Research should be the priority

In the article 'Archive programmes gain momentum' in your October/November 2005 issue, Sally Morris of the Association of Learned and Professional Society Publishers (ALPSP) is quoted as saying that if 'the majority of papers from a particular journal become available for any researcher to find [in an institutional repository] it could lead cash-strapped libraries to stop buying that journal, which would make it no longer viable'.

When I think of cash-strapped librarians I think of all the journals that they cannot afford for their institution's users. Not being able to buy a large proportion of the journals available means that their researchers cannot access and use research results. This has an impact on their research and progress. Cash-strapped librarians must cancel journals. They have nowhere near enough cash to buy access to all journals or even enough to meet their users' potential needs.

Researchers, who are both the producers and the users of the research, can be forgiven for thinking about research access and impact rather than the cash flow to publishers. Is the public funding research, and are researchers conducting it, in order to ensure that cash flows from cash-strapped librarians to publishers' bottom lines?

Morris also suggests that if readers access journals through repositories this will not show up in the usage data so a librarian might decide to cancel a subscription even though the same numbers of people still access the journal.

Librarians must always cancel one journal in order to be able to afford another one. Their budgets do not enable them to subscribe to them all. But a self-archiving mandate does not affect this situation because it applies to the funded articles in all journals. All it does is provide a safety net so that when a librarian is forced to cancel a subscription the users can at least access the author's self-archived draft.

Unless Morris believes that journals add no value that is worth paying for, this scenario does not spell the end of journals; just the end of needless access-denial and impact-loss for researchers, and a lessening of the stress for librarians.

Elsewhere in the article, Leo Waaijers, manager of the SURF-DARE programme in the Netherlands, described the biggest challenge in creating institutional repositories as 'convincing academics that they have something worth preserving and that an institutional repository is the place to do it'.

I believe that it might be more useful to try to convince academics that they have something worth maximising access to today rather than focusing on preserving it for the future. Preservation is not the pressing problem facing research today. Instead, the problems are needless access-denial and the resultant impact-loss for the current generation of researchers.

In any case, worries about preservation should first be addressed with the official journal versions that the libraries pay for, not the home-brew version that they are trying to persuade their own authors to provide.

Stevan Harnad
Moderator for the American Scientist Open Access Forum; research chair in the neuroscience and cognition centre at the University of Québec, Montréal, Canada; and professor of cognitive science at the University of Southampton, UK

Publisher concerns are valid

In reply to Stevan Harnad's letter, I would like to make clear that the concern of publishers about the possible impact on their journal subscriptions/licences of self-archiving in repositories, should the practice ever become widespread, is both rational and well-founded.

As we made clear in our letter to Research Councils UK: www.alpsp.org/2005pdfs/rcuk050805.pdf, many publishers are already noticing indications that when readers can find the same or even 'good enough' versions of content freely accessible on the web, they are happy to use it rather than the publisher's 'added-value' version. In particular, the Institute of Physics Publishing has noted that downloads are much lower for those journals whose full content is available in the arXiv physics repository.

We believe that librarians are rational people, trying to cope with budgets which are inadequate to purchase all the resources their users require. It would, surely, be irrational not to cancel first those journals which your users can do without because they have access to - and are using - freely available versions of the same content. We are currently seeking more information about the role that free availability plays in librarians' cancellation decisions, but in the meantime simply denying that is has any role does not seem very sensible.

We do not argue that publishers have a right to exist and to make money; of course publishing exists to serve scholarship, and not the other way round. Indeed, dissemination is at the heart of the mission of every learned society. We simply ask that those who are advocating policies which risk damaging, or even destroying, journals think about the consequences to scholarship: how would the management of peer review, and the 'branding' of articles as being of particular relevance and importance to a specific community of interest, be funded? And what would be the impact on learned societies' other activities (such as conferences, travel bursaries and research funding)?

Simply to say that publishers could switch to an 'open-access publishing' model is not good enough - the evidence available to date (see, for example, our recent study: www.alpsp.org/publications/pub11.htm and Mary Waltham's analysis for the UK's Joint Information Systems Committee: www.marywaltham.com/JISCReport.pdf) suggests that this is not always a viable alternative.

Sally Morris
Chief executive, Association of Learned and Professional Society Publishers