by Jerry D. Campbell, Chief Information Officer and Dean of University Libraries, University of Southern California

Background

In September 1999, the Association of Research Libraries and OCLC hosted a meeting designated as a Strategic Issues Forum for Academic Library Directors. Held in Keystone, Colorado, and attended by 80 academic librarians, the meeting yielded a somewhat unexpected outcome when a consensus emerged that in the World Wide Web environment the library world is in danger of abandoning its constituency to commercial information services.

The basis for the consensus was not that libraries don’t offer web access but rather the general nature of this access. Our existing library web pages are focused mainly on individual libraries and the resources and services they offer. Consequently, with rare exceptions web access hosted by libraries is not designed to serve as a general entry point for the larger world of web-accessible resources. Neither are the multitude of individual library web pages equal to a single, widely known, dependable beginning place for research in the web environment. For this reason, library users and librarians in search of web-based information turn instead (and are sometimes directed through library web pages) to search engines or information services created by what may be referred to in the new parlance as information.coms1 A number of recently created content providers may also be counted among the information.coms, including netlibrary.com, questia.com, and several ebook providers.

While they appreciated the benefits of such information.com portals, Keystone attendees nonetheless observed that these commercial portals were established on different values and principles than those espoused by the library community. In addition, the information.coms were seen as pursuing different goals and purposes than libraries. In light of this, the Keystone attendees concluded that libraries (and librarians) should take collaborative action to address this situation. In particular, Keystone attendees suggested that academic libraries should develop a full-service, shared web presence that they labeled "library.org." They also made a preliminary effort to describe the nature of a library.org presence and articulate the principles and values that might undergird it (see http://www.arl.org/resources/pubs/br/br207/). In addition, they considered the kind of business plan that might be required for its initiation.

Purpose

The purpose of this paper is to suggest that the Association of Research Libraries should seriously pursue the feasibility of developing a "library.org" web presence. For clarity, this paper will refer to the proposed web presence as the scholars portal, the domain name derivatives of which are available. 2 This paper will also suggest that this effort could best be undertaken in partnership with other agencies including OCLC and the Library of Congress. It is not the purpose of this paper, however, to argue the case for the Keystone Conference principles or outcomes, though that might indeed be an interesting and useful debate.

Similarly, the paper intends to argue only the general case that such a portal--a collective research library presence on the Web--is needed. Thus the paper will not propose and argue for a particular design or set of services for the scholars portal, though a brief outline of these will be suggested as a means of providing substance for the idea. If ARL chooses to pursue the concept, the real particulars of the portal must emerge as the result of careful thought and broad discussion. In the same fashion, the paper argues for the partnership noted above not in order to be exclusive but in order to provide the necessary critical mass of expertise to undertake development of a scholars portal. While increasing the number of partners adds complexity, the consideration of other strategic allies must be part of assessing the concept.

Justifications

With the availability of what already numbers several dozen information.coms, one might reasonably ask whether research libraries should bother taking concerted action to get into the web information portal business. After all, the information.coms offer a variety of different formats and approaches to finding information, and our constituencies are accustomed to using them. In addition, developing a scholars portal would undoubtedly be a complex and expensive undertaking and would require concerted action on a scale that has been difficult to achieve, let alone sustain, within the research library community. Letting a thousand flowers bloom, it turns out, has always been easier than cultivating a garden. And yet, there are sufficient reasons why ARL should consider pursuing such a project.

Among the most obvious of these reasons are certain inherent drawbacks and limitations common among the information.coms that appear when they are regarded from the scholar’s point of view. Though each of these may be understandable, in some cases even expected, they nonetheless present fundamental difficulties for our constituencies. First among such drawbacks stands the undependable nature of the results of information.com services. This stems foremost from the basic reality underlying a ".com," namely, that it must be a net income-producing operation. Thus each information.com is based upon a business model that guarantees profitability. There are, of course, a variety of such models ranging from payment for direct listings to the display of banner advertising, with some being hard to detect and others being graphically obvious. The problem, however, is that each business model determines the architecture of the service and significantly affects its functionality in ways that are difficult or impossible to detect by the customer. In other words, results obtained through information.coms are not returned solely for their accuracy or quality but are based on unknown criteria--a circumstance unacceptable for scholarly research.

In addition, such services do not plumb to the depths of information ordinarily reached by researchers in subject disciplines. Conceptually, one might say that information.coms are horizontal rather than vertical information retrieval services. So except for the most straightforward types of inquiries (those aimed at known sources, high-level key words, aggregated data, current events, and so on), their ability to facilitate what the academy considers in-depth research is extremely limited. In the best of all possible web-based worlds, the subject library of the future might be conceived of as a highly customized, narrowly focused search engine adapted to the character of publications and research habits of a discipline. This is not, however, the nature of current information.coms and there is no evidence yet that one or more will evolve in this direction.

Indeed, the largest drawback of all may be precisely that information.coms in general have not been designed with the particular needs of the research community in mind. Rather, for good business reasons, they appear focused on broad customer bases and the most common information-seeking activities of the general public. This may simply be a necessary condition for success with commercial enterprises. More limited niche markets are certainly not prohibited in the commercial world, and indeed, some are beginning to appear in the education sector. At this time, however, the information.coms with an education focus are aimed more at providing content than web searching and are still able to offer only small and basic collections as gauged by research library standards.

In addition, these and other information.coms are proprietary and distinct and are themselves in some need of aggregation for portal access. 3 For instance, while one will occasionally find information on a university web page about how to search the Internet including a listing of search engines, never will one encounter such a listing when accessing an information.com, since it would be self-destructive to promote one’s competition. Thus, ironically, the information.com world, while replete with help for web information seekers, in the final analysis adds to the overall complexity of the situation.

There are other more proactively positive reasons that ARL, along with strategic partners, should undertake to build a scholars portal. One may be simply that there is arguably no other agency (or group of agencies) with the breadth of awareness, the information, the skills, and the objectivity to succeed at doing so. This is not meant to be a grandly arrogant statement but rather to point out the unique character of the research library community: the libraries themselves, those who operate them, their not-for-profit status, and the potential resident therein for making sense of the world’s largest unmanaged database, the World Wide Web. In other words, research librarians can likely create a scholars portal better than anyone else.

Perhaps less noble but still important is the growing fact that increasingly the world’s business, including the business of research, is becoming web-based. Those agencies that wish to survive, let alone thrive, are busily developing new web architectures and exploring how to migrate significant portions of their business to the web environment. In the academic community, this move to the Web includes internal administrative business functions and increasingly the core functions of teaching and learning. Similarly, in the research library environment, integrated systems and digital library experiments have migrated to web-based functions almost totally. As of yet, however, this move to the Web for research libraries does not include an effort to make sense of the Web itself. Thus, there is an opportunity for ARL both to perform a service and to heighten its visibility by playing a leadership role in designing and developing a scholars portal. Conversely, failure to create such a collective web presence and service may mean that research libraries as a collective, and ARL, will become increasingly invisible to the web generation.

Finally, perhaps the most persuasive arguments in favor of ARL undertaking this project are the magnitude of the need and the significance of the opportunity. Scholars must be able to depend upon the veracity of the results of their efforts. At the moment, there is simply no satisfactory means of certifying results obtained from the information.coms. The proposed scholars portal recommended here would reflect the values inherent in the scholarly community, including the zeal for accuracy and dependability of data. These "academic" biases ideally would be clearly revealed to the researcher on the portal. For these reasons, there may be no more significant contribution that ARL could make to present and future generations than that of providing highly functional, dependable, and academically credible access to the Web.

The Scholars Portal (http://www.scholarsportal.org)

The potential usefulness of a scholars portal is so extraordinary that it is tempting to describe such a wonderful service that it would be practically unachievable. The following description, therefore, must be thought of as categories of possibility that may be mixed in whole or part. They are also clearly not the only possibilities. Indeed, the real work (and the fun) will consist of creating the design of the portal, deciding what to include, and determining where to begin.

Content:

The scholars portal would promote the development of and provide access to the highest quality content on the Web. Through the efforts of a myriad number of agencies the information content of the Web is growing exponentially. What is needed, however, is the addition of information content of academically sound quality. The scholars portal would facilitate the addition of high-quality material by fostering standards, searching across databases, and offering a variety of supporting tools. As a result, libraries, corporations, and many other organizations would be empowered to contribute to an accessible, distributed digital library. The existence and efforts of scholars portal, therefore, would accelerate the growth of high-quality material and facilitate what has been referred to as the global relational research library. 4 Such a library could contribute to a reformation in the format of scholarly publishing and usher in access to a vast and heretofore largely unusable body of original material, specialized resources for communities of scholars, and accumulated scientific data.

Enhanced Services:

With the growing use of asynchronous learning methodologies, there is also an increasing need for extending certain elements of traditional library public services to the Web. This is already beginning to happen through experiments with virtual reference environments. Perhaps the most ambitious test of the virtual reference environment today is the Collaborative Digital Reference Service (CDRS) sponsored by the Library of Congress. The goal of CDRS is to provide professional reference service to users anywhere anytime by means of an international, digital network of libraries. 5 The service will include electronic responses to questions, including document delivery, when appropriate and 24x7 virtual access to a live librarian when necessary. To accomplish this, CDRS will utilize a cooperating group of libraries around the globe in a shared enterprise. Similarly, a group of research libraries have teamed with OCLC to propose the creation of a virtual science chat room where undergraduates will find 24x7 access to a librarian-staffed chat room to receive assistance with information in science, mathematics, and engineering. 6

As these and other experiments succeed in establishing viable models for virtual reference services, the role of a scholars portal would be to discover and promulgate best practices, to expand the subject coverage of virtual reference, and to make such services accessible through a single gateway. In addition, scholars portal would foster the extension of web-based services further into document delivery, provision of specialized supporting materials, experimental shared work spaces, and activities in support of alternative scholarly publishing.

Engines and Tools:

To begin with, scholars portal would provide a number of highly desirable gateway functions. These might include an explanatory guide to information.coms as well as cross-platform access to commercial databases. It might also offer sophisticated electronic thesauri to guide researchers toward areas of interest with precision. The goal would be for the scholars portal to be adopted as the place to start for anyone seeking academically sound information.

Thus, a primary function of the scholars portal would also be to provide researchers with an alternative means of retrieving dependable information beyond the capacity of the information.coms. Its goal would be to provide highly focused search engines adapted to the technical languages of the various academic specialties. By customizing search engines in this fashion and directing them to dependable sources of information, scholars portal would evolve increasingly "intelligent" automated systems and improve the success rate of query systems.

The foundations for moving in this direction are already visible in the emerging capacities of OCLC’S Cooperative Online Resource Catalog (CORC). CORC represents the development of tools capable of automating the creation of metadata for web-based electronic resources, a necessary precondition for access.7 Without CORC and other tools in the CORC suite, the volume of data on the Web would never be harnessed by manual techniques.

Conclusion: The Integrated Whole

Without doubt, the need within the academic community for a scholars portal is high, and the capacity to develop it is resident in ARL, OCLC, the Library of Congress, and other agencies. There is also no doubt that today represents a fleeting moment of opportunity to engage the effort. If the research library community accepts the challenge of and succeeds in creating the scholars portal, it will put a tool of immense value in the hands of the academic community. In doing so, it will for the first time in the web environment, bring together high-quality specialized content, commercial sources of data, viable search engines, and virtual human and machine-based assistance. It will also create an extraordinary and exciting new future for the research library community that draws on the best from its past adapted in form and function for the future. In the words of one ARL colleague, it will have created the real information commons.8

~~~

Postscript

As is indicated in this paper, the development of portals to provide entry points for web-based information has become an increasingly important topic within the commercial information sector. Additionally, academic institutions are using the concept of the portal to address issues of providing access to information in a cohesive manner for members of their communities (see, for example, the Boston College University-Wide Information Portal and the effort by two-dozen Ivy League and state colleges to create portal software that can be shared). Harvard University and the Digital Library Federation have proposed a planning process to define the means by which research institutions can make information about their digital finding aids more accessible through harvesting metadata. Library conference programs, recent articles, and the new journal entitled Portal speak to the library community's interest in portals. The above paper was prepared and presented by Jerry Campbell for discussion by the ARL membership at their May 2000 meeting to consider what role the Association should play in portal development for the scholarly community. Subsequent to the meeting, the ARL Board established a small working group to think through and recommend a practical vision for a Scholars Portal and a possible ARL role in developing such a proposal.

Endnotes

1. For one listing of such search engines and metasearch engines see the excellent website created by the University of Minnesota Duluth Library .

2. The name "Scholars Portal" was suggested by USC library faculty member Deborah Holmes-Wong. Following her suggestion, these domain name variants were reserved: scholarsportal.org; scholarsportal.edu; and scholarsportal.com. For purposes of this paper, the identifying phrase "scholars portal" will be italicized and utilized without an apostrophe.

3. Some members of the commercial publishing industry may be taking a step in the direction of solving a portion of this problem through the development of the Digital Object Identifier, an identification system that would allow the linking of web-based information across the databases of different publishers.

4. Harold Billings (in press), "Shared Collection Building: Constructing the 21st-Century Relational Research Library," Journal of Library Administration (also delivered at the conference on "Research Collections and Digital Information," Oklahoma City, 2 March 2000).

5. Information about the CDRS was supplied to the author by Diane Kresh of the Library of Congress.

6. Information about the Real-Time Reference for Undergraduate Students in Science, Mathematics, and Engineering was supplied to the author by Gloriana St. Clair of Carnegie Mellon University. The project proposal is available online .

7. See http://www.oclc.org/oclc/promo/10520corc/index.htm.

8. Billings.