Stuart Shieber, a panel participant from the Harvard forum and professor of computer science, is quoted in a recent article in the New York Times. The article discusses a proposal sponsored by Shieber to facilitate the distribution of research by Harvard arts and sciences faculty openly on the Internet:

At Harvard, a Proposal to Publish Free on Web

The title is misleading, as it seems that what is at stake is not publishing, but archiving/self-archiving. (I assume that Harvard faculty are not interested in abandoning the peer-review model of publishing in favor of self-publishing on the Internet.) As described in the Times article, the proposal is ambiguous:

“Harvard would deposit finished papers in an open-access repository run by the library that would instantly make them available on the Internet”

A couple of questions immediately occur to me: Are “finished papers” equivalent to published papers? If they are published papers, are they the formatted journal version, or the post-peer-reviewed, but unformatted version? Or are “finished papers” those that are ready for submission to a journal (e.g. pre-prints)? If the proposal addresses published papers, does “instantly” mean immediately after publication, or after an embargo period? Will the proposal establish a repository like the University of Southampton’s Eprints Soton or function as an alternative to PubMed Central ?

Journal policies regarding when and how one can distribute manuscripts vary wildly (at least for journals in the biomedical sciences). Some journals forbid the author from distributing any version of the manuscript, either pre- or post-print. Other journals allow the author to distribute both pre-and post-print versions. Yet the majority of biomedical journals, including Nature, have a policy that falls somewhere in the middle. (The SHERPA/RoMEO project is a great resource and provides an overview of and links to individual publishers’ and journals’ policies.) How will the Harvard proposal fit with journal policies?

Allan Adler, a VP at the Association of American Publishers, suggests that the Harvard proposal is not antagonistic to traditional publishers. According to Adler, publishers reject “mandates”, but not giving authors a choice regarding archiving/self-archiving. (I assume this is because the author is given the choice of violating a journal’s policies or not, which may then have negative consequences for the author.) Publishers who don’t have policies forbidding archiving or self-archiving might still object to the idea that Harvard’s faculty will retain copyright and distribution/licensing rights for articles. An op-ed in the Harvard Crimson by Robert Darnton explains:

By mandating copyright retention and by placing those rights in the hands of the institution running the repository, the motion will create the conditions for a high deposit rate… The Harvard system would have all faculty members grant a non-exclusive permission to the President and Fellows of Harvard to distribute their articles… Anyone who wanted to retain exclusive rights to her- or himself could do so by obtaining a waiver. Of course, those who cooperate with the system will also retain full rights to the publication of their work. By sharing those rights with Harvard, they sacrifice nothing; and they will have the collective weight of Harvard behind them if they resist a journal’s demand for exclusive rights. We have designed a legal memorandum called an author’s addendum to reinforce them in negotiations with commercial publishers.

Since 2006, SPARC has provided just such an author addendum for negotiating with commercial publishers, yet my informal poll of authors and publishers suggests that authors rarely negotiate copyright with journals. This may change with the Harvard proposal. Darnton suggests that the Harvard proposal will give Harvard authors a stronger negotiating position with journals. Authors from other institutions may also benefit in their negotiations. Finally, the proposal seems likely to increase awareness that one can try to negotiate legal rights like copyright when publishing.

As Darnton notes, the Harvard proposal does include an “opt-out” policy, yet is unclear if this will allow authors to opt-out from archiving individual papers or on a global basis. I suspect that the details of policy will have have a huge impact on the number of manuscripts which are self-archived.

Universities have long been interested in retaining some of the fruits of their researchers’ work, whether it has been via institutional repositories, databases, or eprint servers. In addition to making the results of research openly available, and thus benefiting the community as a whole, this proposal seems in part to be an attempt by Harvard to also retain copyright and distribution/licensing rights for their researchers’ work. I’m curious to see how Harvard will license the manuscripts in their repository and what rights they will grant to users of the repository (re-use, re-distribution, read-only?). It will also be interesting to see how publishers respond, given Harvard’s clout in the research community.

Update: More information/commentary in articles/posts from Inside Higher Ed , The Boston Globe, The Chronicle of Higher Education blog, and Slashdot.

Update: Harvard has issued a press release about the initiative. It states that authors will retain copyright and not the university. Authors will instead provide the university with “a nonexclusive, irrevocable, paid-up, worldwide license to exercise any and all rights under copyright”. (Quoted by Chris Armbruster) Based on Robert Darnton’s op-ed in The Crimson, I (mistakenly) assumed that the university would hold the copyright rather than the authors. I have modified some of the statements above to reflect this fact.

Another update: The Scientist reports that the proposal will require authors to post manuscripts as soon as they have been accepted for publication. This implies that the repository is for post-peer-reviewed, but unformatted versions of manuscripts.

Update: Harold Varmus, who was the keynote speaker at the forum, has spoken about the Harvard proposal to Bloomberg.com. Varmus notes the distinction between author self-archiving and peer-reviewed, open access publishing.