Dr. Clifford Lynch, Director,Division of Library Automation,
Office of the President, University of California
"The Changing Nature of Collections in the Digital Age"
The third keynote speaker, Clifford Lynch, Director of the Division of Library Automation in the University of California's Office of the President, surveyed trends affecting traditional materials and several new kinds of information resource which are likely to become very important as time goes on. He sketched a taxonomy--traditional print, electronic print, Internet reources, special collections, data flows, and collaboration environments--and described how each type is developing in very different ways in terms of what they provide, how easy they are to "navigate," who will provide them and at what cost, and who will archive and preserve them. He expressed concern over the increasing fragmentation and loss of coherence, which will make it more difficult for scholars and students using information resources and for those striving to define the roles of libraries.
Libraries have moved very slowly to deal with new media. We typically still spend five percent or less on electronic resources. As some libraries move toward "access" rather than "ownership," libraries may become divided into suppliers and consumers. --and Traditional interlibrary lending arrangements will give way to contractual arrangements which compensate supplying libraries. Commercial vendors will also play an increasingly important role as suppliers. Since monographs don't lend themselves as readily as articles and other forms of publication to electronic distribution, they are increasingly threatened as a viable publication format. Finally, as electronic access becomes dominant, abstracting and indexing databases assume far greater importance than ever before, since they now "define" the information universe for most users. A print journal that is not indexed is nearly invisible. We will have to pay careful attention to how well the A & I databases are covering what scholarship deems desirable.
The legal and economic framework that has emerged is more radically different from that of traditional print than most would have expected. But it is now fairly well defined: license, rather than purchase, with negotiable terms and conditions, and very high costs. The high costs are pushing buyers and sellers to deal with aggregations of resources--selling journals as a large collection, for example, rather than title by title. On the other hand, there is a disconcerting trend toward direct sales from publisher to end-user. This could undercut libraries' role in selecting and organizing materials from all publishers. Though libraries clearly have archival responsibility for print materials, no one has effective archival responsibility for electronic print, and much could be lost before our society determines who is responsible and gives them the necessary legal rights and economic support.
The Internet is a communication channel that facilitates publication and information dissemination. It is not a library, since it is not an organized collection, but Internet resources can be incorporated into libraries, or be used to complement and extend libraries. The main reason so many users are beginning to bypass traditional libraries for Internet resources seems to be the compelling appeal of instant gratification, but we need to do much more to present library and Internet sources as a coherent continuum rather than simply two entirely separate information domains. There are tactical issues, such as whether we should replicate sources locally or merely point to them, how we should catalog or index them, how to insure a reasonable degree of free access, and how far we can go to provide access to what have been traditionally semi-private information domains. Another problem is how to describe, index and provide effective access to the increasing number of information services and digital libraries one can reach via the Net.
Digitization for preservation and access, and publication via the Net, offers new opportunities for special collections, but we must consider carefully the priority this should be given, since costs can be high and the number of potential users often very few. The Net also gives us an opportunity to develop "logical" special collections, drawing together closely related resources physically scattered in many libraries around the world.
Sensor and digital data
Our society is now generating enormous quantities of real-time data flows, from sources such as satellites, telemetry and sensing devices, video surveillance cameras, computer-generated reports of financial transactions, and so on. We must decide how much has lasting research value and needs to be kept, and figure out who should do it and how it will be paid for. Some data flows, such as weather data, clearly substitute for data traditionally acquired and kept by research libraries in print form, and there is no question that much of the scholarly research of the future will be based on retrospective analysis of archives of such data flows. But digital data often requires sophisticated computing capability for effective access, and libraries may no longer be in a position to provide it. Another issue is whether libraries should provide access to the ongoing, real-time data flows that are becoming critical information resources for many people in business, government and the professions.
These can be as trivial as listservs, but include a variety of more elaborate distributed collaborative activities in which people share text and video, talk, operate scientific apparatus, review data and carry out group projects. Since all such collaborative environments can be recorded and saved, and since they constitute an increasingly significant part of the process of scholarly communication, we must decide what parts of these collaborative electronic session should be kept, and what role libraries should play. The technical, cost and copyright issues involved appear far more complicated than even those we face with electronic print and Internet resources.
Few libraries will be able to deal with all these resources, and as we progress through the taxonomy matters become less and less clear or definite. We will have to give careful thought to our priorities, and where we invest our limited funds and staff time. Most important, we will need to maintain an awareness of the full range of relevant resources, and strive to give users as coherent a picture as possible of what is available, how various information sources complement each other, and how each type is most effectively accessed.
Summary by Bill Whitson, UC Berkeley
Return to top Return to Conference Program page