Current Issues in Education > Volume 3, 2000 > Number 6

Citation Information

Willinsky, J. (2000, September 13). Proposing a knowledge exchange model for scholarly publishing. Current Issues in Education [On-line], 3(6). Available: http://cie.ed.asu.edu/volume3/number6/.

Proposing a Knowledge Exchange Model for Scholarly Publishing

John Willinsky
University of British Columbia¹

Abstract

This article argues for developing and testing a new model for scholarly publishing that is based on the exchange economy of ideas that lies at the heart of the research process. As a starting point in developing this model, it reviews current publishing alternatives, challenges the "journal" concept, and outlines the organizational, technical, and economic features of a Knowledge Exchange Model (KEM) for scholarly publication. This preliminary design, growing out of initial website experiments in creating public contexts for educational research and a larger analysis of the public qualities of knowledge, demonstrates how such a model can improve both the academic and public value of research, while also addressing the crisis in library serial costs. The aim of publishing the rationale for this model is to initiate an online discussion (archived here) among readers of Current Issues in Education (CIE) on the general issue and specific features of this proposal.

Table of Contents

• Introduction
• Online Discussion
• Journal's End
• Among the Alternatives
• The Public Knowledge Factor
• The Design of the Knowledge Exchange Model
1. Basic Organization and Responsibilities
2. Review Process
3. Knowledge Exchange Portal
4. XML Indexing
5. Economic Model
• Conclusion
• References
• Notes

Introduction

The Association of American Universities (AAU) and the Association of Research Libraries (ARL) held an unprecedented meeting in March, 2000 to discuss the troubled state of scholarly publishing. Amid the traditional affirmations of research's contribution—"from the lab to the classroom to industry to the public, the advancement of knowledge through research and teaching is an invaluable contribution made by higher education to the public good"—a profound sense of foreboding over the future of research held sway at the meeting. The result was a radical declaration that "the current system of scholarly publishing has become too costly for the academic community to sustain" (ARL, 2000a). It is common to hear of libraries and universities being held to ransom by their faculty's own work, as they face runaway inflation in serial subscription costs largely within science, technology, and medicine (Harnad, 1997).²

The AAU and ARL see no other way forward than leaving the old model of publishing behind; and while they do not yet endorse a new model, they do provide a series of principles for what they term "emerging systems of scholarly publishing." These principles include, not surprisingly, cost containment and reduced time to publication. They embrace the use of new technologies, while insisting that "electronic capabilities should be used, among other things, to: provide wide access to scholarship, encourage interdisciplinary research, and enhance interoperability and searchability. Development of common standards will be particularly important in the electronic environment."

They also venture into far more dangerous territories by proposing that "the evaluation of faculty should place a greater emphasis on quality of publications and a reduced emphasis on quantity."

Focusing on quality over quantity as a means for containing subscription costs is a bit like having fewer, better children to reduce the risk of child-napping – to continue the ransom metaphor.³ The problem is rather the assumption that publishers alone can guarantee the quality and distribution of this public good, which has allowed them to drive scholarly publishing to the brink of unsustainability. Yet these commercial publishing interests are, in many cases, latecomers to an enterprise that has otherwise run as a sponsored exchange economy. That is, the university (with public and private money for salaries and grants) sponsors its faculty's participation in a knowledge economy that is organized around the exchange of research. Faculty are paid by their own university to do research, to publish research, to review research, and to edit research, in addition to the teaching and service they provide. These research activities, once paid for or sponsored, then lead to the exchange through correspondence, conferences, and journals. In offering this research for exchange, the faculty member is not required to recover salary and grant costs.

This exchange is as vital to research as motion is to sharks, and it is undoubtedly facilitated by those who publish it. The cost of an academic journal subscription only covers the publication costs (principally editorial, printing, and distribution) rather than the actual cost of producing the content.⁴ The information is not free (whether it wants to be or not), although the journal publisher gets it for that price; it has been sponsored for the purposes of exchange, most often I would hazard, at public expense.⁵ Given this way of thinking, the crisis in scholarly publishing could be cast as an imbalance between the exchange and service economies of scholarly communication, with the (publishing) services currently driving the process to the brink.

It may be that the publishers are holding this knowledge for ransom because the handwriting on the screen foretells the demise of the Gutenberg galaxy. Certainly, change is afoot, with both authors (such as Stephen King) and bookstores (such as barnesandnoble.com) publishing their own series of e-books, while Napster.com has proven the exchange-economy power of peer-to-peer distribution. Publishers are scrambling in turn for new business models, while a number of prominent universities have opted to go with a for-profit knowledge-marketing start-up, by the name of Fathom.⁶ Against the background of these developments, I am proposing that those who share a concern with knowledge's place within the public sector explore how the academy might strengthen scholarship's original and persistent exchange economy which it has long sponsored.

The Knowledge Exchange Model (KEM) presented here is designed to return the emphasis of scholarly publishing to its original exchange economy. It employs peer-to-peer communication at the institutional level of the research library—with its paramount concern for the organization and conservation of this knowledge—while retaining those publishing services that improve the quality of the research.⁷ This exchange would be managed through a global system of research libraries and learned societies. Economically, this model seeks to demonstrate that redirecting a substantial portion of the research library's current serial acquisitions budget to building and sustaining this, or a variation on this, exchange model will not only improve the scholarly quality of this research knowledge but will, at very little extra cost, afford free universal access to what is, after all, claimed to be a public good. This is, then, a model for fostering scholarly exchange and public education. While I recognize that this model's reliance on new technologies will not make this learning available to everyone by any means, I would ask that it be judged on its promise to increase access to research and to increase the value of that access. The question is whether, compared to the current scholarly publishing system, more people will be able to get at this knowledge in ways that are helpful to them.⁸

The KEM itself is an exchange-economy bargain, a paste-together of borrowed and duly cited (I hope) ideas drawn from those rethinking information systems and scholarly publishing. It also grows out of the Public Knowledge Project research program which, in its efforts to renew the social sciences' social contract, has been exploring the historical and epistemological issues of public knowledge, with a focus on research's particular contribution (Willinsky, 1999). The project has developed a series of experimental websites that have sought to demonstrate the public value of research by directly linking a body of scholarship to, in one case, a series of newspaper articles on an education topic, in another, to the review of a provincial educational policy, and in a third, to an online professional development program for teachers.⁹  The limited success of these small-scale sites in fostering the public's engagement with research, which was combined with an analysis of parallel developments in both the private and public sectors of this knowledge economy, has provided rich lessons in the technical and human issues that come of integrating and supporting research into public settings (Willinsky, 2000). The resulting proposal for a new model is published here to further this participatory design process as a means of inviting a public discussion of how best to work with what is bound to change in scholarly publishing. While this article is addressed to researchers, editors, librarians, publishers, and representatives of learned societies, the idea will also be tested with policy makers, journalists, and the public. This research process will result in an organizational plan, an economic model, and a working prototype for one and possibly more publishing systems. The goal is not to build the One Great System or killer application for scholarly publishing. It is to join with other publishing alternatives to demonstrate, and stimulate thinking about, what can be done to improve the academic and public value of research.

The KEM builds on current efforts to further an emerging global research community, moving from isolated outposts, rich or poor in knowledge resources, into a "distributed information space," to use the term Carl Logoze and David Fielding, of Cornel University, use in their work to connect digital libraries (1998). The KEM proposes the development of what is, in effect, a single database out of a global community of research libraries. In some ways, it resembles the medieval model of European monastic libraries which, taken as a whole, amounted to a distributed database of locally developed resources, linked by an open-source interoperable programming language (Latin) connected by a peripatetic network of scholarly clerics, aided by the 12th century textual technology of the footnote (Illich, 1997).

In response to the contemporary crisis, the Association of Research Libraries has established the Create Change program which "seeks to address the crisis in scholarly communication by helping scholars regain control of the scholarly communication system—a system that should exist chiefly for them, their students, and their colleagues in the worldwide scholarly community, not primarily for the benefit of publishing businesses and their shareholders" (ARL, 2000b). Create Change provides faculty with the "tools" to reestablish control, if only in the shape of form letters for faculty to send to publishers protesting costs and refusing to referee articles for the offending journal. Yet when Create Change's leader, Rick Johnson, the Enterprise Director for the Scholarly Publishing and Academic Resources Coalition (SPARC), refers to how "Gutenberg is remembered because he gave people the gift of access to others' ideas and the vehicle to disseminate their own," it is well to recall what a hard-won gift this access has been, from print in the vernacular languages to universal literacy. A "long revolution" was how Raymond Williams named this corresponding development of democracy, economy, and culture, so closely related to the printing press (1961). Whether the Web will stand as another chapter in this long revolution or as a diverting interlude, like Gutenberg's printing of Indulgences, we cannot know for sure. But this risk is off set, I hope, by the value of rethinking what we want from research; and what we want, I would argue, is no longer well served by the academic journal, in either its literal or virtual forms.

Journal's End

It is difficult to walk away from the library stacks of bound and unbound volumes, the faculty-office buildup of a career's worth of journal subscriptions, and the dignified record that journals can make of a scholarly life. Yet the journal is all about a publishing era dominated by printing presses and postal services. It is the brilliant, three-centuries-old solution to the dissemination challenges faced by, in the British instance, the first corresponding secretary of the Royal Society, Henry Oldenburg. Why send out endless reams of hand-written letters to inform and test what was new among "the Curious and Learned in things of this kind," he realized, when one could capitalize on existing technologies and emerging genres, such as the periodical: "It is therefore thought fit to employ the Press," he explained to readers of the first academic journal in English, "as the most proper way to gratify those, whose engagement in such Studies, and delight in the advancement of Learning and profitable Discoveries, doth entitle them to the knowledge of what this Kingdom, or other parts of the World do, from time to time afford. . ." (1665).¹⁰  I feel exactly the same way about the Web as Oldenburg felt about the Press. It shows considerable promise as a way to gratify those who study and delight in the advancement of learning.

The journal's bundling together of recently received correspondence and "news" for re-mailing has evolved into a commercialized and hierarchical system of journals that has, in recent times, not only led to library's reducing their holdings, but may well be suspected in their prolificacy of impeding the connection and exchange of ideas. Along comes the web and suddenly the “economies” of bundling have far less to offer, while the carry-overs of peer review and prestige can be achieved, one suspects, by others means.¹¹  Still, three-century-old habits die hard and the alternative models that seek to replace it may be futile, at least in our lifetimes, leaving the electronic journal to be the QWERTY keyboards of the twenty-first century.¹² 

Among the Alternatives

Although the KEM is not intended to compete with other alternative publishing models, it does try to go a further step or two. My social science background, for example, has led me to think about research's public role and has enabled me to treat this spread of knowledge as itself a worthy research topic, leading to the developing and testing of a prototype. This model expresses a commitment to more than reduced costs and improved scholarly service; it seeks to study the possibilities of increased public access and presence of research. And just as the other alternatives imagine they can be readily applied to other disciplines, I am starting with a similar proposition, while allowing the professional academic associations to play a critical role in setting the standards for each discipline based on its culture, in areas such as the sciences' use of preprint access, for example.

The most prominent and well-established of the alternatives at this point is "the self-archiving solution," as the Open Archives initiative is calling its system of authors submitting "preprints" to an automated server system for posting to the web, prior to their acceptance for journal publication. This model began with the Los Alamos National Laboratory (LANL) Physics Archive (http://xxx.lanl.gov/), founded in 1991 by Paul Ginsparg. The LANL Archive now houses well over 100,000 articles, mirrored worldwide in 15 countries with over 120,000 "hits" daily, giving rise to similar such systems in other disciplines and locations. This idea of a self-archiving repository is a decided improvement in access over the earlier circulation of "preprints" among corresponding colleagues and "invisible colleges." These preprint archives have led in turn to the Open Archives initiative which is, as its website explains, "a forum to discuss and solve matters of interoperability between author self-archiving solutions, as a way to promote their global acceptance." The Open Archive, which employs metadata and XML for connecting resources across databases, began by considering two models—the distributed searching and a harvest approach—for linking archival databases modeled on LANL.¹³  They rejected the "distributed searching" approach, which is equivalent to a search engine trawling the entire web, “in light of the possible emergence of thousands of institutional and/or subject-oriented archives worldwide” (Ginsparg, Luce, and Van de Sompel, 1999). Instead, they went with the "harvesting model" which “would allow trusted parties—the ones that subscribe to [their] Santa Fe Conventions—to selectively collect data from different archives” using metadata and other protocols. They have agreed that the "providers of data [who set up the archives] use a standard mechanism to state the conditions under which their datasets can be used by implementers of data services [to support end-users]" (Ginsparg, Luce, and Van de Sompel, 1999).¹⁴ 

The KEM, as it currently stands, combines the distributed and harvesting models. That is, it would encourage a global system of institutional archives that would support distributed searching by adhering to universal standards and conventions among academic and research library associations. And for this reason, it is not, as the LANL archive claims for itself, the "low-entry requirement" or the "realistic approach at this point in time" (Ginsparg, Luce, and Van de Sompel, 1999). So why, then, given this archive's "overwhelming proof in practice," and its readiness "for extension to all disciplines," according to its outspoken advocate psychologist Stevan Harnad, would we need to develop and test a more complete revision of the publishing system?¹⁵  Harnad, who is also editor of the University of Cambridge Press' Brain and Social Behavior journal, would "overlay" a refereeing process to the LANL archive, financed by page-charges (at less than $50 per “page”) paid by the author's university or by the government, with the review and editing managed by publishers with their "known and trusted brand-name journals."

Harnad's easy and attractive extension of the LANL archive still has three factors working against its low-entry requirements. Building a system on the pre-print model, which is but a particular sub-cultural phenomenon in scholarly communication, will not well serve all fields (Glass, 2000).¹⁶  Publishers offer no special value, over the PAAs, in managing the peer review system. And, finally, Elsevier Science, the world's largest commercial journal, refuses to allow open access to Open Archive initiative, which means the preprint is freely available while the refereed version is subscription only, doing little to improve the library-threatening economics of access for the system as a whole (Kiernan, 1999). So, my thinking is that researchers, especially those working in related areas of literacy and public education, as I do, need to risk seeming unrealistic and naïve in testing new possibilities for this knowledge that go beyond economy and speed.

There are at least two other proposals that warrant consideration here, if only because they share the spirit of what I am proposing. David E. Schulenburger, the Provost of the University of Kansas, has been inspired by subscription increases of five times the inflation rate among the for-profit journal publishers (and three times for the scholarly societies) to propose a National Electronic Article Repository (NEAR). He seeks an agreement that would ensure that all research articles appear in the NEAR—"a single, publicly accessible repository" sponsored by an association of libraries, universities or a foundation—90 days after their publication in a journal (1998). I would take this to be an initial, transitional step, in taking back control of the publishing process while enabling the journals to continue to offer some value in return for their reviewing and editorial services. As such I think it leads well into the more far-reaching changes posed here and by the Scholars' Forum, put forward by Anne M. Buck, Richard C. Flagan, and Betsy Coles, all of the California Institute of Technology (1999).

The Scholars' Forum is also a response to the failure of publishers to come up with, in the face of new technologies, "a new paradigm for scholarly communication" or even to reduce the costs of online journals, while dangerously usurping the library's archiving function by retaining "the only version of a publication in a digital form." We must then, they declare, "develop for ourselves a new, affordable model for dissemination and preserving results, that synthesizes digital technology and the ongoing needs of scholars." At the center of the Scholars' Forum is to be a consortium of universities which will take responsibility for "the servers containing preprints, journals, proceedings, and abstracts," as well as professional academic associations which provide content expertise through editorial boards and the journals. Peer-reviewed articles can be "aggregated into one or more journals," as well as supported by threaded discussions and automated indexing with metadata tagging. The Consortium will provide technical writers to help researchers with "preparing manuscripts that meet editorial board's submission standards," while its archives will be "future-proofed" with "200-year secure paper" backup copies. Buck, Flagan, and Coles conclude with a call for "conducting meaningful experiments," and the KEM is clearly a variation on this theme, one that, perhaps, most notably, drops the journal as an organizing principle.¹⁷ 

In responding to the Scholars' Forum, Paul Ginsparg expressed concern over the "degree of cooperation and coordination of interested parties in advance of getting anything off the ground (especially considering that the research enterprise is global, not just a national enterprise)," referring to the time required as on the order of monkeys-typing-Shakespeare (SFD, 05/02/99). His point applies no less to the KEM, although one has to allow for the considerable work currently going into developing international standards on metadata and other communication protocols, including those for scholarly work.¹⁸  These global systems can begin to work with only an initial degree of coordination, with further cooperation growing out of their demonstrated value. Part of that demonstration, and clearly the distinguishing feature of the KEM among other publishing alternatives, will be its deliberate contribution to the public value and presence of this research.

The Public Knowledge Factor

Open access to research represents one of the real advances of this new technology over older publishing systems. This access will, after all, support scholarly inquiry on far more of a global basis. It will provide specialized knowledge to professionals and the public at large, especially in such areas as health, justice, welfare, education, and other areas; just as this open access will, more generally, support people's educated curiosity and interests. This access is consistent with the American Association for the Advancement of Science recent talk of forging a new "social contract for science" that includes a commitment on the part of scientists to "communicate their knowledge and understanding widely in order to inform the decisions of individuals and institutions."¹⁹  It fits with how the leaders of the United States and Great Britain have taken a joint stand in declaring the results of the human genome project to be public knowledge, while on a far more personal level, we are seeing the availability of online health information begin to shift the dynamics of the doctor-patient relationship (Freudenheim, 2000, p. A1). This public interest in knowing more of what is known is also reflected in the planned $1.2 billion dollar expansion of CNN's online presence that features research tools and more in-depth coverage (Robinson, 2000).

My own academic responsibilities have led me to consider how universal public access to the broad range of social science and humanities research could contribute much to the deliberative qualities of democracy, providing people with the basis of greater engagement and participation in matters affecting their lives, not only in the voting booth, but in their homes, neighborhoods, and workplace, as well as globally. At issue, for me at least, is how this public good can do more to strengthen this information economy's public sector.

The question I want to test, from a research perspective on developing the KEM, is whether more coherent and intelligible public access to this research can contribute to what political philosopher James Bohman identifies as the democratic need for "more effective deliberation" (1996, p. 2). As "democracy is justified to the extent that it makes possible the public use of common practical reason," which Bohman points out is a widely accepted point, then let us assess whether there are ways of enabling the research that bears on public issues to do more to inform and support that public use of reason (p. 4). Against Bohman's concern over "the lack of cultural resources, and of opportunities to develop various public capacities," I am proposing we make widely available the already existing and largely publicly funded intellectual resources of the academy (p. 108). The printing press earlier proved to be democracy's handmaiden, and I am looking for a further extension of this access to knowledge as a powerful source of political participation and public education.

The very origins of the social sciences date back to nineteenth-century amateur efforts to utilize statistics and other data on behalf of the dispossessed (Smith, 1994). Today, I am asking that any reform of scholarly publishing include improved public access to research.²⁰  As people learn to argue from and work with this research, and with the researchers as well, they may well feel an increase in their "capacity for social agency," to return to Bohman's arguments for improving the deliberative quality of democracy (1996, p. 130). Of course, the introduction of this knowledge will not, of itself, resolve social issues, as it will be used by people and, even more likely, interest groups, of every political persuasion to argue every which way, undoubtedly drawing many researchers into the fray along the way. But if the deliberations may not go smoothly, the greater presence of the research would still stand to raise the general level of public reason and discourse. And its contribution will only increase as this presence affects how researchers frame their projects, knowing that their work now lies within people's reach to a greater degree than ever before.

Now, it may seem particularly unwise to burden an unproven new system such as the KEM with such additional political (and humane) baggage. Yet if not now, in the very reformation of the publishing process, I would ask, then when exactly would be a good time to expand research's public reach, to extend the basis of its public support and interest? The public value of health research has already been well established and figures now in the design of the major, publicly funded online database systems such as MEDLINE.²¹  Even in the rarified atmosphere of astrophysics and the other very hard sciences in the LANL archive, there is room to imagine that universal access is a boon for students and faculty in less advantaged institutions internationally, as it may also occasionally prove to be for the interested autodidact. Certainly, the idea of greater public access would appeal to David C. Bennett, Vice President of the American Council of Learned Councils, which represents 58 societies in the humanities and social sciences, who sees this as a time to establish new connections, especially given the "context of disconnection" which he observes setting scholars off from the public (1997, pp. 3-4). He looks forward to using new technologies "to allow more extensive and effective communication between scholars and non-scholars," and speaks of tailoring "access to scholarly resources for journalists, policy-makers and organization leaders and others with an interest in knowledge" (1997, p. 5).

The repositioning of the library as the core of knowledge's public sphere has been recognized in California, for example, as the Library of California has recently joined with the California Digital Library to increase access to the state's collection of public databases and digital archives, including those of the University of California. The two institutions, which are each but a few years old, have a combined budget of $10 million committed to extending public access, which includes educating the public about these new resources as well as negotiating public access to research collections (Olsen, 2000). Whether to afford scholars everywhere equal access to needed intellectual resources or to increase the presence of this knowledge as a force of interest and action in people's lives, our reform of scholarly publishing needs to capitalize on forces already underway. Such is the idea behind our efforts with KEM as a means of integrating open access with the various components and relationships that define scholarly publishing.

The Design of the Knowledge Exchange Model

1. Basic Organization and Responsibilities

The key institutional player in this model is the research library (Figure 1). The library will mount and archive the research of those employed by the institution, as well as those affiliated through an agreement with other local or foreign institutions. I am proposing that the operating principles and standards for this system be set by global versions of the Association of Research Libraries (ARL) and the various professional academic associations (PAA) that make up the disciplines. The ARL's Research Server System would provide seamless access to the resources distributed among participating libraries.²²  While such a server system could be operated by the government or the PAAs, I have chose to rely on the research libraries for a number of reasons. The ARL is leading the call for a change, as the current system is for them, as I cited above, unsustainable. The library's serial acquisitions budget is the source of funding for KEM, and the research library's staff have considerable experience and training in information science and management. The library also has a good deal of technical infrastructure and support which is designed for managing, distributing, and archiving research and related resources. They are in the best position to ensure that research placed on this Research Server System would adhere to standards of organization, access, support, and archiving that we associate with a fine research library.

Figure 1. Basic Organization and Responsibilities

The ARL currently consists of the top 121 research libraries in North America, and the KEM would ideally work with an expanded, global version of this organization which set the multilingual standards for the Research Server System that included access for publishing the work of researchers who are not otherwise affiliated with an ARL institution. The ARL would work with the PAAs, just as the individual research libraries would work with local faculty, on improving how the system works for disciplines and scholars. The PAAs would also need to form global federations with related academic organizations to develop common, multilingual standards for how knowledge within the discipline is organized and indexed, perhaps with an eye to improving the field's coherence and comprehensibility.²³  The publishers obviously have a diminished role in this model, but they would still be in a position to sell access to their content as well as some of their editorial services, while continuing to produce and market books for a wider audience.

The technical thread that ties the KEM together, enabling a reader to gather research from a variety of library servers and display it in a variety of formats on the screen, is XML which is an Extensible Markup Language that is discretely embedded in a text through a series of tags, much like the hyper-text mark-up language HTML (with more on XML below).²⁴ 

2. Review Process

Although it is left to the PAAs to determine the peer review processes and standards that serve their discipline best, the KEM is intended to enable a more open, systematic, and thorough review. The PAAs would appoint editors and editorial boards, as they often do now, to oversee and manage the publication of research in a division, topic, methodology or concept (Figure 2). Once a researcher placed a work, whether of article- or book-length, in the Research Server System for peer review, it would be directed to the appropriate editor based on its indexing tags for an initial screening (researchers could, of course, hold out for commercial publication or otherwise distribute their work). If a PAA supported "preprints," then the text might be designated "pre-review" and made immediately available. The selection of reviewers for the text would be based on a further match of indexing items, while the PAA could design the distribution, scheduling and monitoring of the reviews to better manage turnaround times.

The quality of the reviews should improve as reviewers (and readers) are afforded greater access to the original data, research instruments, and cited texts.²⁵  In setting the standards for reviewing, the PAAs would be able to preserve such traditions as having the highly regarded review the highly regarded by using an author's impact factor (number of times cited) as one of the match-up points in selecting appropriate reviewers.²⁶  What the model loses with the disappearance of the prestige journal, it can gain with the ready measure of the author's, the article's and the institution's impact factors. The PAAs' different strategies with preprints, review-monitoring strategies, bibliometric measures, awards and prizes, and Open Peer Commentary systems (Harnad, 1997) would be available to the system as a whole for consideration.

Figure 2. Submission and Review Process with Journal-less Publication

3. Knowledge Exchange Portal

The design of a portal or of multiple portals that work equally well for both scholarly and public access to this collective body of research is bound to take some time to evolve, although the National Library of Medicine's MEDLINEplus offers a promising precedent mixing basic research with practical advice.²⁷  I am starting here with a mix of generic portal features and links that we have been experimenting with in the Public Knowledge Project with educational research (Figure 3). Our efforts at helping people to make greater sense of research and how it fits with other sources of knowledge have led us to consider establishing predictable navigational structures that would include being able "to rise up" from a given study for an overview or survey of the field, as well as being able "to drill down" to the particulars of the study's data, research instrument, or sample. One might also be able to move laterally in, say, the four directions of the compass, whether to compare related studies, to consult background materials, to review or initiate discussions of the work, or to visit public sites connected to the study (Figure 4). The goal is to situate the research in ways that increases its intelligibility, contribution, and value, whether for the public, professionals, students or researchers.²⁸ 

The KEM goal of improving both the scholarly and public value of the research includes connecting research to related materials that exist outside of the Research Server System. These potential public and professional interests will depend on the field of inquiry, of course, with professional fields like health, law, education, engineering, social welfare, public policy, likely to be the most active. Still, it could as easily connect astrophysics research to astronomy club websites, archeology to tourist sites, and Shakespeare criticism to performance sites. The portal could enable students and teachers to set up links to the research for their courses. Without having to decide in advance which links are trivial and which meaningful, the system will offer the opportunity of extending the connections through shared XML tags, with the assumption that extending the connections of this work may well prove of value to those within and outside the research community.

Figure 3. Potential Features of a Knowledge Exchange Portal

Against the danger of this public access - given the regular newspaper articles ridiculing seemingly spurious research grants and conference papers sponsored at taxpayers' expense—I would hold that having more people able to learn from what is going on can only reduce the impact of the seemingly sensational or inconsequential project. Academic freedom is not best protected in a democracy, I would hold, by cultivating obscurity and hoping to avoid detection. With time, greater public access and engagement with the research process are bound to affect the formation of this knowledge.

Figure 4. Potential Navigational Structure

XML Indexing

The Association of Research Libraries (ARL) will work with each of the academic associations (PAA), as well as with international standards represented by the Dublin Core Metadata, and the IMS Learning Consortium, and the International Consortium for Alternative Academic Publication, to produce standardized sets of XML tags that adhere to a Document Type Definition (DTD) that covers the format, as well, offering a far-reaching ability to index a study's content (Table 1).²⁹  The XML tags will connect a text to past and future work, to critiques and surveys, along the navigational lines outlined above. The XML tags also enable the display of the document to be controlled by the PAAs and the users.

In preparing their research texts, writers will embed XML tags in the document using a software tool that will plug into their word processors, much as commercially available bibliographic software such as EndNote and Reference Manager do. This indexing tool will prompt researchers with default settings, pull-down menus, and other features to minimize the effort required to index a work. The PAAs will prepare index- settings that are specific to the discipline, with regular updates made available to members as a result of an ongoing review process, with researchers able to customize the tools for multidisciplinary work. Given its features, it might be worthwhile developing a version to plug into the Portal's XML search engine that would improve the engine's ability to serve the discipline. The XML system used in KEM will need to allow for (a) existing systems of tags to be updated and improved to reflect changes in a field of inquiry, as well as for software and hardware upgrades, (b) variants to be used in the texts (e.g., constructivist, constructionism, constructivism) while being represented by a standardized tag, and (c) wild-card or user-defined tags for use by, say, Special Interest Groups or splinter organizations.

Table 1. Potential XML Indexing Options for Research Documents

Although some are likely to balk at the extra work - which compares to writing an abstract, selecting keywords, and preparing a bibliography to proper form—they may well appreciate how it enables them to extend and direct the potential influence of their work, both within and outside of the research community. The other advantage it potentially affords is that it will foster a greater awareness among researchers of how their work stands within a series of research fields and issues. Knowing which work it will stand beside in support and critique—rather than thinking it sits among the unrelated pieces of a journal or by itself in cyberspace—could well encourage them to think about the greater coherence and cumulative effect of this associated body of work, a coherence no less strengthened by critique and dissent, I might add, against the current scourge of fragmentation and isolation that comes from the sheer bulk of work going on.

The XML search engine in the Knowledge Exchange Portal would work with any combination of XML tags (with a full-text search option as well), with the ability to also access XML documents (adhering to a similar DTD standard) outside the Research Server System. The XML search engine would also have a mapping feature which will enable it to track trends in research by graphing the number of studies that had been conducted over a period of time by topic, method, and sample, or it would be able to map a socio-gram that represents who cites whom within a field of inquiry, or present the current information options related to a given study, as in Figure 4.³⁰  The engine should also be able to answer such questions as: Show me only the top-40 cited articles or authors in this topic. What are the leading research methods in this area? What is the top school by citation on that topic? Map this concept's flow of influence and impact. Whose work is having the most public impact in this field, whether in the policy arena or with the media?

The ARL will work with PAAs in establishing common XML standards with outside agencies to enable global searches that include other databases (media, government policies, public programs, and the like.). The individual research libraries could work with local agencies and resources on developing such standards. Until such XML standards are universally adopted, researchers and system developers for the KEM could continue to cite URLs for related sources, documents, and sites. With agreed upon XML standards among PAAs and outside agencies, users would be able to move back and forth through these different orders of knowledge, enabling them to, say, work through government programs, with ability to check related research, public discussions, etc., at any point.

5. Economic Model

The 121 members of the North American Association of Research Libraries currently spend a total of US $480 million on journal subscriptions. The ARL estimates that by 2015, "the cost to support these journal subscriptions will be as much as US $1.9 billion, with individual libraries paying nearly $16 million a year" (ARL 2000b, Issues). The KEM's economic model is based on working with the current budgets of participating agencies, with the bulk of it coming from the serial acquisition budgets of libraries now subscribing to the academic journals (Figure 5). Its goal is to contain, if not reduce, the overall costs of the publishing system, even as I suspect that the system's increase in public access and intelligibility could well encourage greater government and philanthropic support for research libraries. Two possible revenue streams for the libraries maintaining research servers would be to license forms of re-publication to the publishers and to charge for large-scale access by research firms that aggregate, analyze, and resell research results in different areas to corporate, government, and other clients.³¹ 

Figure 5. Economic Model

The KEM would add to the ability of governments and foundations to assess, monitor, and review the state of research fields, providing a tool for exploring the scope, direction, and impact of research activities, and for assisting with the peer-review of grant applications. The funding agencies would also gain new opportunities to strengthen the public value of the research by supporting initiatives aimed at increasing accessibility, engagement, and academic-public document links, all of which are central features of this system. For the foundations—and here I am thinking of Andrew Carnegie dedicating his fortune to the "the advancement and diffusion of knowledge and understanding" (cited by Lagemann, 1989, p. 19)—the KEM would offer greater opportunities than the existing system of scholarly publication for extending globally, the reach of the projects they support among the public and researchers alike (Figure 6).

Figure 6. Contribution of Funding Agencies

Conclusion

Among Stevan Harnad's more vociferous critiques of an alternative publishing model like the Scholars' Forum is the missing motivational factor. Why would researchers want to change? How could this proposed system “compete with the existing journal corpus for authors… when the existing journal labels are the ones that carry the confidence and prestige” (SFD, 05/09/99). The answer, in the case of the model presented here, is what I hope is an irresistible blend of altruism and ego gratification. The researcher's work would now be universally available and perhaps far better positioned to do the work it was intended to do, both within and beyond its specific field of study. The contribution of the work, both within and outside of the academy, would be all the more traceable, all the more ready to be recognized.

But what, then, of the professional academic associations' motivation? What will they offer members, once they give up publishing journals? Well, when you consider that, with the inevitable shift to scholarly e-publication, it will be worthless to give members password-access to PAA "journals" to which their institutions already subscribe. If this membership advantage is about to evaporate anyway, then the PAAs would do well to review what alternative publishing models have to offer. With KEM, they will continue making prestigious editorial appointments both nationally and internationally, as well as working globally and nationally to rationalize the field's divisions and categories, thereby extending the reach of their members' research. They can continue to offer online “newsletter” services to their members (possibly subsidized with related ads) that include job notices, grant opportunities, professional events, and related news. They will provide updated XML indexing tools to support their member's writing and searching. The real value, though, may be in offering members periodic review services, based on interest profiles, covering new publications, research trends and developments in related fields, other guides to staying on top of it all.

Harnad rightly notes that “many scholars and potential scholars the world over have access to very little of the learned literature,” and this is all the more so with journals, I have found, rather than with books, due to the financial commitments entailed in journal subscriptions. Here, then, is the chance to participate in a system that greatly increases access to a wider readership of scholars and other potential readers, a system that makes this knowledge far more a thing of democracy. This I take to be motivation enough for researchers, editors, editorial boards, and learned societies.

That still leaves the question of how a model on this scale could get started, once the economic model, the working prototype, and such participating agencies as the ARL and a PAA or two felt there was reason enough to adopt it. Although it could begin with the equivalent of a single journal and research library, it really needs to demonstrate itself across a discipline. To start where I live, I could imagine the American Educational Research Association, which publishes five journals, working with the ARL to establish a large-scale implementation with a substantial grant perhaps from a foundation or federal agency (much as the Public Broadcasting Corporation was launched some decades ago). It could adopt the KEM prototype's software, portal design, and XML indexing tool to establish a network of ARL libraries (or a central ARL server), while having its researchers submit their work with XML tags with much assistance, perhaps, from the libraries. The U.S. Department of Education and other agencies might begin to coordinate their research resources using this system to facilitate the linkage between data sources, research, and policy. Educational publishers might negotiate licensing fees to the system for universal access to their backlists.

But before all of that, we need to continue the discussion over what we would have of scholarly publishing. As the world of publishing at large is in flux, we need to stake out the critical issues for scholarly inquiry and its public presence. From my perspective, at least, we need to argue for this the place of this knowledge within the public and private sectors, amid the partnerships and sponsorships, of this information economy. I worry that universities are chasing the short-term gains of corporate partnerships and start-ups, with their libraries caught up in easy fixes and technical work-arounds with shared databases and publisher's archives, while what is also needed is a larger vision of what we would have of knowledge. Thus, I argue for taking hold of the larger conceptual possibilities, for seeking a more radically public repositioning of the work that we do in the name of knowledge. The publication of this KEM proposal is meant to provoke a consideration of new systems, in their impossible details and grand possibilities. It is intended to solicit the participation and consultation of this journal's readership in realizing how such systems can be made to work at both a practical and principled level to increase this hard-won knowledge's contribution to the public good. It is about taking greater responsibility for what we imagine we know.

Authors

John M. Willinsky received his Ph.D. from Dalhousie University in 1982. He is currently the Pacific Press Professor of Literacy and Technology at the University of British Columbia, where, from 1990-96, he was Director of the Centre for the Study of Curriculum and Instruction. In 1997-98, he held the William Allen Chair in Education at Seattle University. He has authored some ten books, among which are The New Literacy (1990), Technologies of Knowing (1999), and Learning to Divide the World (1998) which won the Outstanding Book Award of the American Educational Research Association for 1999. His most recent book, If Only We Knew: Increasing the Public Value of Social Science Research (2000), expands on the ideas introduced in the present article. Dr. Willinsky can be reached via e-mail at john.willinsky@ubc.ca.

References

Arnold, K. (1996). The body in the virtual library: Rethinking scholarly communication. In Eds. T. M. Harrison and T. Stephen, Computer networking and scholarly communication in the twenty-first-century university (pp. 383-397). Albany, NY: SUNY Press.

Association of Research Libraries (ARL). (2000a). Principles for Emerging Systems of Scholarly Publishing. Available: http://www.arl.org/scomm/tempe.html.

Association of Research Libraries (ARL). (2000b). Create change: A resource for faculty and librarian action to reclaim scholarly communication. Washington, DC. Available: http://www.arl.org/create/home.html.

Bennett, D. C. (1997). New connections for scholars: The changing missions of a learned society in an era of digital networks. Occasional Paper, No. 36.Washington, DC: American Council of Learned Societies.

Bohman, J. (1996). Public Deliberation: Pluralism, Complexity, and Democracy. Cambridge, MA: MIT Press.

Buck, A.M., Flagan R. C., and Coles, B. (1999). Scholars' Forum: A new model for scholarly communication. California Institute of Technology. Available: http://library.caltech.edu/publications/scholarsforum/default.htm.

Carlson, S. (2000, August 10). Columbia's Fathom site attracts 4 more prominent partners. Chronicle of Higher Education. Available: http://chronicle.com/free/2000/08/2000081002u.htm.

Case, M. M. (n.d.). ARL promotes competition through SPARC: The Scholarly Publishing & Academic Resources Coalition. Association of Research Libraries. Washington, DC Available: http://www.arl.org/newsltr/196/sparc.html.

Danner, D. and Taylor, P. W. (1997). Principle and practicality: Funding electronic access to Washington State government information. Journal of Government Information, 24(5), 347-359.

Evans, A. T., McNutt, R. A., Fletcher, S. W. Fletcher, R. H. (1993). Characteristics of peer reviewers who produce good reviews. In Programs and Abstracts of Second International Congress on Peer Review in Biomedical Publications (p. 11). Chicago, IL.

Fraser, N. (1992). Rethinking the public sphere: A contribution to the critique of actually existing democracy. In C. Calhoun, Ed., Habermas and the public sphere (pp. 109-142). Cambridge, MA: MIT Press.

Freudenheim, M. (2000, May 30). New web sites altering visits to patients. New York Times, pp. A1, C14.

Ginsparg, P., Luce, R., and Van de Sompel, H. (1999). First meeting of the Open Archives initiative, Santa Fe, New Mexico, US, October 21-22. Available: http://www.openarchives.org/ups1-press.htm.

Glass, G. V. (2000). Anonymous FTP: Can we risk it? Dare we publish pre-prints? Paper presented at the annual meeting of the American Educational Research Conference, New Orleans. Available: http://glass.ed.asu.edu/gene/papers/aera2000.

Guernsey, L. (2000, July 18). Books by the chapter or verse arrive on the internet this fall. New York Times, p. A1, C6.

Harnad, S. (1997). Learned inquiry and the net: The role of peer review, peer commentary and copyright. Beyond print: Scholarly publishing and communication in the electronic environment, University of Toronto. Available: http://citd.scar.utoronto.ca/Epub/talks/Harnad_Snider.html.

Hjørland, B. (1998). The classification of psychology: A case study in the classification of a knowledge field. Knowledge Organization, 25(4), 162-201.

Illich, I. (1997). In the vineyard of the text: A commentary to Hugh's Didascalicon. Chicago: University of Chicago.

Johnson, R. (2000 May). A question of access: SPARC, BioOne, and society-driven electronic publishing. D-Lib Magazine, 6(5). Available: http//:www.dlib/may00/Johnson/05johnson.html.

Judon, H. F. (1994). Structural transformation of the sciences and the end of peer review. Journal of the American Medical Association, 272(2), 92-94.

Kaser, R. T. (2000, May). If information wants to be free. . .Then who's going to pay for it. D-Lib Magazine. Available: http://www.dlib.org/dlib/.

Krechmer, K. (1996 January). Recommendations for the global information highway: A matter of standards. Information Standards Quarterly, 8(1), 1-5.

Lageman, E. C. (1989). The politics of knowledge: The Carnegie Corporation, philanthropy, and public policy. Chicago, IL: University of Chicago Press.

Leavy, A. M. & Ganesh, T. G. (2000, February 1). The nascent promise of ejournals: Instances of pioneering use of technology. Current Issues in Education [On-line], 3(1). Available: http://cie.ed.asu.edu/volume3/number1/.

Logoze, C. and Fielding, D. (1998, November). Defining collections in distributed digital libraries. D-Lib Magazine. Available: http://www.dlib.org/dlib/.

Lubchenco, J. (1998). Entering the century of the environment: A new social contract for science, Science, 279, 495-501.

Markoff, J. (2000, July 20). High-tech executives urge action on world's digital divide. New York Times, p. A6.

Oldenburg, H. (1665, March 6). Introduction. Philosophical Transactions, p. 1.

Olsen, F. (2000, June 30), Two digital libraries in California form partnership to expand services. Chronicle of Higher Education, p. A41.

Robinson, J. (2000). Turner plans $1.2 billion expansion and more CNN Online. New York Times, p. C7.

Rosenbaum, R. (2000, August 6). The play's the thing, again. New York Times Book Review, pp. 12-13.

Scholars' Forum Discussion (SFD). (1999). Attached to Buck, A.M., Flagan R. C., and Coles, B. Scholars' Forum: A new model for scholarly communication. California Institute of Technology. Available: http://library.caltech.edu/publications/scholarsforum.

Schulenburger, D. E. (1998). Moving with dispatch to resolve the scholarly communication crisis: From here to NEAR. Association of Research Libraries Proceedings. Available: http://www.arl.org/arl/proceedings/133/schulenburger.html.

Shirky, C. (2000, July 15). Freedom, one song at a time. New York Times, p. A27.

Smith, J. W. T. (2000). The deconstructed journal – a new model for academic publishing. Learned Publishing, 12, 79-91.

Smith, M.C. (1994). Social Science in the crucible: The American debate over objectivity and purpose, 1918 - 1941. Durham, NC: Duke University Press.

Sosteric, M. (1999, February). ICAAP extended Markup Language: Exploiting XML and adding value to the journal production process. D-Lib Magazine, 5(2). Available: http:// www.dlib.org/dlib.

Tennant, R. (2000, May 15). Accessing electronic theses: Progress? Library Journal, pp. 30-31.

Van de Sompel, H. et al. (2000). The UPS prototype: An experimental end-user service across e-print archives. D-Lib Magazine, 6(2). Available: http:// www.dlib.org/dlib.

Van de Sompel, H. and Hochstenbach, P. (1999). Reference linking in hybrid library environments. Parts 1-3. D-Lib Magazine, 5(4), 5(10). Available: http://www.dlib.org/dlib.

Van de Sompel, H. and Lagoze, C. (2000) The Santa Fe Convention of the Open Archives Initiative. Available: http://www.dlib.org/dlib/february00/vandesompel-oai/02vandesompel-oai.html.

Veltman, K. H. (1997). Frontiers in conceptual navigation. Knowledge Organization, 24(4), 225-245.

Vrasidas, C. (2000, July 13) Promises of electronic forms of data representation and scholarly publication. Teachers College Record, Available: http://www.tcrecord.org/.

Willams, R. (1961). The long revolution. London: Chatto and Windus.

Willinsky, J. (2000). If only we knew: Increasing the public value of Social Science research. New York: Routledge.

Willinsky, J. (1999). Technologies of knowing: A proposal for the human sciences. Boston: Beacon.

Notes

1.  I wish to acknowledge and thank Miriam Clavir, Vivian Forssman, Keith Fuller, Gene V Glass, Anne White, and Larry Wolfson for helpful suggestions on improving this article. This work has been supported by the Social Sciences and Humanities Research Council of Canada.

2.  The 121 ARL libraries spend over $75 million a year on an average of 378 Elsevier Science titles alone. Overall, Mary Case, director of ARL's Office of Scholarly Communication, reports that "since 1986, while ARL libraries have canceled hundreds of thousands of dollars worth of journals, they have spent 124% more on serials to purchase 7% fewer titles. During the same period, expenditures on monographs have increased 29% and ARL libraries have purchased 21% fewer titles" (Case, n.d.).

3.  I have discussed the persistent sense of knowledge's excess, going back to King Solomon, in (author, 2000).

4.  Similarly with scholarly books, the author's royalty is not directed at recovering the actual research costs. Although the research is the result of what is legally termed work-for-hire making it, technically, the intellectual property of the sponsor, traditionally researchers retain the right to publish the work with or, in the case of journals, without royalties.

5.  On how the university community may have helped "perpetuate the myth that information wants to be free by making it look free," see Richard T. Kaser (2000).

6.  Fathom represents the University of Chicago, RAND, the American Film Institute, the Woods Hole Oceanographic Institution, the London School of Economics and Political Science, Cambridge University Press, the British Library, the New York Public Library, the Smithsonian Institution, and Columbia University (Carlson, 2000).

7.  I would not argue the originality of any single aspect of this model. For example, Kenneth Arnold, former head of Rutgers University Press, proposed as early as 1991 that "university presses make common cause with campus libraries and computer centers to develop university-based publishing nodes that could take advantage of the Internet and short-run publishing technologies" (1996, p. 387).

8.  On recent government and private initiatives aimed at the "world's digital divide," see (Markoff, 2000). Of course, the hi-tech industry stands to benefit by closing the divide, but before that the publishers stood to benefit from a spread of knowledge which took place, however, on a far smaller scale. Any new system, I am holding , should not just increase research's reach but improve its value.

9.  The Public Knowledge Project website: http://www.educ.ubc.ca/faculty/ctg/pkp/pk.htm.

10.  On the corresponding tradition, Paul Ginsparg: "As a matter of historical fact, the LANL archives owe *all* of their initial growth starting in '91 precisely to the daily e-mail alerts of new material received (according to subject matter). While one might have expected the later web access to replace this in some way, it is a matter of empirical fact that users of the archive insist on retaining the e-mail alert as an essential component." (Contribution to Follow-Up Discussion, Scholars' Forum, 05/02/99, http://library.caltech.edu/publications/scholarsforum).

11.  The commercial side of this has translated into the "disaggregation" of books on the web, which are now being sold by the chapter online, and a rejection of bundled music "album" (Gurensey, 2000; Shirky, 2000).

12.  To hear that "electronic journals revolutionize the way researchers report their work" - because the articles are interactive, multimedia, multi-linguistic, longer, and include the data - is to wonder whether it is not literally more of the same (Vrasidas, 2000, p. 1). It may be worth noting that Charalambos Vrasidas' online article in Teachers College Record is identified in a very un-journal-like-way with the day it was published, an ID number and the date I accessed it, rather than volume and number, suggesting an end to the bundling of a volume of articles.

13.  Metadata refers to the data embedded in a document that describe the document, such as author, title, date composed (see Table 1).

14.  The Open Archives initiative is supported by the SFX system which enables readers to readily click their way from one type of resource to another, from abstract service to Amazon.com, providing for each reference a range of possible related sources to check, which is designed to work with a Universal Preprint Service (Van de Sompel and Hochstenbach, 1999; Van de Sompel et al., 2000).These technologies demonstrate an integration of the different database systems that have developed begging the question of what a more thorough rethinking of the entire system might offer.

15.  Harnad: "The LANL principle has face validity, overwhelming proof in practice, and is ready for extension to all disciplines. Why take something we KNOW works stupendously well and weigh it down with unproven 'add-ons' that could well prevent it from getting off the ground just when it's ready to carry the other disciplines skyward, just as it has done in Physics?" (SFD, 05/01/99).

16.  Rob Kling, a professor of information systems and information science at Indiana University at Bloomington: "They have a Field of Dreams rhetoric. Their particular model doesn't fit all dreams" (Kiernan, 1999, p. A44).

17.  The Scholars' Forum shares a number of similarities with John W. T. Smith's Deconstructed Journal model which would dispense with publishers and substitute Subject Focal Points for journals which would filter, review, and organize studies relevant to its subject area (Smith, 1999).

18.  International efforts, involving USA, Australia and Germany are currently underway to establish an XML standard for electronic theses and dissertations with the intent of offering free access (Tennant, 2000). The UN-chartered International Telecommunications Union (which evolved out of the International Telegraph Union that dates back to 1865) is overseeing the work of governments and businesses to create international information and communication standards for, among other things, the Global Information Highway (Krechmer, 1996).

19.  This is from the speech of the Association's then president, Jane Lubchenco, a zoologist from the University of Oregon (1997, p. 495).

20.  The issue has been identified by philosopher Nancy Fraser as a "lack of participatory parity" for "subaltern counterpublics" as characterizing the current deliberative public sphere in "actually existing democracy" (1992, pp. 121, 123). The scholarly desire to help the dispossessed has been noted, for example, by Linda Charnes, professor of English at Indiana University, who was recently quoted as asking, "We all avow we speak for oppressed voices of class, race, gender and nationality; we don't 'interpret,' we 'intervene'... but is this all we have to offer as critics?" (Rosenbaum, 2000, p. 13).

21.  MEDLINEplus Health Information (http://www.nlm.nih.gov/medlineplus/) is run by the federally funded National Institute of Health.

22.  The Cornell Digital Library Research Group has shown great leadership in this area. Deploying a series of user-interface, collection, naming, index, and repository servers, it has built a test-bed digital library of 24,000 computer science reports distributed across 120 institutions (Logoze and Fielding, 1998).

23.  For a discussion of rationalizing the multiple and overlapping dimensions of such knowledge classification, using psychology as its example, see Birger Hjørland (1998).

24.  For example, the embedded XML tag <p xml:lang="en-US"> could indicate that the article's language is U.S. English. See http://www.xml.com

25.  The impact of space limitations on peer review quality has become a source of concern in recent times (Judson, 1994). For examples of the online publication of data enabling readers to work with the data, leading to its productive reanalysis, see Aisling M. Leavy and Tirupalavanam G. Ganesh (2000).

26.  On the other hand, at least one study in the biomedical field suggests that the quality of review is inversely related to seniority and standing of the reviewer (Evans et al., 1993).

27.  See note #10.

28.  One instance of this sort of technology is SFX, a generic linking system, which is being deployed as part of the Open Archives initiative; it provides an intelligent guide to available resources in relation to specific documents (Van de Sompel and Hochstenbach, 1999; Van de Sompel et al., 2000).

29.  See Dublin Core Metadata (http://purl.oclc.org/dc/); IMS Global Learning Consortium Inc (http://www.imsproject.org) and ICAAP (Sosteric, 1999).

30.  For recent work on such forms of "conceptual navigation," see Kim H. Veltman (1997).

31.  In recommending that public access to government information be cross subsidized by commercial users, David Danner and Paul W. Taylor, senior policy analysts with Washington State, review the legal support for such differential structures (1997).