Information access: what is to be done?
Blackwell Science Ltd
Blackwell Publishing, the company that will be created through the merger of Blackwell Science and Blackwell Publishers, will publish journals for some 500 scientific societies. Last year Blackwell Science paid £12 million (US$17.3 million) in royalties, profits and expenses to its societies. Of the 30 journals new to its list in 2000, 10 were previously self-published and 10 came from other publishers. These societies chose us as publisher largely for our ability to maintain and develop revenues for them and increase the readership of their journals, despite difficult market conditions. This is the interest I am declaring.
Naturally the Public Library of Science petition has been read by many of the editors and officers of societies for whom we work. Several have been intrigued by the idea that early free access could widen the reach of their journal and one editor would like to try it. But not one of our societies has been prepared to take the commercial risk, although they will be watching with great interest those that have. As is clear from some of the other contributors to this debate there are not yet enough data available to enable us to advise our societies; effective online access has only been available for three or four years, and free access is a more recent experiment.
In most cases a journal is an important part of a society's activities and is its major asset, built up over several years by the efforts of its members, and now generating revenue to fund other activities such as meetings and scholarships. There is a close synergy between a society and its journal(s). The suggestion from scientists largely based in North America and working in one subject area that their colleagues from other countries and other disciplines should follow their wishes and support what is largely a North American system - i.e. PubMed Central - has been seen by some as arrogant. Society officers work hard on their journals to ensure a high level of peer review and, as Frank Gannon has mentioned in this debate, outside the USA they do not have large memberships to fund publication. They do not see why their efforts should be put at risk. Not all journals behave in the same way.
It has been suggested that providing free access six months after publication will not reduce demand for primary research journals as usage has fallen off sharply by then. Our experience in molecular biology shows that downloading of articles diminishes sharply within three months of publication and by six months the rate might be as low as 15% of that at the time of publication (and 7% after 12 months). In other subjects the pattern is entirely different. For example, in whole-organism biology there may be some fall off in the early months but after 9-12 months downloading can be back up to 75% of the initial rate, probably as a response to the first citations. Even after three years, downloading can be at 50% of the initial rate. In surgery, there can be an initial delay with the maximum rate being recorded about three months after publication, followed by a gradual decline, although there is sometimes an uplift after around 24 months.
In these cases, where much of the usage is well after publication, libraries could be expected to hold off until access is free because they would not be giving their users much advantage by paying for early access. Only the richer institutions, perhaps in North America, Western Europe and Japan, might be prepared to pay and thus carry the costs of publication. Ann Okerson has already commented on this and that the developing world would be at a disadvantage.
Anyone following this debate should read the excellent book by Tenopir and King: Towards Electronic Journals: Realities for Scientists, Librarians, and Publishers. As they say, journal publishing has always been about readers wanting to reach authors and authors wanting to reach readers. In the same spirit, the Blackwell Publishing vision is 'publishing will flourish', meaning that our authors will reach more readers and vice versa. A combination of new technology and pricing models will enable publishers to achieve what they did not even dream about five years ago. And librarians are supporting new models. For example, at a UNESCO conference in Paris on electronic publishing in 1996 they were hostile towards the idea of basing prices on the size of the institution and likely usage. At a similar conference earlier this year there was no such hostility and indeed they seemed to accept that those that cannot afford current prices, e.g. libraries in the developing world, should be charged very much less.
As always, many of the best publishing ideas will come from the academic community, and any suggestions from a group of eminent researchers should be taken seriously. Roberts et al., however, seem to be using old models. They start with reference to the Alexandrian Library, a flawed and out-dated model as no library can ever hold all information and yet for the first time with the World Wide Web the effect can be realized, along the lines, for example, of E-Biosci and through collaboration between publishers as with CrossRef. Roberts et al. rightly focus on access but this was a problem with hard-copy journals sold only at one subscription price. Their views may well help to create new, stronger models to the benefit of researchers, teachers and students (an often undervalued part of the system) but their threatening stance and lack of understanding of the close relationship between societies and journals and the usage of journals could cause others to shun them.
I have used the same title for this contribution that Lenin used for his 1902 revolutionary pamphlet. I am not suggesting that Roberts et al. have written with the same manic violence but perhaps there is a hint in their approach of the same desire for centralized control by an elite (PubMed Central) and the one-track aggression so often linked to revolutionary zeal.
As an editor has commented to us, if people were asked to sign a petition to do away with paying tax, many would sign. But they then might be the first to complain about deteriorating infrastructure and lower standards of public service. The publication of research is a serious business yet only costs a fraction of the funding of that research. The idea of boycotting journals is interesting and should be (and probably is already) applied by any researcher who disagrees with the policy of a journal. But I hope I have made a sufficient case against a blanket approach based on free access after six months. In many cases free access could undermine the finances of a journal, leading perhaps to lower standards and even its demise.
I suggest that already a great deal is being done to use the potential of new technology to transform the journal and particularly access to its content. Cutting off the revenues necessary to fund this transformation seems short-sighted.