Scenario: How it often is.

Dennis notices in the Tuesday, (Aug. 18, 1998) Wall Street Journal that Toys R Us net tumbled 63% largely due to a lack of new hits. It occurs to him that they may be a prime candidate for knowledge management consulting aimed at increasing innovation. He wonders what the current relationship, if any, is between IBM and Toys R Us. After wandering around www.toysrus.com and www.ibm.com for about 40 minutes, he doesn't find anything really relevant. He tries to recall if he knows anyone who ever worked at Toys R Us but draws a blank. He considers checking a consulting data bank, but isn't sure where to look.

It isn't so much a particular competency that he's interested in as in getting a sense of what Toys R Us is really like; how to approach them; what works and what doesn't work with this particular company or with Toy companies in general. Dennis then recalls that he is supposed to be summarizing his last engagement with 3M. He grimaces, realizing that he has only been 30% utilized this month, but begins the onerous task of entering information into various fields. After an hour, he looks back over his work and has the distinct feeling that somehow, he hasn't really captured what was really important about the engagement. He shrugs, wondering whether anyone else will ever really read this anyway.

Scenario: How it could be.

Because Dennis has been accessing a lot of stories having to do with the role of knowledge management in innovation, a relevant article from the Wall Street Journal about Toys R Us is downloaded to his Thinkpad. Based on Dennis's profile as well as a narrative categorization scheme, a number of potentially relevant stories from a number of sources are cached in his local proxy for fast download when he arrives at work. Some stories are presented in full; longer ones are presented as automatic summaries, though Dennis has the option of reading the full story if it's of sufficient interest. The stories are presented with major words important to his purpose highlighted.

He quickly scans through the stories and finds two of particular interest. In the first, he reads the story of the attempts of the IBM North American Marketing Director for Special Needs to work with Toys R Us. He clicks on a hot link in the story and dials the phone number of the writer. In a conversation, he finds out more about working with Toys R Us; what approach is likely to work and what isn't. Dennis begins to create a plan for a contacting Toys R Us. He also discovers that Lotus Institute is looking for Beta sites for a new tool that aids creativity so he considers the possibility of having the R&D part of Toys R Us become a beta site. He finds three relevant success stories wherein IBM Consulting worked with customers to increase innovation and uses them to help sell his plan.

Dennis does not find himself distracted from these tasks by 3M because he documented all the interesting information about that engagement as he went along by whatever means were convenient at the time. This included handwritten notes, tape recorded stories, and stories typed into Word. He doesn't mind getting the occasional call from a colleague about his story because he knows the favor will be returned. In fact, he's rather proud of his stories about 3M.

Theoretical Considerations.

For thousands of generations, people apparently lived in fairly small communities; spoke within each community in a common language; and shared a common background. They danced and told stories. Communication, community, and knowledge were closely related. The art, artifacts and architecture of the community were integrated. Learning was integrated with life. The lives of individuals were seen as and accepted as part of the community. The community generally saw itself as part of the larger ecology as evidenced by rituals, myths, and common practices.

For the last few generations, more and more people have come to live by a quite different pattern. In this new pattern, information is decontextualized; community and trust have largely broken down; teaching is done by specialists in a separate place. People are segregated by age. Much communication is long distance and low bandwidth (compared with face to face). Associated with these trends, a rationalistic, reductionistic, analytic way of approaching problems has become so pervasive that we scarcely still recognize it as an approach.

Since technological artifacts as limited in scope as the stirrup, the mill, and the plow broadly affected social structure (White, 1962), it is no surprise that the shift in the media of thought and communication; e.g., from an oral culture to a written one, would have even more profound effects. In Orality and Literacy, Ong (1997) points out that storytelling in an oral culture is always contextualized. Perhaps the appearance of writing is causally connected with the appearance of what Daniel Quinn (1997) calls the "Taker" culture. Only cultures with written language appear to believe that "their" truth is "the" truth which everyone must subscribe to. In prehistoric times, tribes did fight each other, but it seems difficult to imagine that a specific tribe with a purely oral culture would formulate as a goal "taking over" another tribe and forcing them to adopt a new language and a new set of stories and practices. The development of the written symbol has enabled people to view words as real things. If we write down our myths, they become "objectively" true; not just something that our community finds it useful and comforting to believe, but something that "is" true. This true knowledge can then be read and acquired by others; unfortunately, this has often been translated into "must be acquired by others."

Of course, at some level, we realize personally that we cannot simply read books about golf, singing, art, law, poetry, religion, or psychotherapy and instantly "become" a skilled practitioner. These skills require actually performing these skills or rituals, usually in the actual presence of a mentor or community that is already skilled and can provide informative feedback as well as emotional support. Yet, the reification of written information is such a strong illusion that many organizations collectively act as though decontextualized information is sufficient input for intelligent action.

In its most absurd form, some people believe that human experience and expertise can be completely captured in terms of data and that these data can be transmitted anywhere in the world to a complete novice so that a novice can perform just like an expert. Or, in another version, a huge organization can capture all relevant information at the periphery; this information can be transmitted to some central point; globally optimized decisions can be made centrally; and the orders to implement these decisions can be transmitted back out to the periphery. Such delusions would seem to be the logical conclusions of a long-term trend toward a more and more completely decontextualized and "objective" view of knowledge. Shenk (1997) argues that it is primarily the sheer amount of information that we are subjected to that is the root cause of many of the "ills" of society.

Is nothing to be done then? Is the human race indeed haplessly travelling down a self-destructive path that began with written language and has accelerated with the ascendency of the analytical thinking and electronic media?

More recently, a number of thinkers in widely varying disciplines have begun to re-examine some of the often implicit assumptions of analytic methods and find them lacking; to see instead that knowledge exists as a web of relationships; that a possibility now exists for reintegrating society; for reintegrating learning and life; for reestablishing older methods of communicating such as storytelling; for fostering the collaborative creation of knowledge by communities. Such views do not reject technology or new media, but seek the possibility of using new media in such a way as to enhance and perhaps even regain some human values (Murray, 1997; Friedman, 1997).

In architecture, for example, Christopher Alexander and his colleagues (1977, 1979) promote an evolutionary, user-centered design process which takes context and ecology into account rather than a separatist and egotistical approach. The object-oriented programming community has used pattern languages as a way of sharing information within a community of expertise. (See http://www.truthtable.com/patterns.html for additional pointers.) People in business have begun to use stories as ways to make design points (http://www.pliant.personal/Tom_Erickson/Storytelling.html) and ways to discover the deeper needs of customers (Lieber, 1997). In business, people are discovering the importance of social practices in knowledge creation (Collins and Porras, 1994) as well as the value of cooperation in fostering the ecology of an industry (Moore, 1996). Rather than focus purely on the competition of ideas to determine which individual idea is correct, there is a move to use a more collaborative conversational method called Dialogue to create knowledge from a group that goes beyond what any one individual can conceive (Bohm, 1990, 1994; Isaacs, 1996). In particular, Dialogue can be useful when each person sees only a part of a complex system. Indeed, both in business (Senge, 1990) and in ecological matters (Robert, 1995), people are beginning to see the importance of taking a systems view and see organizations as more akin to living things than to mechanisms (Thomas, 1995; DeGuess, 1997). (See http://www.truthtable.com/orglearn.html for further references).

The reductionistic and isolationist view of information (as opposed to a systems view) leads naturally to poor HCI design. In such a view, we must first we find out what a series of tasks are supposed to accomplish. Then, we design a computer system for maximum efficiency. Finally, provided that the budget and schedule allow, we build a "User Interface" so that users may now interact with the computer system, enter necessary data, and see the results. While this description seems like a ridiculous parody, unfortunately, it is often not far from the true state of affairs in designing new systems. The consequences of designing a human-computer system in such a piecemeal fashion are predictably little real productivity growth (Landauer, 1996). Perhaps worse, over the longer run, such systems probably contribute to the breakdown of trust; the decay of social capital. The "efficiency expert" who takes such an approach sees no need to support "chit-chat" and other "nonproductive" social activities that may be vital to building social capital (cf. Putnam, et. als. 1993).

Ironically, it is precisely the social fabric -- the feeling of a community that they can work things out together and trust one another -- that will support an organization's ability to survive technology spawned difficulties such as the Y2K problem. If people within and between organizations see the Y2K problem as an opportunity to "cheat" consumers, banks, their company, the IRS -- in short, everyone else for their own short-term gain -- the social and economic magnitude of the problem will be much greater than if people pull together to work around and solve the common problems caused by software failure.

When people constantly interact with systems that isolate them, there are probably still further consequences. People no longer see themselves as belonging to a community and no longer see themselves as part of a larger life. In such circumstances, their own individual body takes on tremendous relative importance. People want to live forever. Individual physical immortality becomes viewed as a necessary goal. In the service of that goal, no price is too high to pay in terms of ecological and social damage, because those are defined as "other" and only the individual physical self is defined as "self." Thus, the isolationist view of information and life play into a positive feedback loop. As people become more isolated, they view their own existence as more important, become more mindlessly competitive both for direct physical assets and for mindshare, and are less capable of taking other viewpoints or acting cooperatively. Ironically, again, this is not optimal even for their own material gain (Axelrod, 1984).

If instead, systems with human and technological components are designed as true systems that take into account the actual nature of human beings as social animals as well as information processing components (e.g., Beyer and Holtzblatt, 1998), not only is narrowly defined productivity much more likely to increase, but the formation of social capital will be increased as well. This will allow people the trust necessary to "work around" problems that might occur when the technology fails. In addition, such systems will tend to support the ability of such groups to solve system level problems through Dialogue and other group processes that are too complex to be solved by individuals. In a broader sense, people who work with systems that foster community will also feel less isolated; turn less to violence and greed to satisfy their own personal needs because they will see themselves as part of a larger community life. This need to design and implement systems with the real nature of human beings in mind is not limited to computer systems of course, but applies to architecture, physical artifacts, procedures, and governments as well.

"Good" HCI design practice then, is viewed here not simply as a more practical way to improve productivity on a specific job. It is conceived of as part of larger movement to use technology to foster a more community-based, more contextualized, more systems-oriented view of human knowledge. The consequences include greater chances for improved productivity in the small, but also, in the large, the consequences may include a move toward greater trust and cooperation; less feeling of isolation; more feeling of connectedness; hence, ultimately, more ecologically sound behavior.

Project Description.

I am embarking on a project at IBM Research in which a small group will be attempting to facilitate the capture, creation, organization, indexing, searching, and utilization of stories. We will do this by working with our users in their real context to determine what their needs are with respect to stories. Stories, we feel, are a natural way to simultaneously foster a sense of community and to share essential knowledge.

In a related effort, we will be attempting to foster the collaborative creation of knowledge by supporting pattern languages. We believe that tools can be build to help "discover" patterns; to mine them; to help keep track of pros and cons; to find patterns appropriate to a given situation. In both projects, we will attempt to promote connections between the real identities of people and their contributions to help sustain reciprocity and repute over time. We believe that electronic media, per se, and having more information available are not in and of themselves bad things; rather, care must be taken to preserve the personal and contextual nature of knowledge in order for it to have a humanizing as opposed to dehumanizing effect.

References Cited

Alexander, C. Ishikawa, S., Silverstein, M., Jacobson, M., Fiksdahl-King, I. and Angel, S. A Pattern Language. New York: Oxford University Press, 1977.
Alexander, C. The Timeless Way of Building. New York: Oxford University Press, 1979.
Axelrod, R. The Evolution of Cooperation. New York: Harper/Collins, 1984.
Beyer, H. and Holtzblatt, K. Contextual Design. San Fransisco: Morgan Kaufman, 1998.
Bohm, D. Thought as a System. London: Routledge, 1994.
Bohm, D. On Dialogue. Ojai: David Bohm Seminars, 1990.
Collins, J. C. and Porras, J. I., Built to Last. New York: Harper, 1994.
DeGuess, A. The Living Company. Boston: Harvard Business School Press, 1997.
Friedman, B. Human Values and the Design of Computer Technology. Stanford: CLSI, 1997.
Isaacs, W. N. The process and potential of dialogue in social change. Educational Technology, Jan.Feb., 1996, 20-30.
Lieber, R. B. Storytelling: A new way to get closer to your customer. FORTUNE, Feb. 3, 1997.
Moore, J. F. The Death of Competition. New York: Harper Collins, 1996.
Murray, J. H. Hamlet on the Holodeck: The Future of Narrative in Cyberspace. New York: The Free Press, 1997.
Ong, W. J. Orality and Literacy: The Technologizing of the Word. London: Routledge, 1997.
Peppers, D. and Rogers, M. Enterprise, One to One: Tools for Competing in the Interactive Age. New York: Doubleday, 1997.
Putnam, R.D., Leonardi, R., Nanetti, R. Making Democracy Work: Civic Traditions in Modern Italy. Princeton, NJ: Princeton University Press, 1993.
Quinn, D. The story of B. New York: Bantam, 1997.
Robert, K. The Natural Step: a Framework for Large-scale Change. Systems thinking in action conference, Boston, 1995.
Schank, R. C. Tell Me a Story: Narrative and Intelligence. Evanston: Northwestern University, 1990.
Shenk, D. Data Smog: Surviving the Information Glut. San Francisco: Harper, 1997.
Senge, P. The Fifth Discipline. New York: Doubleday, 1990.
Thomas, J. C. The long term social implications of new information technology. In New Infotainment Technologies in the Home: Demand Side Perspectives, R. Dholakia, N. Mundorf, and N. Dholakia (Eds.), Hillsdale, NJ: Erlbaum, 1996.
Thomas, J. Biological metaphors for organizational learning. Presented at joint University of Colorado, University of Michigan, IRL, NYNEX symposium on organizational learning, White Plains, NY, 1995. (available at http://www.truthtable.com/bio2.html)
Thomas, J. and Kellogg, W. Minimizing ecological gaps in interface design. IEEE Software, Jan. 1989, 78-86.
Van Der Heijden, K. Scenarios: the Art of Strategic Conversation. New York: Wiley, 1996.
White, L., Jr. Medieval Technology and Social Change. Oxford: Oxford University Press, 1962.