1.1. What are eprints?

An eprint is an 'electronic publication', usually an electronic copy of a research article. Eprints are usually either 'pre-prints' (a copy of an article as submitted for peer-review) or 'post-prints' (the final draft of an article after peer-review and as accepted for publication in a journal). The significant element is 'peer-review'. The eprints movement assumes that the primary aim behind publishing articles is the sharing and dissemination of quality research. Participating in the eprints process is relevant to researchers in all disciplines: humanities, social sciences,physical, environmental and medical sciences. Support and management of the eprints process is especially relevant for library and computing services.

An eprint archive is a central repository of eprints, whether pre- or post-prints. Often an eprint archive will contain significant drafts which track the revision process from initial submission to final accepted draft. An archive may also contain supporting materials such as research data, images etc. relevant to an article. On the other hand, an eprint archive may choose only to contain digital copies of published, peer-reviewed articles.

The process of depositing an eprint is often referred to as 'self-archiving'. The process includes storing a copy of the article in a repository together with information about the article (also known as metadata and likely to include author, title, date of publication/revision, journal title etc).

Once an eprint is in an archive or repository then potentially it is sharable with others, whether researchers, casual browsers or other archives. Eprint archives increasingly 'talk' to one another and share metadata about their holdings using a protocol called the Open Archives Initiative Metadata Harvesting Protocol (OAI-MHP). Exposing and harvesting metadata using the OAI-MHP allows for the building of new services by aggregating metadata from distributed archives based on specific criteria (e.g. subject, geographic location, type of eprint etc).

1.2. What is an institutional repository?

An institutional repository is an eprint archive (or collection of archives) which holds a collection of research outputs from members of that institution. An institutional repository will often be managed within the library services, will have the support of senior management, and be available online for the benefit of both that institution and the wider academic community. Software exists for easily developing eprint archives (even individuals can do it). The real effort for an institution lies in raising awareness about the benefits of participating and managing the process and policies for deposit, access, maintenance and preservation. The Budapest Open Access Initiative has published a useful Guide to Institutional Repository Software and the Scholarly Publishing and Academic Resources Coalition (SPARC) maintains a set of resources relating to institutional repositories. SPARC also has a branch in Europe of which a number of UK institutions (including Oxford) are members.

Existing institutional archives (most in a pilot stage) include:

• Oxford University pilot eprints repository, http://eprints.ouls.ox.ac.uk/
• University of Cambridge, http://www.lib.cam.ac.uk/dspace/ (in development from Jan 2003)
• University of Nottingham, http://eprints.nottingham.ac.uk/
• University of Glasgow, http://eprints.lib.gla.ac.uk/
• University of Strathclyde, http://eprints.cdlr.strath.ac.uk/
• University of California Digital Repositories, http://repositories.cdlib.org/
• MIT, http://libraries.mit.edu/dspace/
• Université de Montréal, http://papyrus.bib.umontreal.ca/
• Australian National University, http://dspace.anu.edu.au/handle/1885/44441
• India Institute of Science, http://eprints.iisc.ernet.in/

The ePrints UK Project harvests metadata from about 30 institutional repositories (and see also the OAI registry of repositories at University of Illinois at Urbana-Champaign.

Institutional repositories should normally be underpinned by an institutional policy on self-archiving. An example of such a policy is provided by Electronic & Computer Science Department, Southampton (http://www.ecs.soton.ac.uk/~lac/archpol.html)

1.3. What about subject repositories?

Subject repositories are eprint archives which collect and manage eprints relating to one or more related subject areas. A number currently exist, mainly within science subjects. Significant subject repositories include:

• ArXiv - http://xxx.arxiv.cornell.edu/ (physics, mathematics, non-linear science and computer science)
• Cogprints - http://cogprints.ecs.soton.ac.uk/ (Cognitive sciences including psychology, neuroscience, linguistics and other related areas)
• CiteSeerX - http://citeseerx.ist.psu.edu/ (computer science)
• HTP Prints - http://htpprints.yorku.ca/ (History and theory of psychology)
• PubMedCentral - http://www.pubmedcentral.nih.gov/ (US National Library of Medicine's digital archive of life sciences journal literature. See also BioMed Central - http://www.biomedcentral.com/info/)
• PhilSci Archive - http://philsci-archive.pitt.edu/ (philosophy of science)
• E-LIS (library and information science) - http://eprints.rclis.org/ (out of date)

1.4. What's the underlying rationale?

Most activity related to eprints assumes that researchers publish in journals and present at conferences in order to disseminate research rather than, for example, to make money. The eprints movement (if it can be called this) concentrates on 'toll-free' access to research outputs. It is accepted that the publication of monographs and text-books on the other hand often result in royalty payments to authors. Journal articles usually do not though journal publishers charge subscription fees. Institutions have to pay these fees in order for their members to access research findings (often the research outputs of their own members). In some cases, therefore, the institution is effectively paying twice for research (by paying the salary of the researcher and paying the subscription fees for the journals in which that researcher publishes). It is argued that authors (and probably their institutions) should seek to retain more rights over the articles they write. Currently, many journals demand the transfer of copyright. Increasingly, however, journal publishers are recognising the demand for open access to scholarly research and will either permit authors to self-archive a copy of the article in an institutional eprint repository, or will permit the non-exclusive licensing of rights from the author to enable the publisher to perform its function. Institutions are taking more interest in the intellectual property rights inherent in materials created during the course of employment.

At a national level funding bodies and governments are increasingly committing to, or considering a committement to, ensuring open access to publicly-funded research data and outputs. Examples include:

• A position statement by the Wellcome Trust in support of open access publishing, http://www.wellcome.ac.uk/en/1/awtvispolpub.html (out of date)
• OECD Committee for Scientific and Technological Policy at Ministerial Level Declaration on Access to Research Data from Public Funding, http://www.oecd.org/document/0,2340,en_2649_34487_25998799_1_1_1_1,00.html
• House of Commons Science and Technology Committee inquiry into Scientific Publications, evidence and forthcoming reports from http://www.publications.parliament.uk/pa/cm/cmsctech.htm
• See also the proceedings of a closed meeting, "National Policies on Open Access (OA) Provision for University Research Output", http://opcit.eprints.org/feb19prog.html

The Budapest Open Access Initiative recommends self-archiving together with support for open access journals ('...a new generation of journals committed to open access, and to help existing journals that elect to make the transition to open access').

The Public Library of Science (PLoS) is 'is a non-profit organization of scientists committed to making the world's scientific and medical literature a public resource'. The PLoS asks supporters only to publish in, or review for, journals which agree to grant free access to research outputs within six months of publication. The PLoS operates on a business model which rather than charging readers (or their institutions) to access research articles instead charge authors a fee for publishing in an 'open access' journal. On open access journals see further:

• Directory of Open Access Journals, http://www.doaj.org/ (Please note: journals included in this directory charge neither readers nor authors).
• JISC and BioMed Central agree deal that enables HE community to publish without charge in any of BioMed Central's 90+ peer-reviewed open access journals (2003-06-17).
• PLoS Biology, New Journal of Physics, the Journal of Experimental Botany, and the International Union of Crystallography are awarded funds from the JISC to allow evolution of these journals into an open access model (2004-03-05)
• OUP experiments with Nucleic Acids Research as an open access journal (authors charged 300 pounds per article in 2004)

1.5. Doesn't this risk breaching copyright?

It is true that many journals demand that authors transfer copyright to the publisher as a condition of publication. The transfer of copyright relates to the version of the article published. The copyright for any other version of the article (e.g. pre-refereed version) belongs to the author. Some initiatives recommend that authors archive a pre-refereed version of an article (preprint) together with a list of changes made subsequent to the peer-review process. Some publishers now permit authors to self-archive an eprint (post-print) within an institutional repository. Publishers usually offer an option for authors to grant a non-exclusive license to publish an article. This ensures that authors retain rights. And in a few cases publishers seek the transfer of rights but will grant back to the author certain rights including re-publication or other forms of dissemination. Authors are advised to investigate available options with the publisher and to negotiate the retention of rights where possible.

You can use the Publisher copyright policies & self-archiving developed by the ROMEO Project and hosted by the Sherpa Project to check the policies of journal publishers you use.

1.6. Is this only about research articles?

The Open Archives Initiative protocol mentioned above and the software for managing the eprint process may be used to deposit and give access to other types of digital object. Conference presentations, working papers, reports and other ephemeral literature are obvious examples. However, the OAI protocol is also in use for exposing information about archive collections, primary texts in electronic form, student publications, books, digital images, theses, works of art, museum objects, online maps, etc. Many of the large digital library projects such as the Perseus Project; University of Michigan Digital Library Production Service (DLPS); National Science Digital Library; and Library of Congress American Memory Project maintain open archives. The OAIster Project at Michigan harvests metadata from 142 archives and enables the searching of over one million aggregated records describing many different types of digital object.

1.7. What are Oxford's plans?

Oxford has been involved in two JISC-funded projects relating to eprint services. The Sherpa Project (led by Nottingham) is initiating the development of openly accessible institutional digital repositories of research output in a number of UK research universities. Through the Sherpa Project Oxford is developing an institutional eprint repository and investigating the issues relating to the eprint deposit, management and access process. Oxford's participation in the Sherpa Project is being led by Oxford University Library Services.

The Eprints UK project, now completed, was led by the Resource Discovery Network, of which the Humbul Humanities Hub within the RTS was a partner. Eprints UK investigated the development of subject-based eprint repositories from metadata harvested from institutional repositories (including from Oxford's pilot repository).

The institutional strategy with regard to the development and support of institutional repositories currently being discussed through, for example, the ICTC's Digital Archiving Group, and the response to the recommendations from the ELISO report.

1.8. References and further reading