Harvard adopts opt-out open-access policy
From Nature 451, 879 (2008):
Harvard University has adopted guidelines under which the 'final drafts' of academic papers written by researchers at its Faculty of Arts and Sciences will automatically be published on the university's website, unless the authors request a waiver. Immediate open access to papers could conflict with the copyright policies of many journals including Cell, Nature and Science.
Many institutions keep open-access repositories of papers but the decision makes Harvard the first US university to sign up to default open-access publishing for its research staff. Although the University of California has toyed with the idea for years, it has yet to agree on a policy.
Stuart Shieber, the computer scientist at Harvard who proposed the scheme, says that any request for an exemption will be granted. The university has not yet worked out how to define what constitutes a 'final' draft of a scholarly paper, nor come up with a time limit for submission.
Critics of open-access policies worry that highly selective journals with large readerships will suffer, and that non-peer-reviewed research will become more prominent.
A longer version of this article is available at Nature News.
There is also an online debate on the Harvard announcement in the Publishing in the New Millennium forum at Nature Network, with several updates and links to further information.
Comments
The Harvard Open Access Mandate is a great step forward for Open Access (OA) and could serve as a model mandate for adoption by all universities worldwide -- if only its one important (but correctable) weakness were corrected:
A mandate is not really a mandate if one can opt out. It is merely a request. And so far all OA policies that have been mere requests rather than mandates have failed because of high opt-out rates (most notably, the original NIH Open Access policy, which was upgraded to a mandate only 2 months ago, three years after it was adopted, and failed, as a request).
The remedy is simple. It leaves the current Harvard policy completely intact: It continues to require (allowing opt-out) that all Harvard authors renegotiate copyright with their journal publishers so as to retain copyright and assign only non-exclusive access.
But to that copyright retention requirement with opt-out is added a deposit requirement without opt-out: the author's peer-reviewed final draft of all (future) journal articles must be deposited in Harvard's Institutional Repository, with no opt-out, but with the option of making the deposit Closed Access instead of Open Access if the author opts out of copyright retention and the journal does not endorse immediate Open Access Self-Archiving.
A no-opt-out deposit mandate has already been adopted by a number of universities and research funders and it has been found to lead to an 80-100% compliance rate within two years. In contrast, mere requests result in compliance rates of 5-15% (and NIH's compliance rate after 3 years was even lower: 4%).
Optimizing Harvard's Open Access Mandate
Some small but crucial changes to the wording of Harvard's OA policy that will immunize its deposit requirement against any opt-outs from the copyright-retention requirement. These changes will greatly increase the probability that the mandate will succeed at Harvard as well as making it suitable for adoption as a model by other universities worldwide.
Interested readers can contact me for further details of these proposals.
(Comment edited for length reasons.)
Posted by: Stevan Harnad | March 2, 2008 12:54 AM