After fumbling the future,Xerox PARC is back with a visionary new director, bright researchers, and amazing new technology.
In 1983, I read an article about mind-amplifying machines. These were as different from the computers of the time as television was from 15th century printing presses. The article's author, Alan Kay, worked at Xerox Palo Alto Research Center (PARC). I concluded that PARC amounted to a hive of zealots bent on changing the course of culture. I wanted in on it. Most of all, I wanted to test drive one of the devices they used to amplify their own minds.
I managed to get a job at PARC writing articles for a Xerox in-house magazine. A year before the first Macintosh computer was sold, I commuted 45 minutes to PARC's rural campus to type on an Alto - the first true personal computer. My job was to interview PARC researchers about their work. They talked about bit-mapped screens, local-area networks, point-and-click interfaces, object-oriented languages. All those futuristic experiments they showed me have diffused so widely (and profitably) during the past ten years that it is easy to forget they were once confined to that building on a oak-spotted hill above Silicon Valley.
The players on the all-virtuoso team that convened then at PARC's computer labs were eager to volunteer as guinea pigs for the technology they wanted to build. They were dedicated to designing their own personal, networked mind-amplifiers, then handing over their creations to non-programmers for intellectual work. One phrase became PARC's hallmark: "The easiest way to predict the future is to invent it." They succeeded so well that nothing has been the same since, including PARC.
The early days of PARC are the stuff of Silicon Valley folklore. Personal computers did not spring naturally from the computer industry. They were deliberately realized by a radical fringe, against all the force of the day's accepted wisdom.
These zealous wizards handed Xerox an astounding lead in information technology in the early 1980s, but by the end of the decade, Xerox watched as upstarts like Apple and Microsoft grew wealthy off Xerox's discoveries. Neither Apple nor Microsoft even existed when the first Altos were designed in the early 1970s; by 1990 either company could have bought Xerox. The tragicomic Xerox saga is recorded in Douglas K. Smith and Robert C. Alexander's Fumbling the Future.
But Xerox hasn't always fumbled the future. The company persists because long ago it redefined itself and reinvented its business. Fifty years ago, Chester Carlson invented a scheme for printing dry photocopies of documents. Few corporations of the day were interested in an expensive, messy, unreliable way to do what carbon paper did cheaply and well. Nonetheless, Carlson's company, Haloid, committed itself to a future in which people would use office copiers for purposes unimaginable in the carbon-paper era. Renaming itself Xerox, the company championed the copiers that soon rendered carbon-paper obsolete.
Having already invented the future twice, then squandering its advantage, could Xerox still have stories to tell? In the fall of 1993, I returned to Xerox PARC for another chance to go back to the future. Where was Xerox? On the road to reinventing itself for the third time.
New, young faces populate the halls today. Video windows and audio communications are built into workstations. Desktop screens have evolved into wall-sized screens, clipboard-size screens, and pocket-size "tabs." The place is still an intellectual wonderland. Again I feel like I've dropped in on an outpost from the future. It's not just the latest gadgetry. Something's happening. The hardware and software shops are still cooking up new goodies, but now anthropologists and sociologists study the nature of knowledge work and the way organizations function. These scientists are as important at PARC today as are inventors of thin-film displays or engineers of RISC processors.
In the 1970s and 1980s everybody at PARC shared a mental model of the future - they predicted that millions of people would soon use screens and windows and mice, linked to high-speed communication pipelines. But now in the 1990s, nothing seems as easy to forecast or design as personal computing and networking were. This time, the folks at PARC are taking their info-tools into relatively uncharted territory - what happens among people.
Personal computers and local-area-networks are boxes of electronics, cables that transport bits. But once they are in place, the effects on organization are more powerful than number-crunching and file-transferring: People use those boxes and cables to think, communicate, and solve problems in new ways. Those new ways are not visible, but they represent the raison d'etre of innovative information processing and communicating tools. These tools, after all, don't generate electrical power or material goods. Through the interaction of minds, they effect changes in human relationships, and through those, changes in organizations. This territory - what happens to people's minds and to organizations after personal computers and networks are installed - is the new terra incognita that PARC researchers are exploring. What they have discovered so far, in the earliest phases of the research, is both startling and inspiring.
Postmodern Computing
The first person at PARC I talked to this time around was Mark Weiser. I had never met this smiling, bearded fellow before, but I recognized the room he was standing in.
I sat on the same couch in the same corner office with the same view of Palo Alto ten years earlier when Bob Taylor was director of PARC's computer science lab. At that time, as now, PARC was a one-stop shop for everything from new microcircuits (and new ways of making microcircuits) to new computer systems and new software. Taylor had been a young administrator who led a stellar, single-minded team for more than a decade, with legendary results. Taylor and his crew were the heirs to Douglas Engelbart's crusade to "augment human intellect" at the Stanford Research Institute.
As silly as it seems today, there was a time when almost everybody in the computer industry was convinced that computers were destined solely for calculation or data processing. Engelbart, a lonely visionary, pushed the radical idea that computers could be used to extend the power of human minds to think, communicate, and solve problems - this was "augmentation." There was a threshold out there, Taylor's early PARC team believed, measured in processing power, where true intellectual augmentation would take place - the way 24-frames-a-second marks the boundary between still photography and cinema.
Bob Taylor's motto had been "Build what you use. Use what you build." One of the first things Mark Weiser shared was his own interpretation of this now hoary PARC wisdom: "You let what you build change you, and you move on." In Weiser's view, PARC learns how to build better tools, and then everybody learns to use them. Then they do it all again, with the new tools. The people at PARC have a hunch that this organizational bootstrapping might be worth more to Xerox than the tangible artifacts the process produces.
The early computer science lab crew stood on a mountain, and the destination was clear. The sights of Weiser and company are more arcane - they are mapping aspects of intellectual work that have yet to be explored.
Weiser wants computers to disappear into the background. When computers become invisible to users, the most important side of human-computer symbiosis (to humans) has a better chance to emerge. How to make them invisible? Make them ubiquitous. "Ubicomp" Weiser calls it.
To Weiser, the intellectual origins of "Ubicomp" lie in the social rather than the technical side of PARC's research: "Theidea of ubiquitous computing first arose from contemplating the place of today's computer in actual activities of everyday life," he writes in a recent paper. "In particular, anthropological studies of work life teach us that people primarily work in a world of shared situations and unexamined technological skills. However, the computer today is isolated from the overall situation and fails to get out of the way of the work."
In Weiser's cosmology, the original "desktop" graphical user interface does not go far enough to get out of the way of the user. Interfaces don't do the job. Neither do agents. And certainly not virtual reality (VR). Weiser writes: "In its ultimate environment, VR causes the computer to become effectively invisible by taking over the human sensory and affector systems. VR is extremely useful in scientific visualization and entertainment, and will be very significant for those niches. But as a tool for productively changing everyone's relationship to computation, it has two crucial flaws. First, at the present time [1992], and probably for decades, it cannot produce a simulation of significant verisimilitude at reasonable cost. That means that users will not be fooled and the computer will not be out of the way. Second, and most importantly, it has the goal of fooling the user - of leaving the everyday physical world behind. This is at odds with the goal of better integrating the computer into human activities, since humans are of and in the everyday world."
Embedding intelligence in the environment is not possible with present-day technology, so Weiser's group started with wall-sized interactive screens known as "boards," clipboard-sized terminals known as "pads," and tiny computers called "tabs." The first stages of Ubicomp bootstrapping incorporated another innovation, first created by Olivetti's research center in Cambridge, England - an "active badge." With an active badge system, every computer you sit down at is your computer, with your custom interface and access to your files, because your active badge sends it information via infrared signals. It is possible to track the locations of other researchers at all times by central monitoring of active badges - a handy tool with Orwellian implications.
In his office, Weiser drew a horizontal line. The far left side he labeled "atoms"; the far right side, "cultures." PARC, Weiser points out, directs research at every part of the spectrum defined by the line between atoms and cultures: toward the "atom" end of the line is PARC's ongoing exploration of the nature of materials. Toward the center lies PARC's work in turning materials into real hardware products. Smack in the middle is the task tackled by Taylor and his group, integrating hardware and software design into information systems. And toward the right, closer to "cultures," are systems and practices - the hunting grounds of social scientists.
The lab's new direction, Weiser says, "recognizes even more that people are social creatures." He referred to his ideas as a form of "postmodern computing," in that he wants to "return to letting things in the world be what they are, instead of reducing them" to data or virtualizing them into illusions. "Ubicomp honors the complexity of human relationships, the fact that we have bodies, are mobile," he said. Tabs, pads, boards, and badges are the first bootstrapping steps in that direction, not the long-term goal. Nevertheless, part of that environment has migrated into productland and is embodied in the "LiveBoard" (see WIRED 1.4, page 36).
PARC teams are way beyond "fumbling the future." They are inventing the future again. But this time they are reinventing their understanding of how to invent the future, as well. There is a sense that they are onto something new, that instead of extending the old computer revolution into new widgets and gadgets, they are at the dawn of a whole new - and potentially scary - revolution.
PARC is an intellectual playground, full of free spirits. How do they feel about the possibility that Ubicomp might lead directly to a future of safe, efficient, soulless, and merciless universal surveillance?
"Some people refuse to wear badges," Weiser says. "I support their right to dissent. And one principle we go by here is to maintain individual control over who else sees anything about us....The answer will have to be social as well as technical."
Corporations as MUDs: The Virtual Water Cooler
When computers disappear into the woodwork, people in organizations will do the same important thing people in organizations have always done: They will tell each other stories. While workstations can amplify the work people do in their physical cubicles and formal job roles, boxes on desks can't approach what happens at the water cooler. One PARC researcher, Pavel Curtis, is looking closely at MUDs, the water coolers of the Internet. He sees them as a way of bringing informal, even playful communication back into organizations. In MUDs, the "Multi-User Dungeons" of the Internet, thousands of enthusiasts create their own dramatic adventures, sometimes vying for points in fantasy games, sometimes just conversing.
Curtis had been working on a new programming language when he became interested in MUDs, precisely because the newest generations of MUDs were not just fantasy worlds, but tool kits that the players used to create programs to enhance their fantasy-narrative games. MUDs were evolving into group-programming languages.
Why not make a MUD programming tool kit that was also an example of another honored PARC tradition, an object-oriented language? From such a language comes MOO, or "MUD, Object-Oriented." Curtis built on the work of Steven White, a student at the University of Waterloo (Canada). In January 1991, he opened LambdaMOO. Hundreds of players flocked to it.
In his office, beyond the screen-pad-tab-strewn atrium outside Weiser's office, Curtis describes being "drawn back into serious MOO research. In 1992, I dropped the other research and started studying MUDs." MUDs as "social virtual realities" had become his personal passion. That passion grew into the Social VR project, in collaboration with Dave Nichols.
PARC began taking MOOs seriously. Weiser actively encouraged MOO research at the computer science lab as a natural adjunct to Ubicomp - both are ways of exploring the informal dimensions of computer-aided communication. Curtis and Nichols simultaneously looked for useful applications - a MOO for astronomers to meet via the Net and share data, including graphics - and started thinking about bigger issues, like multimedia interfaces and whether the medium can be scaled up for millions of participants. "What else is a MOO good for, besides playing games? What are the nonrecreational uses? Could a virtual office be a useful tool for effective telecommuting?" were the questions they started pursuing.
PARC researchers had been experimenting with multimedia conferencing, including video and audio, in Palo Alto and at a sister organization, EuroParc, in Cambridge, England. What would be required to add video to a MOO?
Curtis envisioned Project Jupiter, a multimedia MOO, to include a suite of tools for creating collaborative environments quickly. Users would modify their virtual collaboration space, creating a unique arena for each project.
We sat in front of Curtis's color screen, looking at surprisingly high-quality video as part of his computer display. He showed me how a multimedia MOO would work. Every space within the virtual world would be a place where people could display text, bulletin-board style. But people who were present in the same virtual room, no matter where they were in the physical world, could choose to be in audio and video contact. A researcher in one part of the building, or one part of the world, might encounter another researcher in one of the MOO's "rooms." A simple command would make it possible for those researchers to see each other via video and talk with each other via audio connections built into their computer workstations.
A collaboration space that mutated from role-playing games is a long way from copying machines, or even the electronic office, but it is easy to see Curtis's work as part of a whole strategic vision of tapping people's everyday innovations.
Management by Storytelling
John Seely Brown, Xerox VP, director of PARC, and "JSB" to everyone, came from the cognitive sciences, not the hardware, software, or telecommunications world. He's setting PARC's sights on a bigger target than electronic offices or even information architectures - seeking to reconfigure the modern corporation. As he wrote in the Harvard Business Review in 1991, "The most important invention that will come out of the corporate research lab in the future will be the corporation itself."
JSB, at PARC since 1978, is informal, lanky, bearded, and possessed of a Mephistophelean grin. He's both loose and intense, and came at me from some unexpected directions. His work makes clear that PARC doctrine today is grounded in an intellectual foundation combining aspects of biology, anthropology, organizational and literary theory, along with the ideas of traditional physics, engineering, software and systems, and cognitive science. He envisions a new, dynamic ecology of communications - rather than a static architecture of information.
True to his precepts of informal communication, JSB met with me at a restaurant. Over spinach salad and a view of Palo Alto's University Avenue, we plunged directly into the future of Xerox.
"Co-evolution" and "narrative" were two words JSB emphasized when asked to describe the new organizations he envisions. In 1992, he wrote: "In the past, our emphasis has been on the individual, and most of our tools were aimed at enhancing the effectiveness of the individual mind. In the future, the action will be in leveraging the organizational mind." When we talked about PARC's future, he began by speaking about webs of human relationships, about the organizational mythologies, legends, and fables that bind groups across formal boundaries.
"Companies no longer assemble products so much as they make meaning out of the world," he said.
The narrative device in the latest PARC scenario is the technique known to screenwriters as "the reversal." Anthropologist Lucy Suchman started studying Xerox accounting clerks in 1979. When she interviewed the accounting clerks, the answers they gave her about how they performed their work corresponded to the procedures described formally in their job descriptions. When she applied ethnographic observation, Suchman discovered that the clerks relied far more, as JSB described it in the Harvard Business Review, on a "rich variety of informal practices that weren't in any manual but turned out to be crucial to getting the work done.... Without being aware of it, they were far more innovative and creative than anybody who heard them describe their 'routine' jobs ever would have thought."
Suchman's work reveals that even people who domundane jobs are constantly innovating. But the institutionalized values of almost all organizations force people to hide their innovation. Encouraging the informal practices that people develop within organizations humanizes organizations, and it also taps the hidden power of everyday innovation.
JSB foresees PARC in the center of a revolution, not just in information products but in technology design and production. Researchers and customers, companies and citizens, will coproduce technologies.
Now that we know how to build tiny machines that can amplify our thinking, let's dissolve them into the countertop, the desktop, the hallway, says Mark Weiser. Let's find ways to facilitate informal, imaginative communication among coworkers by harnessing the narrative form that can cause MUD addiction, says Pavel Curtis. As the technology disappears, let's fill the space it leaves with storytelling, says JSB. Personal computers and computer-mediated communication offer ways for people to think about ideas and procedures, and to communicate with one another.
The innovations that emerge from that computer-aided thinking are more valuable to organizations than the hardware that helped individuals come up with them. The relationships between people in organizations are more important than the hardware and software that makes it easier for them to communicate. Using computer-based tools in an atmosphere that fosters storytelling, improvisation, and informal communication is a way of harnessing silicon to amplify the powers of minds and working communities. "A new transparent knowledge medium that supports communities of learning and workers - a medium which helps us help each other," as JSB puts it.
The importance of storytelling surprised the social scientists on the PARC team. Another anthropological study examined service technicians in training at a large corporation and out in the field with customers. The reps in the study relied on a rich repertoire of informal tricks absent from the procedures manual. In formal training, reps are taught to trouble-shoot in an efficient, algorithmic manner. In the field, reps construct a story that makes sense of the diagnostic data - they spin a narrative that explains the symptoms. While they run the machine through all the standard diagnostic tests, they tell each other stories. From the stories, solutions emerge.
Reps and local operators - and the networks of people they each bring to the dialogue - use stories to communicate about solutions to their problems: stories of weird error messages of the past; stories about buddies of theirs who had debugged a problem like this one a month ago. Not only did the front-line workers in organizations innovate to solve their own problems and save their own companies from their own procedure-manual illusions, they created an invisible, informal, potentially immortal knowledge-base of solutions in the form of stories.
I asked JSB about his belief that tomorrow's successful companies will be "knowledge refineries." JSB talked about the many ways we all improvise every day to enable us to work around the limits of space, time, and technology. He wants to use PARC to tap that as yet unexplored innovation power. JSB is convinced that information technology can be used to help people achieve more of their innate potential. But simply implementing new hardware and software systems isn't enough. "You have to change the way people think about working, learning, and innovating," JSB cautions.
If the ephemeral understanding of how work practices change with new technologies becomes Xerox's "product," rather than boxes or bits, where is the company's competitive advantage? JSB believes that if they are right, Xerox will redefine itself radically. "To the root," JSB reminds me, is the original meaning of "radical." PARC's success in the 1970s has indirectly bred questions in the 1990s - about social control of technology, risk assessment, ethical guidelines for using technologies affecting privacy. Dwelling on the big, soft questions - the messy waste products, the effects on our psyches, communities, and biosphere - may or may not spark a spectacular marketplace win, but there's a palpable sense at PARC that it's the right thing to do.
As they look at cultural phenomena, PARC scientists and engineers have not abandoned their commitment to integrated research into fundamental principles and components. Xerox was instrumental in establishing the United States Display Consortium, a cooperative venture to recapture from Japan the technological lead in flat-panel video display technologies. A new active-matrix liquid crystal display has led to a prototype of the highest-resolution screen in the world - at PARC. Malcolm Thompson's new color display screen, with 6.3 million pixels, makes possible an extremely high level of visual representation - as good as or better than paper.
So how will PARC guarantee that this time they won't fumble their new future? Three ways, says JSB. "One, we are more careful about intellectual property. Two, we are working smart - looking for entrepreneurial partnerships to develop ideas quickly. And three, Xerox has radically repositioned its organization so that its corporate strategy is shaped and informed by PARC and PARC is being shaped and informed by corporate strategy."
JSB serves as chief scientist for all of Xerox, relaying PARC's inventions and potential to the rest of the company. "In the old days we threw inventions over the transom to the business divisions. Now it's more like co-evolution."
Boxes and bits - information-processing and communications hardware and software - will still go out the door, as Xerox continues to convert PARC inventions into products. But some ideas might end up as PARC's most influential innovations. If JSB is right, PARC might end up swallowing Xerox, in a new paradigm of ontological management processes that is only dimly imaginable today. Will our organizations end up as rigid tyrannies, with "human factors" designed to conform to technologies? Or will we coproduce more humane organizations, more innovative and more fulfilled humans? It's time for communication technologies to be designed with those questions in mind.
