Copyright 1998 CAUSE. From CAUSE/EFFECT Volume 21, Number 1, 1998, pp. 12-17. Permission to copy or disseminate all or part of this material is granted provided that the copies are not made or distributed for commercial advantage, the CAUSE copyright and its date appear, and notice is given that copying is by permission of CAUSE, the association for managing and using information resources in higher education. To disseminate otherwise, or to republish, requires written permission. For further information, contact Jim Roche at CAUSE, 4840 Pearl East Circle, Suite 302E, Boulder, CO 80301 USA; 303-939-0308; e-mail: [email protected]

Managing Knowledge:
An Interview with Thomas Davenport

Thomas Davenport, Curtis Mathes Fellowship professor and director of the Information Systems Management Program at the University of Texas, Austin, spoke to CAUSE97 attendees about "New Roles for Information Professionals." After his presentation, he talked with CAUSE97 conference chair Gerald Bernbom about the topic of his latest book, Information Ecology.

Bernbom: In your book, Information Ecology, you provide many examples of corporations that have begun to take an ecological approach to information management. Are the concepts you discuss also applicable to colleges and universities, and how might they be applied?

Davenport: Administratively, there are a lot of similarities. I don't think there are any significant problems in applying the ideas of information culture, politics, and staff to universities. It gets tougher in academic computing because it is such a knowledge-oriented environment, but the issues are relevant. I think the need to apply some of the ideas is very strong, particularly the ideas on information engagement. There's a real need for focusing on how people use the information we transmit in academic environments, but because pedagogy and teaching and learning are worlds unto themselves, I think the information ecology concept is less applicable there than in the area of administrative information systems. Thinking seriously about it, there are probably more attempts to make administrative information environments better than where information, knowledge, and learning are the heart of the matter.

Bernbom: Do you see any synergy for people who do structural design, people whose real work is not pedagogy and content, but design of delivery?

Davenport: I probably should have included them by saying that it takes a village to make a good information environment. There's a lot of good thinking in communications schools, and in library science programs, about these issues.

Bernbom: Higher education is in the knowledge business. Could you comment on how topics like information strategy or information architecture apply to the scholarly record?

Davenport: I'm very interested in that issue. While I haven't done a lot of formal research on it, I do a lot of day-to-day observation. Somebody asked me last week, since I write about knowledge management, what I have done at the University of Texas to improve the management of knowledge. I said that's too big a task for me to even think about! It's much easier for me to work with external organizations than to focus internally.

In terms of knowledge or information strategy, part of that is focusing on what knowledge is really important. Universities have a tough time doing that because all knowledge is supposed to be equally important. Increasingly, as physical environments get tougher, people will have to make more decisions about what kind of knowledge we want to generate and transmit and inculcate in people.

The other part of information strategy is deciding what aspects of the information process are really most critical to you. Clearly, in universities we've decided knowledge creation is the thing we care about. We care much less about knowledge distribution, dissemination, use. We produce a vast amount of knowledge that never gets used by anyone but other academics. The average research article never gets cited by anybody. We've proliferated all these channels, but nobody seems to pay attention to them. There's a real need for change in that respect. But the tradition of the culture is so weighted against making any change. The inmates run the asylum in these places; the professors have all the power. Deans and administrators realize that there�s a big problem. They go out to raise money and people say, "After I've graduated, the place is irrelevant to me."

Bernbom: Do you work with anyone in the information business where you see anything going on that could be applicable or parallel to higher education?

Davenport: The biggest area is clearly in consulting and professional services. Most people define consulting as a knowledge business. What they're doing is to help their clients to overcome problems using knowledge and to apply knowledge to business situations. Consulting firms have been extremely diligent in managing their knowledge. It's clearly the leading industry. Most of the large firms have huge repositories of knowledge. For example, Ernst & Young has done a really good job of capturing what they know. They've been more focused on stocking the shelves with knowledge and less focused on motivating people to take things off the shelves and actually use it. They are a role model the university could look to. Andersen Consulting has been extremely aggressive in using technology for pedagogical purposes. The best educational technology I've ever seen comes out of work Andersen has done with Roger Schank, a cognitive scientist at Northwestern University's Institute for Learning Sciences,1 who has a model of scenario-based educational simulations on CD-ROM.

Bernbom: You indicate in your book that good business decisions are based on fact rather than intuition or rumor. What do you see in higher education decision-making vis-à-vis this issue?

Davenport: As my editor kept pointing out to me, one person's facts are another person's opinions. You can get into difficult issues in this area. In general we don't pay enough attention to fact in American organizations, and focus too much on other sources of decisions. For higher education, there's good news and bad news. The bad news is that the things that really matter in universities are hard to measure -- learning, for example. When you come to the things that matter most, it's difficult to agree on what facts really count. The good news is that in state schools -- this is largely due to semi-ignorant interference by legislators -- universities are starting to have to document what they do, how many hours people spend in a classroom. To try to address the learning-oriented issue at Texas, we have this thing called the performance-based instructional system. If you're a faculty member and you want to go to your department chairman to argue for more resources, the numbers are right there as to how many students you have, what classes you teach, how that compares to other places, what percentage of courses are taught by tenure-track faculty. Because of that external influence, there has been a much more fact-based orientation. I've been away from private schools for a while, but I think it's going to spread there as well.

Bernbom: The corporate world has focused a lot on business process reengineering in recent years. As a university faculty member, what do you think colleges and universities need to do to be successful in reengineering their business processes?

Davenport: Unfortunately, I've seen most of the problems of reengineering in the corporate sector be replicated in universities. I've heard someone at my school say, "Oh, yeah, we've reengineered our curriculum." Of course, it didn't change very much! We have adopted the term for our own purposes when it was fashionable to signify radical change whether radical change was really happening or not. The other problematic thing is that the reengineering I've seen in universities -- some of it has even been successful! -- has almost exclusively been in administrative areas. It's the easier stuff to reengineer, but what I would have liked to have seen is for somebody to say, "We're going to reengineer the educational process." That is at the core of what we do.

I've worked with a couple of institutions on one important area, fund-raising, with some success. Reengineering in general had a tough time living up to the hype around it. People made improvements, but what I've learned in both my research and in observing companies is that in reengineering, design of the new ways of doing a particular process is quite revolutionary, but implementation is almost always evolutionary. It takes a lot of patience and time. In universities, you have a lot of smart people. The problem is, when you have a lot of smart people, they want to participate in the design of their own work. (You should do that more, even when people aren't so smart!) Particularly, in universities reengineering projects tend to be quite large. Some of them get unwieldy in terms of how you design a process when you've got forty-two people participating in it. But it's a worthwhile effort. The need for changing how we do our work has not gone away, but the term reengineering is no longer a fashionable way to describe it.

Bernbom: As you've indicated, administrative processes in higher education are somewhat marginal to the core business of teaching and learning. How do you assess the value of investing in processes that are essentially not core?

Davenport: I think they're good things to do. One of the reasons why we focus on them is because we know the other areas are more valuable but so laden with difficulty and peril that we don't want to venture into them. If it's a foregone conclusion that you're not going to be able to touch the core stuff, then you might as well do the more peripheral stuff. Focusing on improving financial processes and financial information is a good thing to do. Institutions have to stay in business and be sound financially. But if it distracts you from looking at the teaching, learning, and knowledge creation processes, then it's dangerous.

Bernbom: In the examples you said were somewhat successful, like fund-raising, were changes in information practices an important part of the reengineering, or were there other things that made it successful?

Davenport: Almost every major reengineering project has some information components. We focus too much on technology and not enough on the information people need. Part of the fund-raising efforts involved circulating information. For example, a company comes to Texas to talk to me about my research. Typically, I have no clue whether we're trying to hit them up for money or whether they've given us money in the past or not. One university I worked with started to enlist faculty much more in the fund-raising process, giving everybody an account-team kind of structure. Everybody on the team, including faculty and even some students, had access to much more information about who had given what in the past and what the goals were for this organization. Clearly it's very useful to know the history if you're going to get people to give you money.

Bernbom: What about information politics in the campus setting? Is it as much of a challenge as in the corporate world?

Davenport: The politics are quite fierce! That's one of the reasons why I focus my efforts more externally than internally. The thing about universities is that we expect that because we're interested in knowledge and learning, we'd be above that sort of thing. Clearly, we're not. I don't know if the politics are any worse, but they may be more overlooked by people who think that because they are in an academic environment, they can leave all those corporate politics behind. The need to pay attention to information politics is quite strong; the political structures in universities are the most difficult to deal with. In most universities, you have some form of information anarchy. Anarchy isn't all bad because it shows that there's widespread interest in information. If you're going to get people together, you've got to enlist a pretty broad stakeholder community, much more so than in most corporations. I don't know if the politics are more severe, but they're much more difficult to manage.

Bernbom: Your book makes the point that the techno-utopian approach is outdated and inadequate, that the focus needs to shift from technology to information and knowledge, and that what we need are chief knowledge officers. What about the people in higher education who carry the title chief information officer?

Davenport: They are no more chief information officers than their counterparts in business! They are largely chief technology officers.

Bernbom: Do you think this role of chief knowledge officer has applicability in higher education?

Davenport: I think there's a real need for it. We manage our knowledge pretty poorly in universities. It's a tricky thing. In universities, almost everybody feels like they're involved in knowledge. It's difficult for a university to say, here's the knowledge czar. Even if I were creating that kind of role, I don't know if I'd call it chief knowledge officer. There's a quite distinguished professor from Japan who was just named the Xerox Distinguished Professor of Knowledge at Berkeley. There was an article in the New York Times about this. They interviewed several other professors at Berkeley who said, "First, there's no knowledge in a business school anyway. Second, we're all professors of knowledge. Why does this guy get the title?" The usual academic jealously and resentment popped up. That would probably be the case if you established a chief knowledge officer. People are more used to the chief information officer idea. I don't know whether chief content officer would work or not. It's a little less presumptuous than knowledge, as a term.

Bernbom: You talk about the different kinds of information support workers who are going to be needed in order to pursue this information ecology approach. Where will they come from?

Davenport: There is some progress in that regard. Probably the best example is in library schools where they're starting to graduate generic information professionals who can operate across information technology environments as well as book- and document-oriented environments. The key thing is that we get together on this, that we not assume that any one group has the territory staked out. The other thing that worries me is that we'll pass each other like ships in the night.

Bernbom: What kind of changes could you envision in an undergraduate curriculum to help develop the skills for the kind of professionals that are going to be needed?

Davemport: At a basic level, we need to raise the visibility of knowledge management. You could argue that in universities, the core of what we do is teach people how to manage information on a personal level. But it never gets addressed in any kind of explicit way. I had one professor in all of my twenty years of education who once took up the issue of when you walk into a library, is it better to go to the card catalog first or to the reference librarian first? He had this vociferous belief that it was better to go to the card catalog first. My sense is that I'd rather take advantage of a human, but we don't address those issues. We don't tell people how to acquire information. Our libraries may offer some Lexis-Nexis searching course every other Wednesday, but it's very peripheral to what we do as institutions.

If we're serious about creating lifelong learners, the skill that matters most of all is teaching people how to find, filter, and act on information. It ought to be the first course that every freshman takes. We need to start to develop schools that address information issues in holistic ways. Library schools are the earliest adopters of this kind of thing. At my school, when the new prospective dean of the library school came in, she came to the business school and said she'd really like to cooperate with us. But a lot of the business school people didn't understand why the relationship was important.

We need to get to the point where when you go to an information function in an organization, whether it's in a university or a company, the person is familiar with a whole variety of information-oriented approaches. But now you walk into the information technology department and they say, sure, we can build an information system for you. You walk into the market research department and they say, yep, we can do a focus group for you. But nobody can see the entire picture. It's very problematic.

Bernbom: In your book, you propose information mapping as an alternative to information modeling. Would the World Wide Web be the place for information mapping, and how would you go about this exercise?

Davenport: The Web is great -- clearly, a very information-oriented tool. It doesn't totally take technology out of the picture as a barrier, but it makes a lot of progress. I'd like to get to the point where the Web is like television and just as you don't hear any spirited discussions of whether my cable box is made by General Signal or General Instruments, you wouldn't hear any spirited discussions of whether I am using Internet Explorer or Netscape or what kind of processor is best. That's the nice thing about television. Technology has not been much of a barrier.

Where would you start? I've thought of this in regard to my own school. There are a lot of information providers within universities. You probably want to start with a task force of different types of information providers to list the information that you have. It might get overwhelming pretty quickly. You'd have to start making decisions about what information is really most important to you. What you find in a lot of companies these days is too much information egalitarianism. The shuttle bus schedule and the cafeteria menu are at the same level in the repository as the R&D organization. You have to decide what's really important to you. You can put it all online if you want, but you have to have a structure where the important stuff is most visible. Otherwise, it can be a morass. Putting it all in one place may be a good idea, but it requires a lot of architecting and navigational aids and some prioritizing about what really matters the most.

There's a concept in the archival management field, vital records. You start by prioritizing your vital records. Those we know about. Maybe it's what's between the vital records and the cafeteria schedule that is the challenge!

When I was attending an information management meeting in Boston I wandered into a records management meeting. I started looking at the agenda. It looked just great! I thought, I really need to pay more attention to what happens here. The only issue, from a vital records standpoint, is we live in an environment where there are so many information providers. No one group can hope to monopolize it, totally architect it. We'd like to be in a place where everybody in a university would have a personal Web page to say, here's what I'm working on. All we can really hope to do is try to provide people some guidelines, some common format, so that the most important information gets seen first, but it won't be the only information there. You'll see pictures of people's kids and pets, but the easiest stuff to get to should be what they're expert in or what kinds of courses they're teaching. You can't control it all. All you can do is hope to guide people where they need to be guided. The Web is also a good paradigm for that. It's a very democratic information medium.

Bernbom: In closing, what do you think are the most important issues today, with respect to knowledge management, for colleges and universities?

Davenport: I'd say there are three important issues. First, addressing what knowledge management means in a university context. I would encourage CAUSE to really pursue that issue in a hot and heavy way. It's critical to the mission of the university, but we do it pretty poorly now. We don't want to let everybody else take the lead on us. Second, figuring out what to do with information technology and pedagogy -- a very difficult problem. All boats have been lifted, but the level of the sea is still very low in that regard. The third thing to worry about is competition from the corporate sector. I visit consulting firms and corporate universities and get very nervous because they are a lot more progressive than we are in higher education in using technology to support their mission. I'm afraid these corporations are going to jump in and take over our charge as creators and distributors of knowledge and educators of human beings. It would be awfully sad if the higher education institution died out.

Sidebar

Information Ecology: Mastering the Information and Knowledge Environment
by Thomas H. Davenport
Oxford University Press, 1997, hardcover, 255 pages, $29,95

Reviewed by Julia Rudy

As competition in the higher education marketplace heats up, the need for an institution-wide approach to information management will become increasingly key to success and survival. And while investment in technology and return on that investment will continue to be major concerns for campus administrators for the foreseeable future, it is important that a focus on technology not overshadow the institution's fundamental, critical resources -- information and people.

In Information Ecology, Thomas Davenport argues that the "status quo approach to information management -- invest in information technologies, period -- just doesn't work." He recommends, instead, a new approach that takes into account an organization's entire information environment. Such an information ecology approach is counter to the traditional "machine engineering approach" that throws technological solutions at information problems. The approach recommended by the author puts people, and how they create, distribute, and use information, at the center of information management. It also recognizes that technology is only one part of the equation -- and sometimes the wrong one -- for creating change: "The effective use of information," says Davenport, "much more than any new technology, can change how an organization runs."

Key chapters of the book emphasize the importance of developing an overall strategy for information use; address information politics, behavior, and culture; explore important information staff issues; describe key steps of a common information management; and present alternative approaches to information structuring and modeling. Each of these chapters includes a helpful "assessment survey" for readers to see how their organizations measure up.

While the examples that Davenport provides to illustrate best practices in "holistic management of information" are largely from the corporate world, the concepts he explores and the processes he recommends are also applicable in academe. I particularly enjoyed the author's discussion of information staff and his position that non-information-technology information providers such as librarians, management accountants, records managers, analysts (business, market, and/or financial), and individual managers and workers are also critical to the information support structure. What is important is that this combination of information professionals must work together to provide, as much as possible, a single interface to a wide range of information sources for users. Worth the price of the book is Davenport's discussion of new tasks for information staff (such as information pruning, adding context to information, enhancing the style of information, and choosing the right medium for information), as well as new support roles in the "television-type" organization -- information innovators, content editors, content directors, information producers, and a chief content officer. With the emergence of the World Wide Web as the central platform for accessing and delivering multimedia campuswide information systems, the academic community should especially appreciate these concepts.

If you have an interest in information rather than technology management, this book is one you won't want to miss.

Reviewer Julia Rudy ([email protected]) is director of research and development at CAUSE.

Davenport	A excerpt of the actual Davenport interview is available in RealVideo (56k) To take advantage of RealVideo you will need to obtain RealPlayer 5.0 from the RealMedia Website. See the CAUSE file help page for details on configuring your browser to view these files.
Watch the entire General Session speech given by Davenport at CAUSE97 in RealVideo (28.8k)

Gerald Bernbom ([email protected]) is assistant to the vice president for information technology at Indiana University.

...to the table of contents