Dear Stevan,
I admire your vision and passion for green open access – in fact we all do here at Elsevier - and for your tenacity as your definitions and concepts of green open access have remain unchanged for more than 15 years. We also recognize that the open access landscape has changed dramatically over the last few years, for example with the emergence of Social Collaboration Networks. This refresh of our policy, the first since 2004, reflects what we are hearing from researchers and research institutions about how we can support their changing needs. We look forward to continuing input from and collaboration with the research community, and will continue to review and refine our policy.
Let me state clearly that we support both green and gold OA. Embargo periods have been used by us – and other publishers – for a very long time and are not new. The only thing that’s changed about IRs is our old policy said you had to have an agreement which included embargos, and the new policy is you don’t need to do an agreement provided you and your authors comply with the embargo period policy. It might be most constructive for people to just judge us based on reading through the policy and considering what we have said and are saying.
With kind wishes and good night,
Alicia Wise, Elsevier
Dear Alicia,
You wrote:
"This refresh of our policy [is| the first since 2004... Embargo periods have been used by us... for a very long time and are not new. The only thing that’s changed about IRs is our old policy said you had to have an agreement which included embargos..."
Is this the old policy that hasn't changed changed since 2004 (when Elsevier was still on the "side of the angels" insofar as Green OA was concerned) until the "refresh"? (I don't see any mention of embargoes in it...):
Date: Thu, 27 May 2004 03:09:39 +0100
From: "Hunter, Karen (ELS-US)"
To: "'harnad_at_ecs.soton.ac.uk'"
Cc: "Karssen, Zeger (ELS)" , "Bolman, Pieter (ELS)" , "Seeley, Mark (ELS)"
Subject: Re: Elsevier journal list
Stevan,
[H]ere is what we have decided on post-"prints" (i.e. published articles, whether published electronically or in print):
An author may post his version of the final paper on his personal web site and on his institution's web site (including its institutional respository). Each posting should include the article's citation and a link to the journal's home page (or the article's DOI). The author does not need our permission to do this, but any other posting (e.g. to a repository elsewhere) would require our permission. By "his version" we are referring to his Word or Tex file, not a PDF or HTML downloaded from ScienceDirect - but the author can update his version to reflect changes made during the refereeing and editing process. Elsevier will continue to be the single, definitive archive for the formal published version.
We will be gradually updating any public information on our policies (including our copyright forms and all information on our web site) to get it all consistent.
Karen Hunter
Senior Vice President, Strategy
Elsevier
+1-212-633-3787
Yes, the definition of authors providing free, immediate online access (Green OA self-archiving) has not changed since the online medium first made it possible. Neither has researchers’ need for it changed, nor its benefits to research.
What has changed is Elsevier policy -- in the direction of trying to embargo Green OA to ensure that it does not Elsevier's current revenue levels at any risk.
Elsevier did not try to embargo Green OA from 2004-2012 — but apparently only because they did not believe that authors would ever really bother to provide much Green OA, nor that their institutions and funders would ever bother to require them to provide it (for its benefits to research).
But for some reason Elsevier is not ready to admit that Elsevier has now decided to embargo Green OA purely to ensure that it does put Elsevier's current subscription revenue levels at any risk.
Instead, Elsevier wants to hold OA hostage to its current revenue levels -- by embargoing Green OA, with the payment of Fools-Gold OA publication fees the only alternative for immediate OA. This ensures that Elsevier's current revenue levels either remain unchanged, or increase.
But, for public-relations reasons, Elsevier prefers to try to portray this as all being done out of “fairness,” and to facilitate “sharing” (in the spirit of OA).
The “fairness” is to ensure that no institution is exempt from Elsevier’s Green OA embargoes.
And the “sharing” is the social sharing services like Mendeley (which Elsevier owns), about which Elsevier now believes (for the time being) that authors would not bother to use enough to put their current revenue levels at risk (and their institutions and funders cannot mandate that they use them) -- hence that that they would not pose a risk to Elsevier's current subscription revenue levels.
Yet another one of the “changes” with which Elsevier seems to be trying to promote sharing seems to be by trying to find a way to outlaw the institutional repositories’ "share button" (otherwise known as the “Fair-Dealing” Button).
So just as Elsevier is trying to claim credit for “allowing” authors to do “dark” (i.e., embargoed, non-OA) deposits, for which no publisher permission whatsoever is or ever was required, Elsevier now has its lawyers scrambling to find a formalizable way to make it appear as if Elsevier can forbid its authors to provide individual reprints to one another, as authors have been doing for six decades, under yet another new bogus formal pretext to make it appear sufficiently confusing and threatening to ensure that the responses to Elsevier author surveys (for its "evidence-based policy") continue to be sufficiently perplexed and meek to justify any double-talk in either Elsevier policy or Elsevier PR.
The one change in Elsevier policy that one can applaud, however (though here too the underlying intentions were far from benign), is the CC-BY-NC-ND license (unless Elsevier one day decides to back-pedal on that too too). That license is now not only allowed but required for any accepted paper that an author elects to self-archive.
Let me close by mentioning a few more of the howlers that keep making Elsevier's unending series of arbitrary contractual bug-fixes logically incoherent (i.e., self-contradictory) and technically nonsensical, hence moot, unenforceable, and eminently ignorable for anyone who takes a few moments to think instead of cringe. Elsevier is trying to use pseudo-legal words to squeeze the virtual genie (the Web) back into the physical bottle (the old, land-based, print-on-paper world):
Locus of deposit: Elsevier tries to make legal distinctions on "where" the author may make their papers (Green) OA on the Web: "You may post it here but not there." "Here" might be an institutional website, "there" may be a central website. "Here" might be an institutional author's homepage, "there" might be an institutional repository.
But do Elsevier's legal beagles ever stop to ask themselves what this all means, in the online medium? If you make your paper openly accessible anywhere at all on the web, it is openly accessible (and linkable and harvestable) from and to anywhere else on the Web. Google and google scholar will pick up the link, and so will a host of other harvesters and indexers. And users never go to the deposit site to seek a paper: They seek and find and link to it via the link harvesters and indexers. So locus restrictions are silly and completely empty in the virtual world.
The silliest of all is the injunction that "you may post it on your institutional home page but not your institutional repository." What nonsense! The institutional home page and the institutional repository are just tagged disk sectors and software functions, of the self-same institution. They are virtual entities, created by definition; one can be renamed as the other at any time. And their functionalities are completely swappable or integrable too. That too is a feature of the virtual world.
So all Elsevier is doing by treating these virtual entities as if they were physical ones (besides confusing and misleading their authors) is creating terminological nuisances, forcing system administrators to keep re-naming and re-assigning sectors and functions, needlessly, and vacuously, just to accommodate vacuous nuisance terminological stipulations.
(The same thing applies to "systematicity" and "aggregation," which I notice that Elsevier has since dropped as futile: The attempt had been to outlaw posting where the contents of a journal were being systematically aggregated, by analogy with a rival free-riding publisher systematically gathering together all the disparate papers in a journal so as to re-sell them at a cut-rate. Well not only is an institution no free-rising aggregator: all it is gathering its own paper output, published in multiple disparate journals. But, because of the virtual nature of the medium, it is in fact the Web itself that is systematically gathering all disparate papers together, wherever they happen to be hosted, using their metadata tags: author, title, journal, date, URL. The rest is all just software functionality. And if the full-text is out there, somewhere, anywhere, and it is OA, then there is no way to stop the rest of this very welcome and useful functionality.)
The Arxiv exception. In prior iterations of the policy, Elsevier tried (foolishly) to outlaw central deposit. They essentially tried to tell authors who had been making their papers OA in Arxiv since 1991 that they may no longer do that. Well, that did not go down very well, so those "legal" restrictions have now been replaced by the "Arxiv exception": Authors making their papers OA there (or in RePeC) are now officially exempt from the Elsevier OA embargo.
Well here we are again: an arbitrary Elsevier restriction on immediate-OA, based on locus of deposit. The Pandora's box that this immediately opens is that all a mandating institution need do in order to detoxify Elsevier's OA embargo completely is to mandate immediate (dark) deposit of all institutional output in the institutional repository alongside remote deposit in Arxiv (which is already automated through the SWORD software). That completely moots all Elsevier OA embargoes. Yet another example of Elsevier's ineffectual nuisance stipulations consisting of ad-hoc, pseudo-legal epicycles, all having one sole objective: to try to scare authors of doing anything that might possibly pose a risk to Elsevier's current revenue streams, using any words that will do the trick, even if only for a while, and even if they make no sense.
What's next, Elsevier? "You may use this software but not this software"?
The Share Button. Although it never defines what it means by "Share Button" (nor why it is trying to outlaw it), if what Elsevier means is the Institutional Repository's copy-request Button, intended to provide individual copies to individual copy-requestors, then this too is just a software-facilitated eprint request.
Whenever a user seeks an embargoed deposit, they can click the Button to send an email to the author to request a copy. The author need merely click a link in the email to authorize the software to send the copy.
So does Elsevier now want to make yet another nuisance stipulation, so the Button cannot be called a "Share Button," so instead the name of the author of the embargoed paper has to be made into an email link that notifies the author that the requestor seeks a copy, with the requestor's email alive, and clickable, so that inserting the embargoed paper's URL will attach one copy to the email?
Elsevier is not going to make many friends by trying to force its authors to do jump through gratuitous hoops in order to accommodate Elsevier's ever more arbitrary and absurd attempts to contain the virtual ether with arbitrary verbal hacks.
There are more. There are further nuisance tactics in the current iteration of Elsevier's charm initiative, in which self-serving restrictions keep being portrayed as Elsevier's honest attempts to facilitate rather than hamper sharing. One particularly interesting one that I have not yet deconstructed (but that the attentive reader of the latest Elsevier documentation will have detected) likewise moots all Elsevier OA embargoes even more conveniently than depositing all papers in Arxiv -- but I leave that as an exercise to the reader.
So Alicia, if Elsevier "admires [my] vision," let me invite you to consult with me about present and future OA policy conditions. I'll be happy to share with you which ones are logically incoherent and technically empty in today's virtual world. It could save Elsevier a lot of futile effort and save Elsevier authors from a lot of useless and increasingly arbitrary and annoying nuisance-rules.
Best wishes,
Stevan Harnad