The Friendly Orange Glow

by Brian Dear

This was not exactly true. Goldman and Pake kept a secret from Xerox historians for more than thirty years. The secret was that Pake was not Goldman’s first choice. Pake was Goldman’s second choice.

  When Goldman had asked Bob Sproull for names while they flew to New York, there was one name Sproull touted that Goldman recognized right away. “One of the guys he mentioned was Dan Alpert,” Goldman says. “He mentioned one or two others who were also well regarded and had made their way in the computer community. And since I knew Dan from my Westinghouse days—I had worked at Westinghouse, as my first job in research was at Westinghouse. And Dan was one of the very well-respected Westinghouse fellows.” One of the other names Sproull mentioned was Ivan Sutherland, the wizard of interactive computer graphics. Goldman himself had come up with the idea of Pake, someone else he knew, because, it turns out, Goldman, Pake, and Alpert had all worked at Westinghouse at the same time, years earlier, when they were all still climbing up the ladder of their careers. Pake had eventually gone on to a PhD and academia, and was now provost at Washington University in St. Louis. Alpert had gone to Illinois, Goldman to Ford. Goldman, four years younger than Alpert, greatly respected Alpert’s leadership and abilities. And now on the plane, Sproull was recommending him. “Since I knew Dan, I just picked up the phone and called him,” Goldman says.

  Sproull had also known Alpert for years. “I first knew Dan Alpert shortly after the war, I believe in connection with the Division of Electron Physics of the American Physical Society. By 1969 I had served with him on a number of committees and at the Institute for Defense Analyses and the Defense Science Board. I had worked at CSL the summer of 1951 and subsequently followed CSL and especially its educational technology. In 1969 I was both a director of Xerox and a consultant and was actively discussing the creation of PARC with Jack. I had a high opinion of Dan as a scientific and engineering leader and administrator, and it would have been natural that I suggested his name to Jack to direct the fledgling PARC.”

  It was the summer of 1969 when Alpert received the call from Goldman. Goldman was delighted to find out that Alpert was thriving at the University of Illinois, busy running the Graduate College and overseeing Bitzer’s lab and its PLATO project, which had just the year before received a large infusion of NSF money to ramp up the planning and design effort for PLATO IV. “I was so happy to see a great guy like Dan Alpert was succeeding in running it properly,” he says. PLATO “was strictly for the educational world, and I think Dan’s vision was just that, to be an available resource for schools, universities, and whatnot, around the country, around the world, for that matter. And I thought, gee, that is a wonderful idea, anything to take the initiative away from IBM. It was a nonprofit kind of thing, with government support, and that would have been great considering the state of technology in those days.”

  Goldman found PLATO intriguing, in fact the whole idea—and market potential—of educational computing was intriguing. Educational computing was something SDS, soon to be renamed Xerox Data Systems, or XDS, was interested in moving into. Xerox already was dabbling in the education market and wanted to dabble more. “The SDS people envisioned the Sigma computer as the ideal educational computer,” Goldman says. “So there obviously should have been somewhere in this case a marriage between what we hoped would happen with the Sigma computers, and the educational world, meaning things like PLATO. Now, we began to see flaws, after we took over SDS, we began to see flaws in their lineup of products, their whole approach to it. They were destined for ignominy very quickly because it was a time since they were producing largely for the research community, not for the business community, and in the ’69 time frame government funds were beginning to diminish. They were starting to have problems once we took them over….So in thinking about what role this new lab that I was contemplating would take, it was inevitable that I would look at where there are holes in the marketplace, that this company that just took over a computer company, and this lab which is supposed to be a lab backing up the computer business, what holes could it fulfill. And it was inevitable that we would think of the educational market. Because the educational market in those days was a market that Xerox was interested in….We’re involved in education, so a computer in the educational world would clearly be of interest to us, and in this respect, PLATO clearly stood out as an option.”

  It was summer, and Alpert was about to go on a family vacation to Snowmass Village in Colorado, where only the year before he had bought a plot of residential land at 8,500-foot altitude, right down the road from the ski slopes he routinely visited year after year. (He’d tried to talk Bitzer into buying some land nearby, but Bitzer didn’t bite.) Goldman had called out of the blue with what Alpert viewed as “a very attractive offer….I would have tripled my salary, plus a bonus.” At this point, Xerox had not decided where to locate the lab, though Goldman’s preference was New Haven, Connecticut, near Yale University. Santa Barbara was another possibility, as was Palo Alto. Says Alpert, “I could pick whatever place I wanted.” He still longed for the rolling hills of Palo Alto and the campus of Stanford University, where he gave the oral defense of his dissertation on the day before Pearl Harbor was attacked, twenty-eight years earlier. The Xerox offer was tantalizing: big money, big stock options, big responsibility, a lot more visibility than he had in his duties at Illinois, and a wide-open future that he could orchestrate. “It was an absolute sweetheart of a deal,” says Frank Propst. “It was just an amazing deal. And Don and I kind of were telling Dan, you know, you’re kind of crazy not to pick this up. This is a remarkable opportunity and something that could be very, very rewarding for you financially.”

  Beyond the beckoning riches, the offer would bring him right back where he’d always wanted to be: industrial research. He had only left Westinghouse because he had become disillusioned by the choice of director for that lab (and the fact that he’d been passed over for the position). Academia had not been his first preference, but the CSL job back in 1957 was attractive enough that he had taken it.

  And now, opportunity was knocking on his door once again. Dan Alpert, and, in a very concrete way, Don Bitzer, Frank Propst, the CERL staff, and the entire PLATO project, had a massive decision to make. Here is where Alpert’s account differs from Propst’s, who says he urged Alpert to accept the offer. “Bitzer and Propst,” says Alpert, “twisted my arm every way they possibly could, to stay. Because, well, they were living a pretty comfortable life. Hell, they had, in their boss, a guy who understood and cared how PLATO went and was supportive of them as individuals.”

  Alpert eventually had a long conversation with Peter McColough, CEO of Xerox. But he still wasn’t sure what to do. Alpert wasn’t very impressed with Xerox’s SDS acquisition, which he viewed as “one of the jarring things” about the corporate environment he would be parachuting into. “I could see,” Alpert says, “that my interests in computers and my talents were somewhat conflicting with the former president” of SDS, Max Palevsky. “It wasn’t much of a company, didn’t have any innovation to speak of, just making cheaper machines, making money by being more productive in production sense but not in a design sense….He and I had different perspectives on what the future of computing was going to be about.”

  There was one thing he was sure about: take some time off, get out to Colorado, and think about it. So he decided to stick to his plans and take his family on the vacation. They loved Colorado and were experienced backpackers around the Rocky Mountains. Out to Colorado they went. But he still could not stop agonizing about the Xerox offer. “I had a hard time making that decision,” he says. “I remember vividly looking up at the sky, the beautiful Colorado sky, wondering what the hell I should do.”

  When he got back to Illinois, he decided to reach out to a trusted friend, Arnold Nordsieck, a theoretical physicist at the university. “I talked to Arnie one time,” says Alpert. “I went out to see him and I told him what my dilemma was, and he said, ‘Well, Dan, follow your dream. Don’t get seduced b
y industrial firms. Keep on to your dream.’ ”

  Bitzer and others remember Alpert as a big worrier. The Xerox decision was a big worry. Alpert also brought the offer up to Bitzer, though was reluctant to discuss it deeply. “He agonized for some time over that kind of decision….He talked to a lot of people about things, but he kept his opinions of what he should do pretty close to his own chest. Now, we used to play squash almost every day then. And he’d bring it up and talk about it and all, and I’d say, you don’t want to do that.” Bitzer was used to the steady stream of job offers Alpert received: mostly high-level administrative positions at other universities. Xerox was a different story altogether, and Bitzer realized this was a good deal.

  Goldman tried to sweeten the deal. Says Alpert, “Jack Goldman said, ‘Well, come on out, Dan, you’ll have a budget, you can take the whole damn PLATO project if that’s what you want to do, to Palo Alto.’ ” Alpert vividly recalls this sweetened deal. “They said you can bring Bitzer here. That’s the way they put it, that’s the way Jack put it. He knew that I was a key figure over at Illinois, he knew they reported to me.”

  Decades later, when asked about Goldman’s sweetening the offer by inviting Alpert to bring Bitzer and PLATO with him, Bitzer paused, surprised. “I don’t think he ever told me that,” he said. “And he wouldn’t say anything like that if he had made up his mind first….I don’t recall Dan talking about moving PLATO out to Xerox, and the reason I don’t is he probably kept that pretty close to his self, he made up his own mind. But I do remember the occasion when everybody was afraid he might leave and I thought he was crucial and I knew it was a good offer and so you can’t say he didn’t want to do that because it’s stupid….You could see him agonize but he’d play very low-key, and we’d go and play squash and he’d talk a little bit about it.”

  There was another factor Alpert would have to weigh in his decision. Bitzer and Propst had formed a company called Education and Information Systems, Inc., or EIS. With the huge PLATO IV system coming down the road, it was clear that a lot of hardware was going to be required for PLATO, and like Owens-Illinois, which had taken on the plasma display manufacturing project, some company was going to have to take on the job of designing and manufacturing the peripherals for the interactive audio and microfiche slide mechanisms for the PLATO IV terminals. EIS was formed to do that. If all went well, and PLATO IV exploded, then EIS—and its owners—stood to do well financially. Dan Alpert had been invited to become an owner of the company and had agreed. (He described the arrangement as “the Troika.”) He too saw the potential for making some money should PLATO IV scale up the way they all hoped. Particularly if the system ever got commercialized, perhaps via Control Data. The scale might be huge. Given these considerations, EIS was something he could not ignore. “In that period,” says Alpert, “I was the guy to whom PLATO reported….I put a tremendous amount of my energy and spirit and life into that project.” If Alpert took the Xerox job, none of these great outcomes might happen, or they might happen without his involvement. If Alpert took PLATO with him to Xerox, one can easily imagine interesting consequences to follow.

  But it did not happen. In the end, he realized he had too much invested in PLATO at Illinois, plus he didn’t like Palevsky, wasn’t impressed with SDS, might lose out on EIS if he left Illinois, and the Xerox lab was still just a blue-sky notion with zero staff. It all added up to a feeling that it was too risky to move, and so Alpert turned the offer down. It was now November 1969. The PLATO IV system was right around the corner. EIS and its “Troika” might benefit greatly from the PLATO IV rollout. And with Control Data looming in the background, it was anybody’s guess how big PLATO could get.

  —

  Jack Goldman would keep Alpert’s rejected offer a secret for more than thirty years. No doubt the rejection stung, possibly reflecting poorly on his ability to attract top talent to Xerox and on the wisdom of launching this new lab. He decided next to recruit George Pake at Washington University, his second choice. Goldman found out that Pake had recently been offered, and then turned down, his own former research job at Ford. He called him on Thanksgiving Day of 1969. He did not tell Pake about Alpert. But Pake suspected that someone else had already been offered the job and turned it down. Pake, no dummy, knew exactly who to call.

  “I got this phone call after I turned down the Xerox offer,” says Alpert, “and first he said, ‘I wasn’t told your name, I wasn’t given your name by anybody, but I put two and two together and I figured you’re the only person it could’ve been.’ So, I said, ‘Yes.’ And he said, ‘So, why did you turn it down?’ ” Alpert gave his reasons, but encouraged Pake to take the job.

  The timing was better for Pake than it had been for Alpert. Pake was growing tired of the politics at Washington University. He accepted the offer, and then, like Goldman, kept the story of Alpert’s prior rejected offer secret, helping Goldman and Xerox, and Pake himself, save a little face. For more than thirty years he would play along with the alternative story that reporters and the historians of Xerox would establish as gospel. Alpert too kept the secret all that time, until asked about it one day thirty-four years later, in a beautiful home he had later built on that 8,500-foot-high plot of Colorado land surrounded by tall pine trees looming over Snowmass Village far in the valley below.

  —

  Pake ultimately decided PARC would be located in Palo Alto instead of New Haven, and began hiring. One of his hires was Bob Taylor, who had moved from ARPA to the University of Utah. Taylor eventually hired Alan Kay, and the rest is history. But one minor detail about Kay’s early days at PARC is revealing, in that it shows a lasting impact of that “big whammy” moment when he saw the plasma panel display in Illinois in 1968. “He wanted me to convert,” says George Pake, “a big chunk of PARC’s budget and resources to research on displays, including panel displays, and I looked at this very hard because a group at Stanford Research Institute was doing research on—I can’t remember what the display technology was—and it was sort of being cut loose by SRI and we could have hired them.” Pake looked around the landscape and saw the work that PLATO and Owens-Illinois were doing on manufacturing plasma displays, and saw that IBM had also begun a major effort to design and manufacture displays, having licensed the Illinois patent. Japanese companies had also begun work on what would become LCD displays, and later, plasma TVs also based on the Illinois patent. Pake realized that there was plenty going on in the industry already. “We couldn’t add much to that,” he says. “I told Alan, I appreciated the cathode ray tubes were too bulky and power hungry but I said why don’t you go ahead and build a prototype…of the Dynabook you wish you had, even though it’s not portable. He had already started to do that….So he did come to me to want to invest in flat panel displays and I just thought we didn’t have the resources to do that. It turned out retrospectively it was a good decision. We couldn’t have got very much farther with the displays that were already being done. And we didn’t have a lot of money either. Compared to IBM we were a drop in the bucket.”

  —

  It is worth taking a moment to ponder what might have happened if Dan Alpert had accepted Jack Goldman’s offer, packed up and left Illinois, and joined Xerox. Particularly if Alpert had done so only under the proviso that the PLATO project come along with him. This scenario is not that far-fetched: Alpert, after all, agonized over the decision for weeks and very nearly accepted the offer. Had he done so, very likely the Xerox lab would still have wound up in Palo Alto, Alpert’s old stomping ground that, despite the distance of time and career circumstance, he missed dearly. Consider the cascade of historical milestones that might—and, likewise, the milestones that might not—have transpired under this scenario. At the top of the list, Alan Kay would never have joined Xerox. He stated this in no uncertain terms to this author when presented with the Alpert hypothetical, such was the level of his dislike for PLATO. Without Alan Kay, there might not have been a leader within PARC pushing for “guar
anteed cycles” at the desktop level, and certainly there would have been no visionary pushing for the Dynabook and all that it entailed. As a consequence, the Alto personal computer would probably have not been made, or if it had been made, it would have lacked all the design details that Kay brought to bear. When asked about this what-if scenario, George Pake remarked, “I think that the whole push toward distributed computing would not have been as strong as it was.” The “Personal Dynamic Media” article might not have appeared in publication in 1977. Perhaps most crucially, there probably would have been no reason for Steve Jobs to visit Xerox PARC in 1979, only to be blown away by what he saw, taking ideas back to Apple, then subsequently coming out with the Lisa desktop computer, followed, famously, by the Macintosh, which changed the world.

  What would Silicon Valley look like today had Alpert and Bitzer and PLATO moved there in 1970? It is easy to speculate. What is certain is that much of the fabled history of the Valley might not have transpired, or might have transpired in quite different ways. It is also certain that PLATO itself would have evolved quite differently, particularly due to the lack, in the corporate office parks around Palo Alto, of the unforeseen, monumentally transformative influence and impact of the waves of creative high school kids and college undergrads who would soon wander into the CERL building and fall under the spell of the Orange Glow, as we will see in Part Two of this volume. Finally, in this what-if scenario, it is possible that PLATO would have died just as quick and ignominious a death as SDS soon did under Xerox, a company that would become known for fumbling the future.

  There is one PARC veteran who, when presented with this what-if scenario, held a more positive view. Unlike Kay’s immediate negative reaction to the notion of Alpert running PARC and introducing PLATO to Silicon Valley, Bob Taylor, the man who had recruited Kay to PARC, says he would have warmly welcomed Alpert. “I would have been a lot happier if Dan Alpert had taken the job,” says Taylor. “I liked Dan a lot, we were friends, and he would have been great to work for. I didn’t get along with George Pake at all. It was awful.”