Harvard mandating open access
Writing about web page http://chronicle.com/news/article/3943/harvard-faculty-adopts-open-access-requirement
The world of repositories is very much full of the news that Harvard University’s Faculty of Arts and Sciences have mandated open access deposit for its academics. There is an "opt out" that they can use, though, if they can't get the publisher's permission.
So it remains to be seen whether this will actually cause any changes in the scholarly communication model, in favour of open access and IR deposit as a standard procedure. Authors might simply call upon the option to opt out, rather than negotiate with their publishers. After all, no academic would want to miss out on getting an article published in a prestigious journal, just because their institution wants a copy of the text for their open access repository!
But it does seem that so much of what can go into a repository is actually dependent on the authors themselves, so pressure from funder mandates and institutional mandates is what is needed to populate open access repositories. It is my impression that authors don't feel comfortable with sharing early versions of their work, and they would prefer that someone from the library just harvested everything they ever wrote for them, from the final versions. But in order for the library to do that, we would have to have permission from the copyright holders. Since the authors themselves usually sign copyright in their work over to publishers under the current publishing model, we can't do what the authors would prefer.
So, Harvard academics will presumably be putting pressure on publishers to use a licence to publish rather than signing copyright agreements, in order that they can deposit the final version into their IR. This might make it easier for academics from other institutions to carry out similar negotiations. The news of the mandate itself might highlight to the academic authors that they can ask for a licence to publish in preference to a copyright agreement. Taylor and Francis' website mentions that they have such a licence already, but I doubt that many authors know about it or request to use it.
It seems to me that we could do with a list of which publishers and/or journal titles will use such a licence to publish, enabling authors to share the text of their final version in an open access repository. Such a list would help authors to have the confidence to ask for a licence, after the (sometimes) months of negotiations that take place prior to their article being accepted. I can appreciate that the academic author will perhaps be willing to sign anything at the end of such a protracted process, just to get their work into publication, with all the pressure that is on them to get published, and to get published in particular prestigious titles.