University Times

Open access publishing has been in the limelight as recent developments have encouraged more free publication of scholarly literature.

This is the first in an occasional series on the impact of the digital age on academic publishing.

The National Institutes of Health’s policy requiring NIH-funded researchers to submit their work to its PubMedCentral archive upon acceptance for publication went into effect April 7. Additionally, as of May 25, the NIH will require investigators to include a PubMedCentral reference number in applications, proposals or progress reports that arise from NIH-funded work they author or co-author.

A move by Harvard University’s arts and sciences faculty in February to give the university license to post copies of their published works online stands to further open the door to free open access to Harvard professors’ scholarly publications.

Pitt isn’t far behind. An open-access University repository into which Pitt scholars could voluntarily post their scholarly work is being developed and is expected to be in place by fall, said University Library System director and Hillman librarian Rush G. Miller. “I’m sure a number of universities will follow suit.”

While open access initiatives are gaining momentum, the concept is not new. The advent of the Internet invited such dissemination of information, long ago prompting discussions in the scholarly community on how to address the new technology.

Pitt Provost James V. Maher and Miller were among the participants at a 2000 conference on scholarly publishing in Tempe, Ariz., that yielded guidelines that would hold even as new ideas for systems of scholarly publishing arose. These have become known as the Tempe Principles. (See box.) The nine principles and additional information are available at www.arl.org/resources/pubs/tempe.

Maher presented a brief history of the issues in light of the Tempe Principles and gave an overview of Pitt’s initiatives at the April 1 Faculty Assembly meeting.

By way of background, Maher said rising journal costs, tight budgets and the advent of evolving digital publishing technologies created a “perfect storm” that came to a head in the early 1990s, prompting moves toward a reorganization of scholarly publishing.

ULS’s Miller confirmed that rising costs of journals — with increases averaging about 10 percent a year — put the squeeze on university libraries, which were forced to cut subscriptions or divert money from book budgets to maintain journal subscriptions.

And, as libraries halted subscriptions, commercial journal publishers raised prices to maintain their profits, creating a cycle that fed upon itself.

Miller said librarians were talking about the “serials crisis” when he entered the field more than 30 years ago, and today “it’s still as much of a crisis as it ever was.”

The system of scholarly publishing is a big circle: A faculty member gets an idea for a research project. The professor is a university employee who uses university resources — computers, libraries and perhaps travel grants — to conduct the research that is submitted as an article to a publisher. The publisher adds value by editing it, managing peer review and by printing and distributing the work. The publisher then sells the journal back to the university libraries that make up the bulk of their subscribers.

“The market is the same people that paid to produce it,” Miller said. “We created and funded it and gave it away to the publisher who then sells it back at an exorbitant rate,” he said. “That’s what’s breaking apart.”

While some journal publishers have been responsible about their pricing policies, others increased costs to whatever the market would bear, backing some budget-strapped librarians into a corner.

Although Pitt has maintained its collections, “Many universities just gave up,” Miller said. As a result, many faculty and students found their access to materials sharply curtailed. “It became clearly noticeable that scholars in many universities did not have access to the basic tools they need to conduct their research,” Miller said.

Information technology gives access to more journals, but most online journals still aren’t free. At Pitt, users with University accounts have access to thousands of journals because, like many other schools, Pitt pays a license fee based on the number of users. Outsiders can’t use the online collections except in limited circumstances.

Miller said the Internet opens up new ways of distributing knowledge, creating concerns in the scholarly community. For instance, anyone could post scholarly material online without peer review, one of the biggest concerns in the academic community. While it might be acceptable in some disciplines, posting material without prior review could be harmful in others, such as medicine or other sciences. “It’s a very dangerous idea,” Miller said, noting that it was among the issues discussed at Tempe.

The group addressed issues of cost, access and fair use as well as the issues of maintaining archives and ensuring data users’ privacy.

The principles, which Miller said have stood up well over time, were developed despite the fact that it’s still unclear exactly what forms the distribution of scholarly knowledge will take.

The document sets forth principles so that whatever system evolves, “certain things remain inviolate,” he said.

Maher noted that Pitt has done well in several areas addressed at Tempe. Among the principles are that electronic capabilities should be used to provide broad access to scholarship and be securely archived so they remain available.

Beginning in 2001, Maher said, Pitt began putting doctoral dissertations online and since 2004 Pitt has been posting dissertations digitally, giving scholars worldwide the opportunity to view Pitt doctoral students’ work from wherever they are. Previously, the bound dissertations could languish on a library shelf to be opened perhaps once or twice, if at all, by someone who came to the library in person. Pitt’s database of 2,000 dissertations has been accessed a half-million times, Maher said, adding that many of those hits are from international users. “We do think the dissertations from Pitt are having an impact on the scholarly world,” he said.

Pitt also hosts discipline-based electronic archives of scholarly publications for philosophy of science, minority health, European integration and aphasiology, which are available to anyone with access to the Internet. Maher said most major universities are trying to do their part by hosting at least one open access archive; he knew of no other schools hosting as many as Pitt does.

In addition, ULS is partnering with the University of Pittsburgh Press to publish digitally a number of titles from the press’s Latin American series.

Progress has been slow, Maher said — not just at Pitt — in pressing journal publishers to support making scholarly work available at reasonable costs and to grant copyright agreements that give faculty access to their own works.

Some publishing contracts prevent professors from being able to use their own work in the classroom, Maher noted, adding that the University has developed revised contracts with more favorable provisions for authors that faculty can insist publishers use instead of the standard agreements.

Maher noted that it is difficult for universities to pull together to force change on publishing companies for fear of antitrust lawsuits; publishers have demonstrated a willingness to sue, he said. Instead, the battle is being fought one faculty member at a time.

The new NIH requirements strike a compromise that takes into account commercial publishers’ concerns alongside a desire to make research paid for by the American public available to the public. While authors must submit their work to PubMedCentral, the journals are given a year to market their product before the submission becomes available in the open access archive.

Among publishers’ arguments against open access is that it will decrease their subscriber base. Miller said librarians disagree with that argument, noting that the currency of journal information is what’s of most value to faculty and that he knows of no library that would cancel a journal subscription in order to wait for the material to become freely available later.

Also, an article submitted for publication typically doesn’t compete with the version that is published because the finished product has been polished and edited.

Looking ahead, Miller said he believes open access publishing will become the norm for most disciplines, although what form that will take is not yet clear.

He expects the NIH policy will have an impact on all research, noting that some publishers already have begun offering to submit journal articles to PubMedCentral on behalf of authors — making themselves more attractive by relieving authors of one more burden. He envisions that publishers may expand the practice to include submitting articles to a repository of the author’s choice, creating a different sort of competition.

Miller noted that journal publishers do contribute important value-added services for authors — managing peer review and editing, for instance. For that reason, he believes scholarly publishers will remain viable players even as open access models come to dominate the field.

“Whatever emerges, there’s still going to be billions of dollars being spent,” he predicted.

Tempe Principles

1. The cost to the academy of published research should be contained so that access for faculty and students can be maintained and expanded.

2. Electronic capabilities should be used to provide wide access to scholarship, encourage interdisciplinary research, and enhance interoperability and searchability.

3. Scholarly publications must be archived in a secure manner so they will remain permanently available.

4. The system of scholarly publication must continue to include processes for evaluating the quality of scholarly work.

5. Copyright should be managed in a way that assures faculty access to and use of their own published works in their research and teaching.

6. In negotiating publishing agreements, faculty should choose journals that support the goal of making scholarly publications available at reasonable cost.

7. The time from submission to publication should be reduced in a manner consistent with the requirements for quality control.

8. The evaluation of faculty should place a greater emphasis on quality of publications and a reduced emphasis on quantity.

9. In electronic as well as print environments, scholars and students should be assured privacy with regard to their use of materials.

Adapted from the Association of Research Libraries “Principles for Emerging Systems of Scholarly Publishing,” www.arl.org/bm~doc/tempe1.pdf

—Kimberly K. Barlow