Preservation of Access to Subscription Electronic Journals in Australian University Libraries : a Discussion Paper for the CAUL Electronic Information Resources Committee
Christine Maher
David Groenewegen
Gail James
November 2002
Executive Summary
Arrangements for preserving access to electronic journals currently or formerly subscribed to by Australian university libraries are very unsatisfactory. Trends for publishers to build and maintain their own archives involve serious dangers for both the publishers and libraries; but the magnitude of the task is such that individual libraries will be unable to provide long-term solutions on their own. Long-term (defined as at least 100 years) access preservation is contingent on the development of new arrangements for electronic archiving, which will require collaboration between libraries and publishers in order to be technically and financially sustainable.
It is recommended that CEIRC and CAUL sponsor further research and a project to examine the feasibility, security effectiveness, cost effectiveness and intellectual property implications of “within Australia” third party arrangements and to compare the cost effectiveness and security of such arrangements with projected collaborative electronic archives hosted overseas.
Preservation of Access to Subscription Electronic Journals in Australian University Libraries : a Discussion Paper
This paper started life as a response to the concerns expressed by CAUL Datasets Coordinators at meetings over the last two years that the licence provision arrangements for continuing access to subscription electronic journals after cancellation of titles were in many instances manifestly inadequate for maintenance of long-term access to the subscribed materials, even though many licence agreements promise to ensure ongoing access to cancelled e-titles. Tony Millett’s paper[1] for CEIRC outlines the considerable variations from publisher to publisher in the licence arrangements for continuing access to e-journals after cancellation and substantiates the concerns of the Datasets Coordinators.
It quickly became apparent to the authors that preservation of access to e-journals whether as cancelled or ongoing subscriptions, is dependent on the successful archiving and preservation of e-journal content. Flecker points out that “The issue of long-term archiving and preservation of e-journal content has become one of increasing importance” to research libraries. He further comments that concerns about the archiving of e-journals have retarded the move from print to electronic-only subscriptions at the same time that duplication of print and electronic journals is unlikely to be sustainable over time.[2] Raym Crow states in his paper for SPARC on Institutional Repositories that “Digital preservation and long-term access are inextricably linked : each being largely meaningless without the other.” [3]
Continuity of access to e-journals looks to be increasingly problematic regardless of whether a library cancels or continues to subscribe to particular titles. Publishers of e-journals have assumed de facto archiving responsibilities for these journals and the high costs of doing so may endanger their commercial viability. The U.K. Dept. of Trade and Industry Report “Publishing in the knowledge economy; competitiveness analysis of the UK publishing media sector” sees the “cost of maintaining a digital archive” as a threat to the existence of the journal publishing industry.[4] Examples of publishers undertaking archiving of their own e-journals are American Chemical Society, Elsevier Science and Institute of Electrical and Electronic Engineers.
Australian academic libraries in the past few years have invested very large sums of money in subscriptions to electronic journals. As long as the e-journals merely duplicated existing print subscriptions issues of continuity of access and archival responsibilities seemed not to be very pressing, because the libraries still relied on the print copy to be the archival copy. Similarly, ambiguities about whether libraries are leasing (access only for the duration of the subscription) or licensing (implying an ongoing right of access to content) e-journals when they subscribe seemed not to matter as long as the print copy was still available.
In an environment where financial exigency combined with end-user demand for e-journal access anywhere at anytime is compelling many Australian academic libraries to cancel print subscriptions and rely entirely on electronic versions of journals, ensuring that these libraries have long-term (defined as at least for 100 years) access to the material that they have paid for has now become of utmost concern and is the key issue to be examined in this paper.
Archiving
Archiving can involve three different but related issues. This paper does not attempt to address all the problems and issues related to these areas, so some definitions are in order to make the focus of the paper clear:
1. Digital preservation is concerned with ensuring that material created or stored in a digital format is able to be accessed in the future. The rapid advances in computing technology over the past twenty years have shown that:
· Storage formats become outmoded, and therefore inaccessible. A commonly used example are the 5 ¼ inch floppy discs, which were in common use during the 80s and early 90s, but which are as good as useless because modern PCs no longer come with the correct drive installed.
· Operating systems become obsolete, making programs difficult to run, and thereby making the information inaccessible. Many DOS based programs suffer from this problem in modern operating environments.
· Hardware becomes outmoded, and thereby prevents access to information. A recent example of this has been NASA’s reliance on early 80s computer parts to run the Space Shuttle program (http://www.iht.com/articles/57527.html).
· Software changes, which can affect the way that information is viewed or accessed. This can be a common problem with web pages, which were designed to run on a particular browser version. Information may be lost or rendered unreadable if viewed in a different browser, or even a different version of the same browser.
Strategies for addressing these problems are referenced in the bibliography at the end of this paper; however, preserving material is not the primary focus of this paper.
2. Access efficiencies are concerned with making it easier and cheaper to access electronic material. This may involve creating a mirror site (i.e. a duplicate site, which contains all the same material) within Australia, so that Internet traffic costs are reduced, and access is (theoretically) quicker. This mirror site would have the potential to act as an archive (as all relevant electronic material would be copied to it), but this would not be its primary purpose. A mirror site would create its own maintenance and licensing issues.
3. Electronic archiving is defined, for the purposes of this paper, as ensuring ongoing access to electronic material that has been paid for by Australian University libraries.
The issue of electronic archiving is complicated by the question of what level of archiving will satisfy the needs of our clients and who assumes the responsibility and the costs of archiving.
1. Ensuring that all scholarly information is preserved by someone, somewhere, is an electronic equivalent of the mandatory deposit schemes of the past. The feasibility of this is greatly affected by the digital preservation issues discussed earlier, and also by the continuing existence of the publishers of the material, many of whom refuse to allow archiving outside of their own organizations.
2. Ensuring that Australian libraries have long-term access to material that they have paid for. Where a subscription is cancelled, continuing access to the paid-for issues of the e-journal is reliant on:
· The continued existence of the publisher, and its willingness to offer this service.
· The publisher actually being able to offer this service – when Monash tried to enforce this at Ovid in early 2000 it became clear that Ovid weren’t quite sure how this would actually work, and the solution was less than ideal, although it has since improved.
· Equality of treatment by the publisher of current and cancelled subscriptions – Elsevier have made clear that cancelled titles might not be eligible for the same range of services as ongoing ones.
· Archival formats which may quickly become technologically obsolescent, e.g. cd-roms as provided by Project Muse and EMERALD.
· Ability of libraries to pay additional archival access fees where these are required by the publisher – this trend is increasingly evident e.g. American Chemical Society. If this trend were to be generalised to most publishers of e-journals then it is unlikely that many university libraries could sustain the costs in the longer term, or at least while they are simultaneously continuing to act as long-term repositories for books and print journals.
Ways Forward
A number of operational and research projects have identified a range of potential solutions to the problem of who assumes responsibility for e-journal archiving and how to support the likely fairly heavy costs involved.
Open Archiving
Open Archiving (sometimes known as E-Print Archives or repositories) has been mooted by some as the long term solution to the problem. The essential theory of this movement is that higher education and research bodies are paying for information twice – once to write it, and again to obtain a journal subscription to enable other members of the institution to read it and similar materials. In return they get a peer review process and distribution of their ideas, but the feeling is that the costs are now outweighing the benefits of this system.
The current proposal is that these institutions should store electronic copies of all the research publications of its staff on its own servers, or on general subject servers, and that these should be made freely available to all. The assumption is that once enough institutions take this on, all scholarly publications will be available to all scholars, and journal subscriptions will be able to be cancelled, thus freeing up funds for the cost of the local archive. This movement is gaining a good deal of momentum, but may not necessarily fulfil the library role as archivist (as Michael Day pointed out[5]), and they rely on full compliance by all authors to make sure every article is available –which may not be the case.
Relevant sites include:
· http://www.eprints.org/
· http://www.openarchives.org/
· http://www.arxiv.org/
Publisher Archives
One of the major dangers of relying on publisher e-journal archives has already been briefly described above, i.e. the high cost of maintaining such an archive for a large suite of journals and the resulting pressures on the commercial viability of the publishers undertaking such archiving responsibility. The recent move by the American Chemical Society to require separate payment of an annual subscription fee to its archive of e-journals on top of the annual subscription fee to a rolling few years of the current issues of its journals is a potent indicator of the sorts of costs which need to be covered in order to sustain the archive. Dual subscriptions to maintain both the current access and archival access have budgetary implications for libraries which call for further analysis and comment.
Another aspect to this process of publisher self-archiving is that some publishers are digitising back files of their print titles and incorporating the newly digitised retrospective material in the current subscriptions, whose prices then rise to reflect the cost to the publisher of the retrospective digitisation. The danger arises where libraries have no choice in the matter of whether they wish to subscribe to the retrospective electronic content or not. ACM Digital Library appears to have taken this route and IEEE intends to follow suit.
Donald Waters, Co-Chair of the Task Force on Archiving of Digital Information created by the Commission on Preservation and Access and the Research Libraries Group, has also identified another equally important concern with publisher archives and that is “whether the material is in a preservable format and can endure outside the cocoon of the publisher’s proprietary system”. Waters goes on to state that “One necessary ingredient in a proof of archivability is the transfer of data out of their native home into an external archive, and as long as publishers refuse to make such transfers, this proof cannot be made”.[6]
The lack of trust in publisher archives is also cited by Sarah Thomas as an outcome of an informal Project Harvest( one of the Mellon Foundation funded projects) survey in which “90% of respondents preferred multiple custodians rather than a single-party preserver” and “many publishers were insufficiently aware that others did not trust them to archive materials responsibly or to be the sole custodian of their output”. [7]
Third Party Solutions
Waters comments elsewhere in the paper cited above that an unexpected outcome from the Mellon Foundation projects on e-journal archiving is that “new organizations are likely going to be necessary to act in the broad interest of the scholarly community and to mediate the interests of libraries and publishers”. (p.87) In other words it is unlikely that either libraries acting on their own or publishers acting on their own will be able to provide a sustainable solution to the access and archiving issues.
The Andrew W. Mellon Foundation projects involve collaborations between a number of stakeholder organisations who share the objective of finding cost-effective long term archiving solutions for subscription electronic journals. These organisations include in the US the Digital Library Federation, Council on Library and Information Resources and Coalition for Networked Information. The Mellon Foundation has financed projects undertaken by Cornell, Harvard, and Yale University libraries, the University of Pennsylvania Library, New York Public Library and Stanford University Library. It is notable that several of these projects involve collaboration with the publishers of the journals. Further details about the projects can be found in the Appendix to this paper.
In the UK projects are under way under the auspices of the British Library Research and Innovation Centre, JISC Committee for Electronic Information and JISC Preservation Focus and the U.K. National Preservation Office. These projects and the Web addresses where corresponding documentation may be found are also listed in the Appendix to this paper.
Costs of Long-Term Archiving of E-journals
A recently released report from one of the Mellon-sponsored projects, that undertaken by Yale University Library in collaboration with Elsevier, entitled “YEA: the Yale Electronic Archive, one year of progress” establishes that a collaborative e-journals archive is now technically feasible, although significantly, issues of financial feasibility and sustainability will become the object of investigation in the next phase of the project.[8] The discussion in the Chapter titled ”Some Economic Considerations” makes it clear that electronic archiving costs are likely to be high to very high and that a range of payment models for users of archival e-journals will be required to support different levels and types of access. (p25-30)