It sounds difficult to believe now, but all of computing had to be invented. The ideas, techniques and metaphors which enabled the immense advances in hardware since the 1960s, and made them usable for the average user did not spontaneously appear in the ether, as one would these days tend to think, but very intelligent and highly driven people had to dedicate their lives to defining and implementing them. This book is the story of how the invention of computing unfolded in one of the most famous, coveted places for research on all aspects of computing: The Xerox PARC.
At the time of the founding of PARC in 1970, Xerox was a monopoly in the highly lucrative market of office copiers, thanks to the patents on a number of key technologies, and a highly trained and agressive sales force. The future was coming at it, however, in full swing: There were a couple of years left for the patents to run out, and computers were turning out to be a very important part of the “office of the future”, a domain where Xerox didn’t have an answer or even a running project. McColough, the CEO of Xerox who took over from Wilson, the legendary CEO considered to be the founder of Xerox, saw the writing on the wall, and hired Jack Goldman to become the chief scientist of Xerox. It is striking how many Ford executives Xerox hires in the run of this book. Goldman is one of the first, and also the most positive, as later hires turn out to be “bean counters” who didn’t think themselves in the technology business. It was Goldman who initiated the founding of PARC, with the first hires moving into the building close to the Stanford campus in July 1970.
When DARPA and later ARPA gave computing sciences their initial momentum following the Sputnik shock, one major focus was on making computing widely available. Since the existing kinds of computers were behemoths running in university basements, significant research was directed into time sharing systems for dividing and securing computing resources. J. C. R. Licklider, the legendary head of the Information Processing Techniques Office (IPTO) subdivision of ARPA, was key in funding these projects, but his vision went beyond these massive machines towards a “man-computer symbiosis”, which required research into more visually driven systems and HCI (human-computer interaction). Licklider’s efforts to fund such projects was continued by his follower in the same position, Robert W. Taylor. Taylor didn’t have the credentials of a PhD from a top university, but he was great at selecting and supporting researchers. This, and the fact that he was probably the person in the USA who was most intimate with the handful of practitioners of the very young domain of computer science, made him the ideal candidate for leading the PARC - which he didn’t get, due to his lacking a PhD, but he was made the director of the Computer Science Lab (CSL).
Through Taylor’s network, cunning and negotiation skills, PARC became host to the best computing technology talent at the time; so much so that the president of MIT at the time accused Goldman, Xerox’s chief scientist, of “destroying the ability of universities to teach computing because we were grabbing all the good people”. The story of where all the scientists and directors at PARC came from and how and why they ended up at the PARC takes on a 1001-nights-esque feeling, with stories embedded in stories. All important figures, Butler Lampson, Alan Kay, Chuck Thacker, Charles Simonyi (whose story is the most striking) etc. get the same treatment, in some cases following their career a number of positions back. This sometimes gets difficult to follow, as all the company, university and lab names leak into each other to create a mental mud ball of elite institutions.
For most people in this book, PARC is where they arrive, after going through education in other places. I think the reason for their arriving at PARC, instead of going on to other research institutions, is that they join others that are as inventive, driven and accomplished. The place was teeming with ideas for the future but also ingenious solutions to the daily problems. A great example is how Starkweather, inventor of the laser printer, managed to connect his prototype printer to the computer it could work with, following its sudden relocation to an office 2 kilometres away, with a highway in between. Figuring out that there was a line of sight between the buildings, he bought four telescopes, directed lasers between the buildings through these, and used this connection as a data line. As this example demonstrates, PARC scientists were able to implement their ideas using off-the-shelf products, e.g. the new Intel chip for their clone of the PDP, or lasers and telescopes for data transfer.
Once they settled in and created their own time-sharing computer by cloning the PDP, the PARC people set off to invent the office of the future for Xerox, with a visually driven, networked computer at its center. The result was the Alto, the first computer that worked with the modern desktop metaphor. It could be programmed with Smalltalk (invented and developed by Alan Kay’s group), controlled with a mouse, was networked with the other Altos in the same office over Ethernet (invented by Robert Metcalfe), and had a large “bitmapped” screen (I couldn’t really understand what that means, but it was revolutionary at the time). All these and more made Alto an ideal machine for accomplishing office work on: “One could now write memo, letter, article or dissertation and with the push of a button see it printed in professional-quality type” (p. 192). As the computers were networked with a reliable and fast implementation of Ethernet, an early form of networking culture evolved around the Altos: “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 without ever meeting their partners face-to-face.” (p. 212) The book makes clear that one reason PARC people invented the future and came this close to even selling it was that everything was built for actual use. Taylor imparted to everyone the one principle that “the things they built had to be designed for daily use” (p. 107). Lampson, based on previous experiences, proclaimed that “[they] should decide never to make something at PARC that isn’t engineered for a hundred users” (p.108). Such constraints went into the design of all components. The networking component, for example, was expected to be reliable, but at the same time not cost more than 5% of the overall cost of the computer (p. 185).
It might sound up to this point as if PARC were a bunch of uber-nerds having the time of their lives while creating incredible things, and you must have sensed right at the last comma that this wasn’t the case. Taylor organized his group around the existing culture of “alpha nerds” discussing matters in what they held to be a rational manner, and everyone who has been in or seen such environments knows that they are neither rational nor free. To start with, most - well, nearly all of these people are white men. Exceptions are Adele Goldberg and Diana Merry, both in Alan Kay’s group, and Lynn Conway who pioneered modern VLSI technologies at PARC. I couldn’t find a single mention of a person of color. Conway even actively avoided joining the team that worked on the Alto, because she found the environment, especially Lampson’s behavior, too agressive: “You can be so confrontational and challenging about how smart you are that you can’t always see that somebody else has got this cool idea” (p. 306). An important part of PARC culture was the Beat the Dealer sessions, where an applicant (or sometimes just a researcher looking for listeners) presented their idea, and then the audience went on to aggresively dismantle it. As the people in attendance were among the smartest and most knowledgable in their areas, the presenter would have a hard time if they weren’t exceptionally well-prepared. Some people, however, sensed early on that the treatment wasn’t always fair: Robert Metcalfe, inventor of Ethernet, remembers how applicants from certain elite universities were handled with velvet gloves, while “if you were some poor schmuck from the University of Arizona, they’d grill you and it was all over” (p.149).
So what happened to the Alto? Why do we remember Apple as the inventor of the graphical desktop, and not Xerox? Xerox management is what happened. The original Alto left a deep impression on anyone who used or even witnessed it; those intimate with computers of the time knew right away that it was a game changer. A couple of Altos were even sold and installed outside Xerox, to the Swedish Telecommunication Agency and, weirdly enough, Jimmy Carter’s White House. The Alto was actually even put into a better shape more suitable for the general public, and was slated for release: “For a few short, glorious weeks, official Xerox policy was to serve the growing market for electronic word processing with the Alto III, a programmable personal computer that would bear the same relationship to the competition’s glorified typewriters as a Harley does to a tricycle” (p. 263). But management reneged on this decision, ostensibly for reasons of production costs (despite Xerox’s own experts disagreeing), putting its weight behind a new model of typewriter instead - yes, an actual typewriter. It’s impossible not to conclude that this was an egregious case of mismanagement. Xerox did end up marketing a number of technologies invented at PARC. The most profitable new product Xerox marketed, the laser printer, was invented there by Gary Starkweather. Ethernet was also published as a standard, and Xerox sold hardware for it. Apart from these handful of succesful products, Xerox very clearly missed the opportunity to become the company that invented and profited from personal computing.
One of the most interesting chapters in the book is dedicated to what is probably the best-known event in PARC’s history: Steve Jobs’ visit. Hiltzik is able to paint a much more detailed and nuanced picture of this visit than folklore has it, thanks to the interviews he conducted with the major figures involved. There were, for example, two visits, and not just one, because Kay’s team refused to show Jobs and his leading developers the most exciting things on their first visit. Jobs actually had to use his clout with the higher-ups to make them “order” a more detailed, all-revealing demo. At this second visit, Bill Atkinson, the famous developer who led the development of the user interface of Lisa (precursor to the Mac), got a chance to watch the people who wrote the software for Alto actually use it, and pepper Goldberg and Ingalls with questions. He later said that, more than the actual implementation, what drove him was to see that the feats they were demoing were in fact achievable. Another misconception is that Xerox didn’t “get” anything from this visit, but only divulged tech it could have itself developed into a product. This is not completely true, as the visit was a quid pro quo, wherein Xerox would get a chance to invest into Apple in the last funding round before it went public. This reads like the most lucrative investment chance in the history of venture capital, and Xerox did go ahead to invest, but in a very Xerox-managementy move, the company divested itself of the investment before the IPO, and didn’t get to reap the benefits.
Hiltzik’s prose is far from being academic or boring. He sets up story arcs that span chapters, and connects the figures’ personal histories and idiosyncracies to the ongoings in the lab. He knows where to inject the occasional anecdote, without the book turning into a collection of pithy stories. If you are working with computers, the text might come over as sparse on technical detail, which is completely understandable, as it is written as a journalistic piece for the wider audience, but it is never misleading or exaggerated.
This is quite a big book (and hence the long review), but it’s never boring or wordy. If you are interested in the history of computing, or modern technology in general, it should definitely be on your reading list.