The solution is simple. All libraries should announce that in a year's time they will stop buying any Elsevier publication. The world would be a better place without them. They are as outdated as hand-loom weavers.

I've been following the debate on the web and in the mailing lists and I have a little bit of a different perspective for a number of reasons. At Mendeley, I was involved in the development of the policy and helped to make it more friendly to sharing platforms. Also, having immersed myself in the technical details, it seems like the two titans quoted above are confused about what the new policy actually does. I have no problem putting some blame on Elsevier for not communicating it as clearly as it could have, but I also think the heated arguments above may keep those who stand to benefit from the new policy - institutional repository managers, researchers, and sharing platforms such as Mendeley - from being able to take advantage. So the debate above was mostly a resurrection of a long-standing disagreement between two large personalities, but the Association of Research Libraries has concluded there's little chance that a library would cancel subscriptions as a result of increasing green OA. They make a very good argument here: http://policynotes.arl.org/?p=80 , essentially taking apart the idea that green OA is a threat to publishers. I would argue that the idea that green OA is somehow a threat to publishers does more to slow adoption than the actual reality. Furthermore, all the heated exchanges and emotion obscures the fact that sharing of research is much more permissive under the new policy. Whatever your feelings about Elsevier (and I've been sharply critical for years), if you take an objective and informed look at the new policy, it's better in a number of ways: Private sharing on social platforms is now explicitly OK. Everyone was doing it before, but now they're officially in support, clearing the way for institutional adoption of new tools and services to support researchers. Repositories can now systematically ingest Elsevier content without Elsevier wanting to have an agreement in place. This dismantles the very odd bit of the prior policy which only allowed this if your institution didn't have a policy encouraging it. There's no retroactive removal being requested. The embargo applies uniformly across everyone, but everything you could post publicly and immediately before, you can do so now. There's a NEW category of author manuscript, one which now comes with Elsevier-supplied metadata specifying the license and the embargo expiration date, that is subject to the embargo. The version the author sent to the journal, even post peer-review, can be posted publicly and immediately, which wasn't always the case before. At Mendeley, we pushed hard for this, precisely because there's no way to tell, at scale, the difference between a preprint and what they were previously calling an author manuscript, so we worked to help develop the new category of tagged manuscript. Having these tags will now vastly ease the job of a repository manager, because making the Elsevier-branded accepted version public after the embargo is now something that can be handled by the software itself. There's no need for anyone to manually keep track of dates. So all that is why, even for an open access advocate like me (who personally prefers gold CC-BY for science), I think the new policy is a step forward for open access and I think when the rams stop locking horns and the dust clears, the whole herd will realize the pasture is indeed greener.

CROSS PURPOSES Paul, no, the purpose of “repository-focused green open access” is not -- and never has been -- "to prompt libraries to cancel journal subscriptions, forcing publishers to restrict themselves to coordinating peer review at a vastly reduced price from that currently charged for journal-focused “gold” open access" (though that is indeed what I think will be the eventual outcome). The purpose of green open access, mandated by universities and funders, is (and always has been) open access. Open access means immediate (un-embargoed), free online access to the final peer-reviewed drafts of peer-reviewed journal articles. http://legacy.earlham.edu/~peters/fos/2008/08/greengold-oa-and-gratislibre-oa.html Berners-Lee, T., De Roure, D., Harnad, S. and Shadbolt, N. (2005) Journal publishing and author self-archiving: Peaceful Co-Existence and Fruitful Collaboration. http://eprints.ecs.soton.ac.uk/11160/ Harnad, S. (1995) A Subversive Proposal. In: Ann Okerson & James O'Donnell (Eds.) Scholarly Journals at the Crossroads; A Subversive Proposal for Electronic Publishing. Washington, DC., Association of Research Libraries, June 1995. http://www.arl.org/scomm/subversive/toc.html Houghton, J. & Swan, A. (2013) Planting the Green Seeds for a Golden Harvest: Comments and Clarifications on "Going for Gold". D-Lib Magazine 19 (1/2). http://www.dlib.org/dlib/january13/houghton/01houghton.html Sale, A., Couture, M., Rodrigues, E., Carr, L. and Harnad, S. (2014) Open Access Mandates and the "Fair Dealing" Button. In: Dynamic Fair Dealing: Creating Canadian Culture Online (Rosemary J. Coombe & Darren Wershler, Eds.) http://eprints.ecs.soton.ac.uk/18511/ Vincent-Lamarre, P, Boivin, J, Gargouri, Y, Larivière, V and Harnad, S (2015) Estimating Open Access Mandate Effectiveness: I. The MELIBEA Score. JASIST (in press) http://eprints.soton.ac.uk/370203/

OPENNESS William Gunn (Mendeley) wrote: “[E]verything you could post publicly and immediately before, you can do so now. There's a NEW category of author manuscript, one which now comes with Elsevier-supplied metadata specifying the license and the embargo expiration date, that is subject to the embargo. The version the author sent to the journal, even post peer-review, can be posted publicly and immediately, which wasn't always the case before…” Actually in the 2004-2012 Elsevier policy it was the case: Elsevier authors could post their post-peer-review versions publicly and immediately in their institutional repositories. http://users.ecs.soton.ac.uk/harnad/Hypermail/Amsci/3771.html This was then obfuscated by Elsevier from 2012-2014 with double-talk. http://openaccess.eprints.org/index.php?/archives/961-Some-Quaint-Elsevier-Tergiversation-on-Rights-Retention.html And it has now been formally embargoed in 2015. Elsevier authors can, however, post their post-peer-review versions publicly and immediately on their institutional home page or blog, as well as on Arxiv or RePeC, with an immediate CC-BY-NC-ND license. That does in fact amount to the same thing as the 2004-2012 policy (in fact better, because of the license). http://openaccess.eprints.org/index.php?/archives/1155-In-Defence-of-Elsevier.html But it is embedded in such a smoke-screen of double-talk and ambiguity that most authors and institutional OA policy-makers and repository-managers will be unable to understand and implement it. My main objection is to Elsevier’s smokescreen. This could all be stated and implemented so simply if Elsevier were acting in good faith. But to avoid any risk to itself, Elsevier prefers to keep research access at risk with complicated, confusing edicts.

For Harnad: Eisen is right, really. the reality that web savvy academics just post whatever version they can get away with on personal blogs or on Academia/Researchgate, as well as in other media. The nuances of Green vs Gold are 'academic' really. Many of us are well over Elsevier's business model, despite signing copyright waivers, and some of us will not even publish with them at all. I for example publish in Gold journals when I can: if I cannot, I make the text available wherever I can. I also devote considerable time and my weekends to running a Gold OA journal, which is I think the future model, rather than university Green archiving of preprints. The current publishing model would crash very quickly if we all behaved like this.