The days before every workstation came equipped with a PC, Internet connection and e-mail can seem quaint to us now—a simpler time when the clacking of typewriters filled the air in office buildings, an instant message was walking over to someone’s desk and talking to them, and file clerks were entrusted to organize and protect top secret information vaulted inside steel cabinets. Nevertheless, to a large extent, today’s sprawling array of software apps, wireless gadgets and social networks owe their existence to a team of researchers that was assembled 40 years ago in California’s fledgling Silicon Valley to envision and create “the office of the future”.
Xerox established its Palo Alto Research Center (better known as Xerox PARC) in June 1970 as a West Coast extension of its research and development laboratories. PARC researchers proved wildly successful in pioneering many contemporary business technologies—the PC (the first was called the “Alto”), graphical user interface (GUI), Ethernet local area computer network (LAN) and laser printing, to name just a few. Xerox, however, was considerably less successful (and less interested) in commercializing much of PARC’s technology itself, leaving the door open for Apple, IBM, Microsoft and others to capitalize on PARC’s innovations.
PARC may have missed out on becoming a household name, but few could deny that the organization has demonstrated an uncanny ability envision technology way ahead of its time. Researchers there continue to work in dozens of areas, including water treatment, renewable energy generation, organic and printed electronics, and artificial intelligence. Earlier this month the National Science Foundation chose a team that includes PARC as one of four project teams to participate in the Future Internet Architecture program. PARC, which was incorporated in 2002 as a wholly owned independent subsidiary of Xerox, is part of a project (with nine universities) worth about $8 million to develop an architecture called “Named-Data-Networking,” which seeks to create a more flexible and secure network by identifying data through names (rather than numbers alone) and routing it based on those names.
We caught up with David Biegelsen, a charter member of Xerox PARC and currently a research fellow, to talk about PARC’s early days, its bittersweet successes and its future. PARC is holding a formal celebration of its 40th anniversary on September 23.
[An edited transcript of the interview follows.]
How did you come to join Xerox’s Palo Alto Research Center (PARC) in 1970? What were your expectations at that time? I was at Washington University in Saint Louis doing work on low-temperature physics when I met George Pake. He had offered to sponsor me for a junior fellowship at Harvard, but I said no thanks, much to my young wife’s chagrin. I wanted to go to an industrial lab. Later on George told me that he was going to start an industrial lab in Palo Alto and asked if I wanted to join. That was the extent of my interview with PARC.
What was PARC’s original mission, and how has it changed over the ensuing four decades? The charge given to George was, I believe, rather vague, and it took us at least a couple of years to figure out what was meant by “the office of the future” and what research we could do that would have a serious impact on Xerox’s business. The hope when I came to PARC was that we would have a five- to seven-year honeymoon during which we could do something really radical. [Joseph Carl Robnett] Licklider and Bob Taylor [who became the head of PARC’s computer sciences lab] had a notion of computers being able to talk to one another, a notion that gave rise to the ARPANET. But their vision wasn’t just having networks between computers. They wanted people to be able to interact with other people in a transparent way. Doug Engelbart [inventor of the computer mouse] had a similar vision at SRI [originally known as the Stanford Research Institute]. PARC’s Alto provided an incredible gravitational center for this work and pulled people together around a common goal of building the office of the future. That’s still our goal, although we’re an independent [wholly owned] subsidiary of Xerox now. We, and now many, many groups around the world, are still working on the implications of the technology we first demonstrated back then.
The first truly modern PC was the Alto, which featured a GUI and a mouse, and was developed 11 years before the Apple Macintosh in 1984. What are your recollections of the Alto? Did you use one either at work or home? We probably built a couple hundred of them at first, and every one of us had our own. The Altos had Bravo [a word processor program], an Ethernet connection, and servers for extended memory as well as printing. The Alto replicated various office tools, but pretty soon those other tools were left so far in the dust that comparing them with the personal computer seemed pretty lame.
PARC’s secretaries were given Altos as soon as possible so we could get real feedback [early industrial ethnography] on how they were used. The Alto maybe cost $20,000 in the early 1970s, but we knew that the costs would come down rapidly. Because we could spend money to build systems that would perform the way less expensive computers of the future would, we could use these stand-ins as platforms to develop all the software tools and interactivity that would be enabled. We thus had a real leg up on future competition. Unfortunately, it was very hard for some people to understand the implications of this. At the time, it was believed that with executives being males they would never learn how to type, and that it would be too expensive to give to secretaries. So who was going to buy one? You could already have access to a shared mainframe if you needed a computer and knew how to use one.
Mostly, the Alto was used for document processing—writing documents, modifying documents [using the new cut-and-paste features]. But what surprised everyone the most was their e-mail capability and how efficient it was. First, it was set up on the local area network and then it was expanded to a broader network within Xerox. I don’t remember exactly, but I probably sent and received four or five e-mails a day at that time. PARC’s location on the west coast was chosen to give researchers access to SRI, Stanford University, and other resources in the Palo Alto area. The distance provided a nearly 5,000-kilometer buffer between PARC researchers and Xerox corporate management in Rochester, N.Y., but it also made it difficult for the two sides to get on the same page. What are your thoughts on PARC’s location? Both of those statements are true. PARC’s location was a very thoroughly thought out decision by George Pake, [Xerox Chief Scientist] Jack Goldman and other Xerox executives. In the first 10 years or so Xerox did an incredible job of buffering us against the recession that hit the Silicon Valley [in the early 1970s]. While Berkeley Computer Corp. and others were cutting staff, we were bringing people in. George definitely wanted it to be near one of a small set of universities, but being near one on the west coast kept PARC researchers out of the day-to-day operations that took place in the corporate offices and labs back east. We had a much longer-term horizon than was allowed for many people in the labs in Rochester, N.Y.
Still, our communication with the corporate offices was not so much limited by the distance. I think it was really limited on both our ends by our different visions for the future and about how to commercialize the things we developed.
PARC is known for developing a wide variety of technologies, including laser printing, the Ethernet, the GUI, ubiquitous computing, blue lasers, MEMS (microelectromechanical systems ), natural language processing, flexible and printed electronics, and the first PC. What do you see as PARC’s greatest contribution to society? PARC’s biggest legacy was the Alto. We weren’t just creating a personal computer, we were creating personal computing, even as most companies, like IBM, were moving away from this direction because there didn’t seem to be a need for a personal computer. But PARC allowed people to come together and work on the systems that had to be developed around the computer, like Ethernet, the GUI, and all the software personal computers need and can enable. More important than the physical platform was allowing the interpersonal collaborations to occur that led to new tools. This is how evolution works—toolmaking is one of the things that allows humanity to evolve.
PARC has been criticized for not capitalizing on many of its inventions. For example, Apple took the mouse mainstream, and it became standard on all PCs when IBM-compatible computers switched from MS-DOS to Windows. What was Xerox’s thinking in not commercializing certain technologies? I was not directly involved in the computer developments, but [as an organization] it was hard because we wanted Xerox to bring to the world all of these advances that we found so exciting. It’s like being a child; you get all antsy to get it out there, and we were frustrated that it wasn’t getting out there. But also like a child, it was our own naiveté that brought out a $20,000 very capable computer that nobody really could afford to buy many of. It’s easy to overlook that these costs are very hard to drive down. You need suppliers and lots of other infrastructure to help scale the price down. The frustration wasn’t just Xerox’s inability to realize and profit from the gold mine they were sitting on, there were many other aspects.
What’s been the biggest misconception about PARC over the years? That the primary technology impact to come out of PARC was the personal computer. It’s really “personal computing” that’s our legacy, the ability to support interpersonal computing and collaboration. PARC is also responsible for many other technologies—obviously laser printing as well as high-power and other types of laser diodes that are the backbone of our telecommunications network; amorphous semiconductors, which became the backplanes for LCD displays; and now printed organic electronics which can enable flexible, lightweight displays and other devices. PARC has also spawned many companies like Adobe, 3Com, Spectra Diode, just to name a few. Many of Microsoft’s products like Word got their starts here as well.
What do you foresee in PARC’s future? PARC’s future is much like its past—providing our customers and the world with radically new ways of interacting with each other and interacting with our physical surroundings. [Former PARC computer scientist] Alan Kay said the best way to predict the future is to invent it. I would tweak that to say that you have to invent multiple futures because you don’t know which one[s] will prevail.
