Key Open Access Concepts

This section provides a brief introduction to some of the major open access concepts needed to fully utilize the bibliography. It is not intended as a complete guide to open access (see articles in the "1.1 Overviews" section of the bibliography for such introductions).

Open Access Defined

The "Scope of the Bibliography" section provided an excerpt from perhaps the most influential open access statement, the "Budapest Open Access Initiative." The important things to note in the basic definition are that open access deals with peer-reviewed articles or preprints and that free access to these works is not equivalent to open access. Open access also requires no restrictions on how published material is subsequently used except to require that proper attribution of the work be given to the author and that authors retain control over the integrity of their work.

In practice, what makes open access possible is that it is relatively inexpensive to distribute electronic articles on the Internet. Consequently, the open access movement focuses on various electronic publishing strategies.

While the BOAI provides a landmark definition of open access, other groups' statements have somewhat different views, and the concept of open access is still evolving. For example, the "Bethesda Statement on Open Access Publishing"1 also includes a requirement that:

A complete version of the work and all supplemental materials, including a copy of the permission as stated above, in a suitable standard electronic format is deposited immediately upon initial publication in at least one online repository that is supported by an academic institution, scholarly society, government agency, or other well-established organization that seeks to enable open access, unrestricted distribution, interoperability, and long-term archiving (for the biomedical sciences, PubMed Central is such a repository).

In spite of an emphasis on peer-reviewed articles in key statements, open access can be applied to a variety of scholarly works that are produced by scholars without the expectation of payment. Although this is likely to be a later development, Peter Suber has suggested that it may even be applied with greater difficulty to scholarly materials that do involve author payment or profit.2

The BOAI statement suggests two strategies for achieving open access: self-archiving and open access journals.


Self-archiving can be achieved in at least three ways: (1) putting articles on author Web sites, (2) depositing articles in disciplinary archives, or (3) depositing articles in institutional archives and repositories.

Copyright Concerns

Self-archiving is contingent on authors having the legal right to electronically distribute their articles. This is challenging because many scholars relinquish their copyrights to publishers and, historically, many journal publishers did not want to consider articles that had been distributed as electronic preprints for publication. Since the late 1980's, there has been a growing trend for authors to want to retain their article copyrights. This is primarily the result of the advent of free scholar-produced journals (which often let scholars retain copyright), an increased awareness of copyright issues as a result of the serials crisis, the vigorous self-archiving advocacy efforts of Stevan Harnad and others, and the rise of the open access movement. Authors who retain their copyrights can then grant publishers the limited rights that they need to effectively distribute their works and/or they can put their works under a license that grants certain rights to all potential users and distributors.

A barrier to author granting of rights has been that framing the proper wording of license agreements is a complex process requiring significant legal knowledge. The Creative Commons has greatly facilitated the use of author license agreements by developing a variety of standard agreements that authors can easily select and utilize.3 The Creative Commons Attribution license meets open access requirements.

In recent years, publishers have been more open to considering preprints for publication, granting authors the right to archive their works, and allowing them to retain copyright if they request it.


Self-archived articles may be preprints (i.e., draft articles that have not been peer-reviewed or edited) or postprints (i.e., final, edited versions of peer-reviewed articles). The term used for both is "e-prints." Certain scientific disciplines, such as physics, have a long history of e-print distribution. Initially, this was done by scholars mailing colleagues preprints (or reprints). Later fax was used, then e-mail.

Author Web Sites

While some authors archived articles on FTP or Gopher sites, it was the widespread utilization of the Web starting roughly in the mid-1990's that resulted in a significant growth of personal e-print archives. A key problem with such archives is that they can be unstable, as authors move from institution to institution, retire, make other life changes, or die. As will be seen later, e-prints from such archives are not made as easily visible to the research community as those in disciplinary archives or institutional archives and repositories because they cannot be easily harvested.

Disciplinary Archives

In the early 1990's, formal "disciplinary archives" began to displace scholar-to-scholar distribution in some scientific disciplines. A disciplinary archive provides access to e-prints for one scholarly discipline or multiple scholarly disciplines. Keep in mind that some disciplines have many subfields, and that specialties that draw on many traditional disciplines are increasingly common. Some disciplinary archives provide access to diverse scholarly works, not just e-prints. Usually, a disciplinary archive can be searched and browsed.

The most famous disciplinary archive is probably arXiv, which covers physics, mathematics, non-linear science, computer science, and quantitative biology.4 It was established in 1991.

It is important to keep in mind that some disciplines rely more heavily on articles than others, and that some disciplines that rely heavily on articles do not have a strong tradition of using e-prints. Consequently, there can be significant disciplinary differences in receptiveness to open access.

Institutional Archives and Repositories

Where disciplinary archives provide access to the worldwide literature of one or more fields, institutional archives and repositories focus on the literature produced by a single institution.

An institutional e-print archive may contain e-prints written by scholars from many departments, research centers, or other units. Or, it may only contain the e-prints of a single unit.

An institutional repository includes a variety of materials produced by scholars from many units, such as e-prints, technical reports, theses and dissertations, data sets, and teaching materials. Some institutional repositories are also being used as electronic presses, publishing e-books and e-journals. DSpace at MIT is a notable example of an institutional repository (the DSpace project began in 2000).5

Typically, e-print archives and institutional repositories can be searched and browsed.

Freeware is usually used to support these efforts. Popular choices include EPrints6 for institutional e-print archives and DSpace7 or Fedora8 for institutional repositories. (The popular EPrints software has also been used for disciplinary archives and institutional repositories.)

Open Access Journals

After self-archiving, the second major BOAI strategy is open access journals. Open access journals allow authors to retain their copyrights, but may require that they agree to license their articles with the Creative Commons Attribution license or a similar license.

Open access journals are primarily electronic journals (print editions are sometimes offered as an optional fee-based add-on). Once the first electronic copy of a journal has been created, the costs of distributing it on the Internet are negligible compared to the costs of distributing additional print copies of a conventional journal. Open access advocates also note other cost savings implicit in their approach, such as the elimination of the need for access controls. Still, open access journals cost money to produce and distribute, especially since they are peer-reviewed and edited like conventional journals. Various funding strategies are in use, but the most common are direct author fees, institutional memberships to sponsor all or part of author fees, funding agency payment of author fees, grants to open access publishers, institutional subsidies (such as paying the salaries of journal editorial staff), and priced add-ons (such as recommendation services, current awareness services, or print editions).

Open access journals may be included in index and abstract services. The Directory of Open Access Journals is a major finding tool, which permits searching at the article level for some journals.9

Preliminary research suggests that the "impact" factors of open access journals can be at least as good as those of conventional journals.10

Three organizations play a major role in the publication and archiving of open access journals: BioMed Central, the Public Library of Science (PLoS), and PubMed Central.

Established in 2000, BioMed Central is a for-profit publishing company that publishes over 100 open access biomedical journals.11

The Public Library of Science is a nonprofit organization that, as of August 2004, publishes one open access journal (PLoS Biology).12 A second journal (PLoS Medicine) is expected to be launched in October 2004. The PLoS started in 2000. It first activity was to circulate an open letter that was intended to convince biomedical publishers to make their journals freely available within six months of publication. Roughly 34,000 scientists from 180 countries ultimately signed the letter, pledging not to publish in (or otherwise support) journals that did not meet this requirement by September 2001. When this letter did not invoke the desired response, the Public Library of Science began to publish its own open access journals.

PubMed Central is a freely available life sciences journal archive that is run by the National Center for Biotechnology Information of the National Library of Medicine.13 Journals must meet certain editorial standards to be included in the voluntary archive. As originally conceived in 1999 by Harold Varmus (who was then the Director of the National Institutes of Health), PubMed Central (then called E-biomed) had a broader mission that included e-prints; however, this original vision was very controversial, and it was significantly modified by PubMed Central's launch in 2000.

Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH)

Since open access works are scattered across many disciplinary archives, institutional e-print archives, institutional repositories, and open access journals, it can be difficult for scholars to locate all needed works on a particular subject. It requires scholars to search one system after another in a serial fashion. To deal with this problem, the Open Archives Initiative14 developed the Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH) to allow search systems (called service providers) to retrieve metadata about open access works from archives and repositories (called data providers) and aggregate this data so that it can be searched with a single query. The first version of OAI-PMH was released in 2001.

There are several notable service providers that can help users locate relevant e-prints. The University of Michigan Digital Library Production Service's OAIster service harvests metadata from over 340 data providers and provides unified searching of this metadata.15 The Digital Library Research group at Old Dominion University maintains a similar, smaller scale experimental service called Arc.16 The Open Citation Project's experimental Citebase service uses citation ranking to enhance search result displays.17 The Open Archives Initiative maintains a list of other service providers.18 Major search engines, such as Google and Yahoo, have begun to index metadata from various data and service providers through cooperative projects. These projects supplement existing efforts by search engines to index e-prints, allowing them to overcome special technical problems associated with indexing some of these items.

Government Inquiries and Legislation

There has been increased government scrutiny in the U.K., the U.S., and the European Union of the conventional scholarly publishing system, with particular attention being paid to the perceived high cost of scientific, technical, and medical journals and the fact that much research in these areas is government funded.

In the U.S., legislation was introduced in 2003 that would put works that are "substantially funded" by government money into the public domain (the "Public Access to Science Act," informally called the "Sabo bill" after its sponsor, Representative Martin O. Sabo).19

In 2004, the UK House of Commons Science and Technology Committee concluded a significant inquiry into scientific publishing that resulted in a report (Scientific Publications: Free for All?).20 This report recommended that articles resulting from government-funded research be deposited in institutional repositories, which would be established at all UK higher education institutions, and that funds be made available to pay open access journal publication fees for such articles (authors would need to apply for these funds).

In 2004, the U.S. House Appropriations Committee recommended that articles that result from NIH grant-funded research be deposited in PubMed Central upon acceptance for publication.21 If NIH funds were used to support any publication costs, the articles would be made immediately available. Otherwise, they would be made available six months after publication. NIH would develop a plan by 12/1/04 to implement the recommendation in FY 2005.

Also in 2004, the European Commission announced it would conduct a major study of the scientific publication markets in Europe, which would be completed in 2005.22

Developing Countries

While not strictly an "open access" strategy, an important closely related effort by traditional publishers and others has been the creation of special journal access arrangements for developing countries, whose scholars may otherwise have very limited access to frequently expensive journals. These arrangements provide free or reduced cost access to journals. Major initiatives include the Access to Global Online Research in Agriculture (AGORA)23 and the Health InterNetwork Access to Research Initiative (HINARI),24 which provides access to biomedical journals. On a smaller scale, the novel Ptolemy Project provides medical researchers and clinicians in the developing world with free access to licensed e-journals and other e-resources by making them research affiliates of the University of Toronto.25

Open access is taking root in developing countries, and, in addition to other benefits, it provides a way to increase the visibility of research from these countries. For example, Scientific Electronic Library Online (SciELO) is an innovative scientific electronic publishing cooperative that focuses on providing open access to Latin America and Caribbean journals.26


1. "Bethesda Statement on Open Access Publishing," 20 June 2003,

2. Peter Suber, "Creating an Intellectual Commons through Open Access," 2004,







9. Lund University Libraries, Directory of Open Access Journals,

10. James Testa and Marie E. McVeigh, "The Impact of Open Access Journals: A Citation Study from Thomson ISI," 2004,








18. Open Archives Initiative, "Registered Service Providers,"

19. Peter Suber, "Martin Sabo's Public Access to Science Act," SPARC Open Access Newsletter, no. 63 (2003),

20. House of Commons Science and Technology Committee, Scientific Publications: Free for All? (London: Science and Technology Committee, House of Commons, United Kingdom Parliament, 2004),

21. Peter Suber, "NIH Open-Access Plan: Frequently Asked Questions," 2004,