up. For a long time he felt afraid that West would leave and that he would have to take over the group, but gradually, during many long talks with West, he became fairly comfortable with the possibility.

  One day in September, at a fairly large meeting attended by bosses from outside the group, West criticized Rasala for some mistake. Feeling that he had deserved the mild rebuke, Rasala wasn't angry. Afterward, however, West took him aside. West explained that he was being accused of criticizing other groups but never his own. That was why he had publicly rebuked Rasala in the meeting: he was trying to prove the allegation false. He told Rasala he was very sorry he had done it.

  West, Rasala felt, was growing rather desperate, and it made I sad. At the beginning of November, Rasala remarked: "The ast two weeks I've had strange conversations with West. He's beginning to teach me the loose ends. I get the strange feeling I'm in finishing school."

  In retrospect, Alsing was convinced that West stayed around til the machine was safe from local politics. In November West ailed the team together and told them that he was leaving them; he was going to that distant country, the upstairs of Westborough, to take another job, connected with marketing, a job that would take him often to Japan. Thus he supplied them one answer to the question of what happens to computer engineers who pass forty. West said that Rasala, Wallach and Alsing would run the group now. He told them that their new projects were all lined up and tired them, "The fundamentals are right." He said he expected

  I to continue, as before.

  Some people in the basement wore broad smiles that day and many of the team's young engineers felt indifference. To the West loyalists, it seemed as though the earth had moved.

  In mid-November West went away on a business trip and Rosemarie cleaned out his office. The last item to go was the clock in the oak case. A technician who had worked under West for some years got a cart and put the clock on it. Rosemarie started leading him down the hall toward the elevator. Alsing saw them go by his office and he got up from his desk. "I'll hold the clock on the cart," he offered, "in case we go over a bump." Rasala joined the procession and so did Jon Blau.

  When they got upstairs, they were lost. They had to ask directions to West's new office, from "some stranger in a suit," as Alsing put it. West's new abode was like his old one — narrow, windowless. They put the clock on the floor and then they all sat down. Rosemarie bustled out, and in a moment returned. "You guys comin'?"

  Alsing said, "No, we're having a meeting in West's office."

  "It's our three-o'clock Friday meeting," said Rasala. Finally, they left. Going back downstairs, Alsing had the feeling he had just attended a funeral.

  I visited West not long after that ceremony. He confessed: "I feel awful withdrawal symptoms. I wake up at three a.m., sweating, worried and so on. It's a real problem becoming just an everyday Joe." But then he got up off his couch and lingered by his fireplace, a beer in his hand. "It's really tempting for me to look back on Eagle and bask in it now," he said. He made his long "Ummmmmmmmmh" and went on. "Next question you gotta ask yourself—is that a trap? "Ahhhhhhhhnd, the answer is: Yeah, probably."

  On his coffee table, lying among heaps of magazines, I noticed, was a tall stack of books about the Orient. West was going to help show engineers at the Japanese company Data General had partly acquired how to build Data General computers. He was going to become something of a far-wandering engineer again. He was going to help launch Data General's invasion of the Far East. I thought that someone ought to warn them.

  West was grinning now. He was saying: "Japan! The Orient! Man, I could really get into building machines in Japan!"

  West had risen from the small death of leaving the team. In departing, Josh Rosen had felt reborn. When I looked him up many months later, I learned that he had gone to work for another computer company and was defining the architecture of another 32 bit computer. He hadn't gone to a commune, except perhaps in a comparative, figurative sense. He was working eight-hour days and five-day weeks, and he professed himself happy at last. Dave Peck quit Data General, discovering in the process that after having worked at the company practically since its founding, he was entitled to a pension: when he turned sixty-five, he would receive the grand sum of fifty-three dollars a month. Steve Wallach gave the speech he had once dreaded, describing Eagle's architecture to a jury of peers, at a meeting of a society of computer professionals, and when he was done, they got up and applauded — "the ultimate reward," he said. But Wallach left the Eclipse Group to work in another part of the company, and a year or so later he left Data General. As for Carl Alsing, he too resigned from Data General and took a job in California — "at much higher pay," he said, but he added, "That wasn't why I left." Both he and Wallach had felt unappreciated by the company. In the project's aftermath, Alsing had most of his former responsibilities taken away. "I feel," he said then, with a small smile rather like West's, "that I am free to go."

  Unquestionably, Data General was changing. For one thing, the company was falling on hard times. Not long after Gallifrey Eagle was wheeled carefully down the hall to Software, Data General released a disappointing financial report, their first in many years. Profits were off. Over the next year and a half, the value of a share in Data General plunged, then rose, then fell again. More disappointing earnings reports were filed and speculation about the causes began in earnest. Business Week blamed de Castro's style of management mainly, but appeared not to understand that style any better than anyone else. It did seem that the company had let some of its product line become outworn, and while there was no telling just what had been lost by not fielding a 32-bit machine sooner, it was clear that without Eagle, Data General's present troubles would have looked far worse then they did.

  I did not think Data General had the look of a doomed enterprise, but rather of one that was suffering harsh growing pains. During this period, what seemed to be an unusually large number of people in crucial positions left the company; they included important sales personnel and several vice presidents — among them, Carl Carman. Downstairs, remembering that the Eclipse Group had once been called "a strong foundation," one of the team asked, "If we're such a great foundation, who are we holding up?"

  For at least some of the old hands, the feeling that they had gone underground to build their computer had never been stronger. Surfacing, they found the company in apparent turmoil. And what they returned to seemed nothing like the hero's welcome that they had expected. Some of the problem was inward and inevitable. A number of partum blues"; they spoke of feeling "that empty spot." Some of the problem lay in the local political situation. In the aftermath of West's departure, psychologists came down to visit the team and handed out questionnaires. These seemed designed, one young engineer felt, to find out what was wrong with the group, and he couldn't understand it. Why were they being asked that question, given what the team had accomplished? As for the top managers of the group, they felt that they were under attack. That situation eased within a few months, but the effects lingered. For the senior managers of the team, the project had this bitter end: that far from feeling rewarded, they thought that they were being neglected and maybe even punished for what they had done. Obviously, the company didn't intend to produce such a sour finale, but clearly those in authority didn't take in timely fashion all the right preventive steps.

  Bitterness wasn't general or enduring. Many in the group got promotions, business cards, a few free trips and, though they were a very long time in coming, stock options. To their pride and delight, a number of the team's junior members got to participate in the publication of technical papers about the machine. Most important, many got to play pinball — in the sense that they went to work o
n those interesting projects that West had concocted and left behind for them.

  In the fall of 1980 the Eclipse Group was disbanded and its members dispersed into several new and smaller groups. Some of the old crew mourned the team's passing. One said, angrily: "It was a group that was formed and achieved this remarkable thing for the company, and the company has deemed to reward that group by blowing it up. It's really sad." Many others, however, shrugged. They felt that the group's demise had been inevitable, I the one hand, and that it wasn't really dead on the other. After the reorganization, Ed Rasala decided to leave Data Genal, and although the authorities tried hard to keep him, he fi-

  chosen heir, Rasala was the last of the old group's managers to leave the team, and so his going had the status of a milestone.

  West stayed with the company. He had never imagined that the rule of pinball would be a lie for his loyal lieutenants, and he had dreamed of the group perpetuating itself. It took him a while, but eventually he seemed to find for himself a workable attitude toward the departures of his friends and the team's demise. "Yeah, it's all blown apart, but the ethic's still in place," he said. "In some sense spreading that around may be beneficial."

  West added: "It was a summer romance. But that's all right. Summer romances are some of the best things that ever happen." EPILOGUE

  Long before it disbanded formally, the Eclipse Group, in order to assist the company in applying for patents on the new machine, had gathered and had tried to figure out which engineers had contributed to Eagle's patentable features. Some who attended found those meetings painful. There was bickering. Harsh words were occasionally exchanged. Alsing, who during the project had set aside the shield of technical command, came in for some abuse — Why should his name go on any patents, what had he done? Someone even asked that question regarding West. Ironically, perhaps, those meetings illustrated that the building of Eagle really did constitute a collective effort, for now that they had finished, they themselves were having a hard time agreeing on what each individual had contributed. But, clearly, the team was losin its glue. "It has no function anymore. It's like an afterbirth," I one old hand after the last of the patent meetings.

  Shortly after those meetings, Wallach, Alsing, Rasala and Wes received telegrams of congratulations from North Carolina's leader. That was a classy gesture, all agreed. The next day Eagle finally went out the company's door.

  EPILOGUE

  In New York City, in the faded elegance of the Roosevelt Hotel's gilded chandeliers, on April 29, 1980, Data General announced Eagle to the world. On days immediately following, in other parts of the country and in Canada and Europe, the machine was presented to salesmen and customers, and some members of the Eclipse Group went off on so-called road shows. About a dozen of the team attended the big event in New York. There was a slick slide show. There were speeches. Then there was an impressive display in a dining hall — 128 terminals hooked up to a single Eagle. The machine crashed during this part of the program, but no one except the company engineers noticed, the problem was corrected so quickly and deftly. Eagle —this one consisted of the boards from Gollum — looked rather fine in skins of off-white and blue, but also unfamiliar.

  A surprisingly large number of reporters attended, and the next day Eagle's debut was written up at some length in both the Wall Street Journal and the financial pages of the New York Times. But it wasn't called Eagle anymore. Marketing had rechristened it the Eclipse MV/8000. This also took some getting used to.

  The people who described the machine to the press had never, of course, had anything to do with making it. Alsing — who was at the premiere and who had seen Marketing present machines before, ones he'd worked on directly — said: "After Marketing gets through, you go home and say to yourself, 'Wow! Did I do that?'" And in front of the press, people who had not even been around when Eagle was conceived were described as having had responsibility for it. All of that was to be expected —just normal flak and protocol.

  As for the machine's actual inventors — the engineers — most who came seemed to have a good time, although some did seem to me a little out of place, untutored in this sort of performance. Many of them had bought new suits for the occasion. After the show, there were cocktails and then lunch, and during this time most of the engineers stuck together, like young men at a dance. At lunch, they occupied a table all their own. It was a rather for 290 EPILOGUE

  mal luncheon, and there was some confusion at their table as to whether it was proper to take first the plate of salad on the right or the one on the left.

  West came, too. He did not sit with his old team, but he did talk easily and pleasantly with many of them during the day. "I had a great talk with West!" remarked one of the Microkids.

  He wore a brown suit, conservatively tailored. He looked as though he'd been wearing a suit all his life. He had come to this ceremony with some reluctance, and he was decidedly in the background. At the door to the show, where name tags were handed out, West had been asked what his title was. " 'Business Development,'" he'd said. At the cocktail party after the formal presentation, a reporter came up to him: "You seem to know something about this machine. What did you have to do with it?" West mumbled something, waving a hand, and changed the subject. Alsing overheard this exchange. It offended his sense of reality. He couldn't let the matter stand there. So he took the reporter aside and told him, "That guy was the leader of the whole thing."

  I had the feeling that West was just going through motions and was not really present at all. When it was over and we were strolling down a busy street toward Penn Station, his mood altered. Suddenly there was no longer a feeling of forbidden subjects, as there had been around him for many months. I found myself all of a sudden saying to him: "It's just a computer. It's really a small thing in the world, you know."

  West smiled, softly. "I know it" None of it, he said later, had come out the way he had imagined it would, but it was over and he was glad.

  The day after the formal announcement, Data General's famous sales force had been introduced to the computer m New York and elsewhere. At the end of the presentation for the sales personnel in New York, the regional sales manager got up and gave his troops a pep talk

  "What motivates people?" he asked.

  He answered his own question, saying, "Ego and the money to buy things that they and their families want."

  It was a different game now. Clearly, the machine no longer belonged to its makers.

  ACKNOWLEDGMENTS

  Many people — in the end, there were hundreds — worked on the project that this book describes. I could not name them all in the text, nor does it seem just to attempt to name them in this note, for undoubtedly I would miss someone. I have not even been able to name all of the hardware engineers who participated. Several made important contributions and are not mentioned. Others who also did a great deal of the work are awarded only a small amount of space in the narrative. To all, I offer my regrets, and I hope that they do not share them. In any case, no one who reads this book should imagine that its characters make up a complete cast.

  My thanks to all my friends and acquaintances in the world of computers, and to Susan T. and Muffet for help and hospitality. For their editorial assistance and encouragement, my thanks to Upton Brady, Peter Davison, Robert Manning, Sue Parilla and Michael Brandon. Thanks for general all-around help to Maureen Brown, Avril Cornel, Nina Engelhardt, Louise Desaulniers, Natalie Greenberg and Martha Spaulding; and for special assistance, to Paul Rich. Thanks for listening, to Stuart Dybek, Fred Hapgood, Jon A. Jackson and Mike Rosenthal; and for their excellent, ACKNOWLEDGMENTS 293

  timely counsel, thanks to Messrs. Rob Duggan, George Hall, James T. Kilbreth, Greg Pilkington, Ike Williams and Tim Rivinus; and for painstaking work on my behalf, my thanks to Mark Kramer. And many thanks, of course, for all of the above and more, to my parents and to Fran.

  />