Open Access Archivangelism

Comments invited on:

Harnad, S. (2010) Opening Research on the Web: Hastening the Inevitable. Internet Evolution. 10 June 2010.

Stuart Shieber's recommendations in his "A proposal to simplify the University of North Texas open-access policy" are very good and their spirit should be followed, because that would indeed simplify and clarify the UNT policy, currently somewhat longer and more complicated than it needs to be.

But there is an even simpler way to put it (though it does involve three components, not just two): (1) Immediate deposit required. (2) Immediate OA strongly recommended, but not required. (3) Rights retention and license strongly recommended, but not required. That's all.

Stuart's recommendations for simplifying the rights reservation and licensing details are well taken; and they do make the policy as a whole more consistent and coherent. But (1) and (3) alone, leaving out (2), would not make it clear enough to authors what the real contingencies are, even though Stuart is quite right that, by implication, (2) is in a sense implicit in (3). Waiving (3), however, does not entail waiving (2), any more than waiving (2) or (3) implies waiving (1).

There is no "Harvard-style" vs "Harnad-style" approach. That initial difference vanished completely as soon as the Harvard mandate was upgraded to add an immediate-deposit requirement, without waiver, to its original license requirement, with waiver.

The rest is just about clarifying the contingencies. (1) Yes, you have to deposit immediately, no matter what. (2) No, you don't have to make the deposit OA immediately, if you have reason not to; it is just strongly recommended. (3) No, you don't have to reserve the specified rights and grant the license if you have reason not to, it is just strongly recommended.

I think the current Harvard version still does not make the three contingencies sufficiently clear and explicit (although they are latent in the Harvard FAQ), whereas the UNT version does. Streamlined along the lines Stuart suggests, UNT will do so even better.

Ceterum censeo, I am confident that the extra rights that Harvard seeks in (3) (basically amounting to "Libre OA") will eventually come with the IDOA territory, following as a natural matter of course, with time, once the IDOA version of the OA mandate becomes widely adopted. IDOA guarantees only about 63% immediate Gratis OA plus 37% "Almost-OA" today. But once it is universally adopted, the rest of the dominoes will fall, leading first to 100% Gratis OA, and then to as much Libre OA as authors and institutions want and need. The trick is to come up with a policy model, today, that is strong enough to do the trick -- not so strong as to impede or retard universal adoption, but strong enough to ensure compliance: I think the UNT version (with Stuart's recommended tightening) will prove to be that optimal model.

See: "Which Green OA Mandate Is Optimal?"

Stevan Harnad
American Scientist Open Access Forum