of the American Society for Information Science and Technology       Vol. 27, No. 3       February/March 2001


Go to
Bulletin Index

bookstore2Go to the ASIST Bookstore



John Seely Brown: Leveraging the Social Life of Information in the E-Age:
Idea Sparkers

by Steve Hardin

"Context must be added to information to produce meaning. If we're to move forward, we must not limit ourselves to merely looking ahead, but we must also learn to look around. Learning occurs when members of a community of practice socially construct their understanding of some event or text and then share their understanding with others." This message was the focus of a keynote address delivered by John Seely Brown, chief scientist of Xerox Corporation and former director of Xerox Palo Alto Research Center (PARC), to nearly 400 persons at the 2000 Annual Meeting of the American Society for Information Science and Technology.

Co-author of The Social Life of Information and author of more than 100 papers, Brown (or JSB as he is mostly called) began by explaining that his presentation was not "a formal talk." Instead, he said he would just "throw out idea sparkers," in the hope that they would be evocative for audience members to consider after they left the session.

JSB started his talk by noting there has been a tendency toward "endism" lately. Various observers have predicted the death of the university, the nation, the city, the firm, distance, the library and the book. On the one hand, we say technology is destroying all these things; on the other hand, we read about mega-mergers while we read that the firm is going away. Maybe there are some fundamental misunderstandings of what made the world work. We need to step back and think more about the social fabric in which we live.

Part of the problem, JSB said, is tunnel vision. As an example, he referred to Bill Gates' 1995 book The Road Ahead. Brown described the book as "simple technological determinism." In one sense, the road ahead is straight. In another sense, though, it's filled with twists and turns. In fact, when you think about it, we never had technological determinism. There are some technologies we fight to keep, he said: "I love the fact that my notebook never needs booting up." The book and the fax we fight to keep. He related the story of a woman who uses fax machines instead of e-mail because she can scribble a personal note at the bottom of a printed letter and fax it; in that sense, the fax lets her reach out and touch the receiver with a bit of her personality in a way that e-mail doesn't. And then, there are technologies that always seem to bite back, noting that he arrived well before he was scheduled to give his talk because he anticipated problems booting up his presentation.

Tunnel vision leads to tunnel design. Tunnel design offers no context, no meaning, nor any sense of being located. If you glue two toilet paper tubes to your eyes, you'll lose your peripheral vision. Now try walking around the world, and you will discover that you will collapse into a "twitching heap." Stuff just pops into your center vision always surprising you and often disorienting you. Peripheral vision does a great job preparing the center for what's coming next. But the world you see as you look at your computer screen is very much like looking through these tubes a kind of tunnel vision. The GUI metaphor worked well at first because the computer screen mirrored the desktop. The lack of peripheral vision wasn't much of a problem. But then that GUI got lifted up to the World Wide Web.

"Now as we point and click we literally leap all over the world no sense of where we were, no sense of where we are now and no sense of where we need to head next. If we can't find a way to feel located by reconstituting a sense of periphery as we search and surf, we will always end up feeling disoriented," he said.

Consider the newspaper: "Why does scanning a website feel so different from scanning a newspaper?" he asked. It's because the newspaper has been designed to take peripheral vision into account. Articles are contextualized by the other articles around them. You can constantly scan the periphery while you read the center. You can discover information that you don't know you want to know. A problem with knowbots and other smart Web search tools is that they are good at finding things we know we want to know, but not at finding things we don't know we want to know.

Much of learning depends on hearing other points of view   something that the periphery (both physical and social) facilitates. We also know that newspaper stories that start above the fold are more important than stories that start below the fold. Stories that start on the front and jump to the back page are more important than stories that jump to the middle. We know all these cues subconsciously. You learn over a long time how to negotiate the newspaper.

So, JSB asked, "Is a personalized newspaper even when it is printed any longer a newspaper?" It certainly delivers information, but does it create news? Does it provide a common experience? It's the fact that you have a common experience that holds the society together while it's flying apart in many other areas. For that matter, he asked, "When does a preprint become a paper?" A preprint conveys information; a published paper has the power of the publishing institution behind it.

JSB said he finds that documents themselves are "curious beasts." It's the combination of text and context, he said, that produces meaning. He related the story of Paul Duguid his co-author who was searching in some European archives and saw someone smelling old letters. Asked what he was doing, the smeller responded that he was a medical historian tracking the spread of cholera in the 18th century. In those days, people put vinegar on their letters in an effort to retard the spread of the disease. He was sniffing for the smell of vinegar to track the spread of cholera, but Paul also a historian suddenly realized that text of the letters he had been studying now had quite a different meaning when the 'context' contained a whiff of vinegar. Indeed, when the writer claimed everything was fine but the whiff was present, perhaps things weren't quite so fine, after all. If you didn't understand the context vinegar and its use how could you interpret the text correctly?

We're getting better, Brown said, at using sophisticated techniques for filtering, searching, sorting, selecting, preference matching. But look at how these things fail, in terms of their ability to read the context. For example, he said he buys a lot of books from Amazon.com. He got a message from Amazon.com one day telling him there was a great book that matched the preferences of others who'd bought books similar to the ones he buys. In fact, it was recommending he buy his own Social Life of Information. The machine was missing the context.

We think of the world, he said, through Cartesian lenses: "I think. Therefore, I am." This attitude has led to theories of pedagogy modeled on pouring a pitcher of knowledge into a student's head. A better set of lenses, Brown said, is provided by, "We participate. Therefore we are." In fact, he said, "In participation with others, we come into being." The sense of how you participate and honor the other has a lot to do with the growth of your personal identity. "Understanding is socially constructed," he said. Things are learned in and through discussion.

For example, JSB said, he used to think that troubleshooting was basically a logical process. But he went on to tell a story about how his own epistemological beliefs got changed when he hired some anthropologists to live, work, learn and play with a community of troubleshooters. He learned that troubleshooting for these folks was a form of narrative construction crafting a story that explained the symptoms of the faulty machine. When a machine fails, they come up with a story explaining the symptoms. If they can't solve the problem, they call in a colleague who adds to the narrative. Eventually, they fill in the blanks with more narrative. The machine is diagnosed when they have constructed a story that covers all the symptoms and tests. After a day of troubleshooting, they sit around the table and share stories itself a process of telling, listening, challenging, refining and so on. They socially construct something that's believable by everyone at the table. That's where real learning knowledge sharing starts to happen. 

Brown discussed creating and sharing knowledge in a community of technical representatives. The tech reps in this group first communicate through a 2-way radio system and then the Web. Tech reps' ideas go through an instant vetting process with their peers. When an idea gets refined to where their peers like it, it finally gets posted to a knowledge base. At that point, it's sufficiently tested that tech reps are willing to act on it. Not only that, but the person who came up with the germ of the idea gets credit for authoring it; the procedure becomes an authoring process. The technique creates a "virtuous circle" connecting the formation of social capital and intellectual capital. Creating social capital within a community of practice helps to shape personal identity as those whose contributions make a big difference soon become more central members of that community of practice.

On the Web, there's a fluid boundary between production and consumption. We need to better understand how this could radicalize both learning and knowledge creation. Consider, for example, the open source (Linux) movement. In this community you are expected to borrow someone else's code, add to it, improve it, extend it and use it for new things, if you want. A clever idea that has been embedded in software that was also meant to be read can spread quickly and can be better understood since the 'reader' encounters it in context a context that can also be executed. And again, there's something interesting here with the deep interplay between the formation of intellectual capital and social capital.

Consider the "tragedy of the commons," in which a public good, a public space, tends to get overused and consumed. We know that once a commons gets to have many people using it, it loses its own self-regulation and quickly gets overused and destroyed. As we move from the physical world to the cyber world, we may be able to move from the tragedy of the commons to the "comedy of the commons." (Brown pointed out that term did not originate with him.) Why comedy? Because in a comedy, the more the merrier, the richer it becomes; or, said in modern terms, we get network effects that underlie an increasing-returns economy. We have a great opportunity here to move from the tragedy to the comedy, JSB stressed. We've learned that just putting information out on the Web doesn't work; the commons must be landscaped for better use. Indeed, JSB wondered if this was not one of the fundamental design challenges for ASIST.

As an intuition pump for thinking about landscaping a virtual commons, he reflected on the architectural studio as an example of a powerful social learning environment. In the studio, all work-in-progress is made public, for interaction, critiquing and learning. It's the exact opposite of the academy, where work-in-progress is generally kept secret, especially in the biosciences. In the architectural studio, when the master comes by to critique the student's work, the other students hear the critique, too.

One important Web attribute is anonymity: "On the Internet, no one knows you're a dog," which creates both opportunities and problems. Brown said he stumbled across a seven-year-old child who in midtown Manhattan had a fascination with penguins. There were very few kids on his block who were interested; neither were his teachers. He surfed the Web and found a college's discussion group on penguins. He'd link and lurk, and occasionally carefully compose a question, and send the discussion into the group and get an answer. But no one knew they were talking to a seven-year old boy. The boy was, in essence, beginning a cognitive apprenticeship.

Brown suggested that we might think of the Web as fostering a learning ecology. There is a vast collection of overlapping virtual communities of interest, cross-pollinating each other, constantly evolving and largely self-organizing. In a learning ecology we find that pedagogy becomes husbandry.

In the digital age, the university will foster a variety of experiences. We should let the virtual augment the physical, loosening the bonds among facilities, faculties, students and credentialing. JSB said the distance learning people are getting it wrong in that that they think most learning happens inside the classroom. But a tremendous amount of learning actually happens outside the classroom (but on the campus) in the informal discussions and bull sessions that are fostered by the shared experience in the class.

"The way forward," JSB concluded, "is paradoxically not to look ahead, but to look around."

Steve Hardin is associate librarian, Indiana State University, Cunningham Memorial Library, 3391 Idaho Street, Terre Haute, IN 47803-3878. He can be reached by telephone at 812/237-7685 or by e-mail at shardin@indstate.edu. John Seely Brown is chief scientist at Xerox Corporation. He can be reached by e-mail at jsb@parc.xerox.com

How to Order

ASIST Home Page

American Society for Information Science and Technology
8555 16th Street, Suite 850, Silver Spring, Maryland 20910, USA
Tel. 301-495-0900, Fax: 301-495-0810 | E-mail:

Copyright 2001, American Society for Information Science and Technology