OA is not just a technical question about how to finance journals or launch repositories

20 avril 2005

Peter Suber is the Open Access Project Director at Public Knowledge, a public-interest advocacy group in Washington D.C. focusing on information policy. He’s also a Research Professor of Philosophy at Earlham College and Senior Researcher at the Scholarly Publishing and Academic Resources Coalition (SPARC). He has a Ph.D. in philosophy and a J.D. from Northwestern University. He is the author of the SPARC Open Access Newsletter and editor of the Open Access News weblog. He was the principal drafter of the Budapest Open Access Initiative, and sits on the Steering Committee of the Scientific Information Working Group of the U.N. World Summit on the Information Society, the Publishing Working Group of Science Commons, the Advisory Board of American Library Association Information Commons, and the Board of Governors of the International Consortium for the Advancement of Academic Publishing. Lingua Franca magazine named him one of « Academia’s 20 Most Wired Faculty » in 1999. He has been active in promoting open access for many years through his research, speaking, and writing. (For more details, see his home page).

Question : As an American, what do you think of the way things are shaping up in Europe regarding Open Access, especially since the Berlin Declaration ?

Peter Suber : The cutting edge of the OA movement is now in Europe. In mid-2004, both the US and UK looked like they might mandate OA to publicly-funded research, a rational and long-overdue step. But by late 2004, the US National Institutes of Health (NIH) had retreated from the strong policy recommended by the House of Representatives and the UK government had rejected the strong policies recommended by the House of Commons. In February 2005, however, the Berlin3 conference in Southampton issued exactly the policy we needed to move forward —namely, to require deposit in OA archives and to encourage publication in OA journals. The framers of this policy were largely European and most of the universities, laboratories, and funding agencies now signing on to the policy are European.

The Wellcome Trust in the UK announced last November that it would require OA to Wellcome-funded research. Now CERN will do the same for CERN researchers. While the UK government rejected the OA recommendations of the Gibson committee last November, the RCUK may adopt the key parts of those recommendations any day now. Some of the institutions signing the Scottish Declaration on Open Access will require OA and the rest will encourage it. A February conference in Ukraine recommended that Ukraine mandate OA to publicly-funded research. Finland has officially decided to encourage OA without requiring it, as have major research institutions in France (INRA, CNRS, INRIA, and INSERM), major research institutions in Germany (Max Planck Society, German Research Foundation, Deutsches Forschungsnetz, HRK), the DARE network of Dutch universities, 32 research institutions in Italy, and important institutions in Austria, Belgium, Estonia, Greece, Hungary, Norway, and Sweden.

OA is making progress on every continent, but it’s moving fastest in Europe. In one sense this is an unimportant fact, especially since science is international. Progress has to be fastest somewhere and at the moment that place is Europe. What’s important is that Europe leads in the soundness and coherence of its OA policies, not just in the speed of adoption. For example, institutions looking for direction should adopt the Berlin3 policy over the NIH policy.

I don’t think there’s any national or regional competition for the lead on OA. But if there is, it’s a strange kind —competition to serve everyone everywhere, not to win glory or distinction. It’s a competition in which the leaders want to reduce rather than enlarge the distance between themselves and everyone else.

Q : What do you think of the final NIH Public Access Policy and the effect it could have on the future of Open Access in the US ?

P.S. : I’m disappointed with the final version of the NIH policy. The original version was good and publishers succeeded in watering it down. The original version would have required OA to a very large body of publicly-funded research. (The NIH is the world’s largest funder of medical research.) But the NIH reduced the requirement to a request. The original version would have required OA through PubMed Central within six months of publication in a peer-reviewed journal. The six-month embargo   was a compromise with the public interest designed to win the support of publishers. But when publishers were not satisfied and demanded more, NIH allowed delays up to 12 months. Yet even the 12-month figure is deceptive, since there is no deadline whatsoever. Now that compliance is optional, NIH grantees might never deposit their work in PubMed Central.

The public comments focused on the stronger, earlier version of the policy, and even Elias Zerhouni, the Director of the NIH, admits that the public comments were overwhelmingly supportive. Dr. Zerhouni explains that he made additional concessions to publishers in order provide « maximum flexibility for maximum participation ». But the way to assure maximum participation is to require it, not to make it discretionary.

While the NIH does not require grantees to deposit their work in PubMed Central, it « strongly encourages » them to do so « as soon as possible » after publication. Dr. Zerhouni clearly hoped that flexibility plus strong encouragement would lead most grantees to authorize release their work very soon after publication, and therefore that the new policy would achieve better results than the original version of the policy with a six-month embargo. One by one, however, publishers are announcing that they will not allow their NIH-funded authors to deposit their work with NIH earlier than six or 12 months after publication

The NIH has a statutory license to provide OA to NIH-funded research. But it has decided to put it to one side and rely instead on publisher consent. When publishers do not consent on NIH’s terms, it lets them go their own way in the name of flexibility when it should turn back to its statutory license.

I’ve written more extensively about the weakening of the policy, for example, in both the February and March issues of my newsletter. But in short, the last-minute concessions NIH made to to publishers will delay public access to publicly-funded medical research and deprive the public of access to some of it altogether. This harms the public purpose served by the NIH and violates the NIH’s own criteria for the policy.

Q : Do you feel that the current co-existence of two approaches, one concerning publishing and the other concerning institutional archives, could hinder Open Access or could their complementarity be a positive factor for Open Acces ?

P.S. : The two approaches are complementary. OA journals provide peer review, while OA archives do not. OA archives provide instant dissemination of new findings (and a time-stamp for those concerned about priority), while OA journals do not. OA archives also tend to provide long-term preservation alongside their access-improving function.

As we move from low-volume self-archiving to high-volume self-archiving, and approach 100% self-archiving for new work, then it’s likely that many subscription-based journals will be threatened. We will need OA journals to provide peer review. Likewise, as we move from scattered OA journals to comprehensive coverage of the research landscape, with high-quality OA journals in every research niche, then we will need OA repositories for preprints and to mirror the OA journals and assure authors and readers that their contents will always be OA. Above all, in the long transition period in which we do « not » have OA journals in every research niche, we will need OA repositories for authors who publish in non-OA journals.

For these reasons, I believe that a thriving system of OA archives will stimulate OA journals, and a thriving system of OA journals will stimulate OA archiving. It would be a mistake to cultivate just one when we could cultivate both. Some friends of OA disagree on which should be cultivated first, or which need is more urgent. But I don’t think we have to choose. It’s healthy for the wider movement that some active and committed people see a more urgent need to cultivate archives and others see a more urgent need to cultivate journals. This is simply division of labor that improves our efficiency and helps us reach our goal faster.

Q : It would seem, that except for arXiv, the repositories created by universities and research organizations these past few years are encountering problems in recruiting contents. For you, what are the reasons for such a lack of interest on the part of authors ? And what can be done about it ?

A study by CIBER (March 2004) showed that most senior researchers knew very little about OA. A nearly contemporaneous study by JISC and OSI (February 2004)showed that when researchers do know about OA, they support it in large majorities. So the problem is not researcher opposition, but researcher ignorance and inertia.
The good news is that the spread of knowledge helps OA, just as OA helps the spread of knowledge.

Too many scholars don’t know about OA archiving and its benefits. For example, they may know about OA journals but not OA archives. They may know about OA archives but not realize that it only takes a few minutes to deposit their work. They may not know that OA significantly increases citation impact. They may not know that OA archiving is compatible with publishing in a non-OA journal. They may not know that 80% of surveyed journals allow authors to deposit their postprints in an OA institutional repository. They may have a groundless fear that archiving their preprints will make them ineligible for later publication, when in fact the number of journals with such policies is small and declining.

Too many scholars overlook their self-interest and see OA primarily as a political gesture or an act of charity. They need to appreciate that OA will make their work more visible than any kind of priced publication (in print or online), and will make it easier for readers to find, apply, build upon, and cite. Scholars are very busy, but they’re not too busy to do research that they love. They’re not too busy to take unloved follow-up steps, like submitting their manuscripts to journals, responding to referee comments, sending offprints to colleagues, or sending updated bibliographies to deans or department chairs. They find time for these unloved steps because they understand the connection between them and career-building. What they have to understand is that OA is career-building, whether it is through OA archives or OA journals. Enlarging one’s audience and impact is career-building.

We have to develop an academic culture in which failure to deposit finished work in an OA archive is as unthinkable as failure to type it legibly and submit it for publication.

In an article for Nature last June, I tried to be more specific on how to educate authors about OA, assist them in providing OA to their own work, and create incentives and remove disincentives for providing OA to their own work.

Q : If the Life Sciences communities and of course the Physics community, seem to be quite attuned to the Open Access issue, what do you think is happening in other scholarly communities ?

P.S. : There’s no doubt that OA is moving more slowly in the humanities than it is in the STM fields (science, technology, medicine). There’s no doubt that OA in some scientific fields, like field biology, is moving more slowly than in other scientific fields, like biomedicine. In fact, there’s no doubt that there are many differences among the disciplines relevant to the funding of OA. But at the same time, there’s no doubt that OA will improve research productivity and benefit authors and readers in every field, including the humanities. Scientists seized these opportunities sooner than humanists, but humanists are definitely starting to see the possibilities. For example, just this month Alun Salt called for OA in archaeology and Roy Rosenzweig wrote an exemplary argument for OA in history. Last year I analyzed some of the reasons why OA is moving slowly in the humanities and some of the steps that we can take to promote OA in the humanities. There are impediments, but I’m not worried. OA is desirable and attainable in every field, and it’s natural that it would move more quickly in some fields than in others.

Q : In recent years, we have seen that the Open Access debate was no longer confined to the scholarly world, but was spreading to the « civil society » as seen during the last World Summit on the Information Society. What do you feel could come out of this ?

P.S. : One important aspect of recent OA progress is that many kinds of institutions —universities, libraries, foundations, non-profit organizations, and governments— are realizing that OA is in « their » interest, not just in the interest of individual scholars. So while scholars still control how much OA there will be, many institutions are now looking for ways to influence the decisions of scholars and to create incentives for them to decide in favor of OA. This is progress in two ways. First, it means that these institutions have new clarity about their own interests and, second, it means that more scholars will provide OA to more scholarship.

During the evolution of the NIH public-access policy, a turning point came when the academic and library groups were joined by non-profit advocacy organizations representing medical patients. Policy-makers may respect academics but they don’t fear their opposition they way they fear patient-advocacy groups like the Genetic Alliance, the Arthritis Foundation, or the Spina Bifida Association of America. It was critical when a large number of these groups came together under the auspices of the Alliance for Taxpayer Access to demand OA to publicly-funded medical research.

Beyond this, the World Summit itself is very helpful in going over the heads of officials focused narrowly on different issues and constituency groups. I’ve often pointed out to US policy-makers that our country is already committed to the principle of OA through the WSIS Declaration of Principles and Plan of Action, as well as the OECD Declaration on Access to Research Data From Public Funding. The same appeal could be made to policy-makers in Europe and many other parts of the world. These agreements are meaningless unless we make use of them.

What’s key is to realize that OA is not just a technical question about how to finance journals or launch repositories. It’s about sharing knowledge and accelerating research. These are or ought to be critical interests outside the academy, since even non-researchers depend on research advances. Some organizations outside the academy already realize this, such as the patient-advocacy groups. But we bear the burden of educating the rest of society that our current system for distributing scientific knowledge hinders research and all the benefits of research, from medicines and technologies to environmental health, economic prosperity, and public safety.