Major proponents for OA Talk on How to Best Realize Permissions To Archive Postprints
Summary: Many Institutional Repositories (IRs) are not run by researchers but by "permissions professionals," accustomed to being mired in institutional author IP protection issues and institutional library 3rd-party usage rights rather than institutional author research give-aways. The solution is to adopt a sensible institutional (or departmental) deposit mandate and then to automatize the deposit procedure so as to take Repository Managers out of the decision loop, completely.
The optimal deposit mandate is to require Open Access deposit of the refereed final draft, immediately upon acceptance for publication, but there is a compromise for the faint-hearted, and that is the Immediate-Deposit/Optional-Access (ID/OA) Mandate:
The only thing standing between us and 100% OA today is keystrokes. It is in order to get those keystrokes done, at long last, that we need OA mandates, and ID/OA is a viable interim compromise: It gets all N keystrokes done for 62% of current research, and N-1 of the keystrokes done for the remaining 38%. For that 38%, the "Fair Use Button" will take care of all immediate researcher usage needs for the time being. The robots will have their day once 100% deposit mandates prevail and the research community tastes what it is like to have 62% OA and 38% almost-OA world, at long last. For then those Nth keys will inevitably get stroked, setting everything to Open Access, as it should (and could) have been all along....
Comments by Peter Suber:
Funding agencies can bypass permission problems because they are upstream from publishers. Researchers sign their funding contracts before they sign their copyright transfer agreements with publishers. Hence, if funders are firm (the Wellcome Trust is the model here), they can mandate OA and publishers will have to choose between accommodating their policies and refusing to publish work by funded researchers. But institutional repositories are downstream from publishers and cannot bypass permission problems in the same way. I don't care much who runs an IR. But we shouldn't assume that a university-level OA mandate will make permission problems disappear. (I'm not saying that Stevan makes this assumption.)
Most journals already permit postprint archiving, so no permission-seeking is necessary. But in the minority of cases for which it is necessary, it's a great boon to have some permission-seekers on staff to assist authors.
When the relevant permission problem is an embargo, Stevan's remedy is exactly right. Authors should deposit their work immediately upon publication, but only flip the access switch from closed to open when the embargo expires. During the period of closed access, the repository can provide OA to the metadata and the author can email copies of the text to readers who request them.
Source: P Suber Open Access News of 13 June 2007