Online Databases: Open Access Alternatives

From the ongoing discussion in the online version of Nature, to articles in the Chronicle of Higher Education and even USA Today, open access publishing is a hot topic. But open access publishing can have many different definitions, and pros and cons vary with the definitions.

The University of Lund, Sweden, in its Directory of Open Access Journals (DOAJ), which now includes almost 1100 titles, defines open access journals as "journals that use a funding model that does not charge readers or their institutions for access." This definition builds on the Budapest Open Access Initiative (BOAI) demands, which include the right of "users to read, download, copy, distribute, print, search, or link to the full texts of these articles." DOAJ also takes a stand against publisher embargoes. To be included in the directory, "journal[s] should offer open access to their content with no delay" (www.doaj.org). The journals in DOAJ cover all topics, with science, technology, and medicine representing about half.

A broad definition includes many publication and distribution schemes. E-journals that are published, distributed electronically, and subsidized by universities, government agencies, and volunteer organizations are the most common. In addition, collections of separate articles or research reports could fit the definition, including e-print servers such as arXiv.org, institutional repositories, and author web pages. These collections are not covered in DOAJ, so in reality access to scholarly articles goes beyond the 1100 listed.

The Open Archives Initiative (OAI; www.openarchives.org) encourages authors and institutions to make articles freely available in a standard OAI format. These articles may be preprints, copies of articles published in traditional peer-reviewed journals, or material distributed only via the web.

More common in the popular press is the addition of the word publishing as a qualifier. Most descriptions of open access publishing imply, in addition to no cost to users, a price model where the author or author's institution pays for an accepted article. Here the process of publishing bears a cost, while use is free. Although author page charges are not a new phenomenon, Public Library of Science (PLoS) and BioMed Central are two well-known new examples. Both have been heavily subsidized by grants, including a $9 million, four-year grant to PLoS from the Moore Foundation. Many debates focus on the sustainability of these systems.

For a standard "article processing" fee of $525 (and between $1000 to $1500 for a few journals), BioMed Central will make peer-reviewed articles openly available within the journals it publishes (www.biomedcentral.com). PLoS is newer, with PLoS Biology starting in 2003 and PLoS Medicine in 2004. Both journals charge a fee of $1500 per published article. Print subscriptions are available to individuals or institutions for $160 per year.

In lieu of payment, BioMed Central also offers institutional "memberships" to universities with many potential authors. In Finland, the membership agreement with FinELib, the National Electronic Library of Finland, covers the cost of publication for all of Finland's 25,000 publicly funded researchers and teachers.

Open access publishing is especially attractive to companies and small colleges or universities that are likely to have many more readers than authors. A downside is that a membership fee sounds suspiciously like a subscription fee. Some big universities worry that their fees are an unfair burden, forcing them to pay for open access by others. Some are concerned that author fees will come out of the library budget. Scientists in developing countries worry that without subsidies they will be less able to publish and the topics of interest in their nations will be less likely to be represented. Research and scholarly publishing have costs (although estimates of the exact per article costs vary widely), whether volunteers, institutions, authors, or libraries pay.

Open access journals are having an impact. Web of Science selected 191 (out of 8700 total). According to Thomson ISI, only the top 10–12 percent of 2000 journals evaluated yearly are accepted for inclusion in Web of Science. The rapid addition of so many open access journals is striking. A study by ISI shows "no discernible difference in terms of citation impact or frequency with which the journal is cited" between traditional and open access journals (www.isinet.com/oaj).

Journal publishing and payment models are in flux. Author payments, membership fees, institutional commitments to repository development, and self-archiving in e-print servers or other web sites coexist with lease agreements and traditional subscriptions. No one answer is a panacea, capable of solving library budget woes, access to high-quality literature, and collection development issues. But neither is it time to throw out any of the options as we work to find the best models for libraries and scholarship.