Elsevier, the Research Works Act and Open Access: where to now?

If Elsevier calculated that its withdrawal of support for the Research Works Act (RWA) would neutralise the arguments stirred up around academic publishing, I think the company is mistaken. I certainly hope so.

Things may have gone a little quiet on this issue in the blogosphere since the removal of the immediate cause of irritation, but there remain good reasons to keep looking hard at the vexed state of academic publishing, in particular at the issue of open access, and to keep thinking about how things might be improved.

I’m sure I am not the only one who has had his eyes opened by this affair. Though I haven’t written on the topic for a couple of weeks, I have still been engaged with it. As a working scientist, the issue of publishing just keeps cropping up. Having been made aware of the problems — and some of the associated complexities — I want to try to get to the end of my thought processes, however convoluted the journey.

The past

Elsevier abandoned support for the RWA at the end of February but did so rather grudgingly. There was little acknowledgement of the boycott and so the withdrawal struck many commentators as a tactical retreat by a company that had little notion of enacting real change.

Elsevier’s move may have cooled the argument a little but it is by no means dead. For one thing the number of signatories of the Cost of Knowledge petition has continued to rise — at the time of writing it stands at over 8200. I suspect the embers of the argument are being kept warm by discussions in common rooms and offices around the country. The publisher’s support for the RWA has had something of akin to the Streisand effect: I’m sure more people than ever are aware of open access and thinking more clearly about what it means for them.

The present

I find myself thinking differently. It would be an exaggeration to say I had been radicalised but my determination to seek out open access avenues in publishing my work is stronger than ever. To give just one example, I was invited a week or so ago to contribute an article to a special review issue being put together by an Elsevier journal. I was pleased to be asked — such invitations are a nice acknowledgement of your expertise — and normally I would have agreed without hesitation.

But this time I paused for thought. As is common in such situations, I would be expected to write the review without payment but now I asked myself why I should contribute to Elsevier’s very hefty profit margins by writing for free an article that would be held behind a paywall. So I replied to the editor to say I would only accept the invitation if Elsevier agreed to make the resulting review article freely available. I added that I would consider my labour as payment of any ‘author processing charge’. Unfortunately, we were not able to come to terms so I will not be contributing. Three months ago the outcome of this invitation would have been different.

The decision in this case was relatively straight-forward. This would have been a single-author paper since none of my present group is working in the subject area of the review. It would have been a more difficult choice had there been scope to involve other members of the group — I have their careers to consider. Nevertheless, I think that in future I shall always test the open access options before considering invitations to write reviews.

The future

The issue of open access should remain prominent over the coming months. In the UK the topic will get a fresh airing when the government’s Finch committee, set up last year to examine open access, reports later this Spring.

Independently, the UK research councils (RCUK) are reviewing their open access policies and just last week published draft proposals (PDF – have a look, it’s only 6 sides). While these new proposals don’t go as far as the open access policy of the Wellcome Trust, they represent significant movement in the right direction. The key changes from the current stance are:

  • Specifically stating that Open Access includes unrestricted use of manual and automated text and data mining tools; and unrestricted reuse of content with proper attribution.
  • Requiring publication in journals that meet Research Council ‘standards’ for Open Access.
  • No support for publisher embargoes of longer than six months from the date of publication

There’s a nice summary of the draft document on the Nature News blog and an initial critique here.

In the proposals support remains for the hybrid model, allowing either Gold open access, where authors pay ‘author processing charges’ and the publisher makes the final formatted version freely available immediately, or Green open access where an author-formatted version of the paper is provided free via ‘subject-based or institutional repositories’.

There are residual problems with the proposals that I hope can be cleared up in the consultation period before the new RCUK policy is set in stone. Firstly, the draft proposals retain the dual funding model for payment of open access charges — they can be charged directly to the grant if publication is within its lifetime or from indirect costs (overheads) paid to the host institution. As I have written already, this complex system does not work in practice. It will take good will and organisation on the part of universities and funders to enact a system that provides proper support. If this is to involve negotiations with each institution, it would appear to undercut RCUK’s aim to find ‘more efficient and effective mechanisms to support Open Access’.

The second major problem is likely to be agreeing what constitutes an allowable repository for those authors trying to go down the Green open access route (which has the advantage of not incurring any charges). I ran into difficulty on this route with a recent paper (published in Structure, an Elsevier title) because the staff running my institutional repository at first thought that my agreement with Elsevier did not allow me to deposit the full text of the paper. Happily, an email to the publisher resolved the problem within a few days (please have a read), but I was left wondering how many other articles have not appeared in institutional repositories because of similar misunderstandings (to say nothing of institutional or authorial lethargy over deposition).

A subject-based repository would have been a far preferable option since it makes papers easier to find. However, I discovered that my publishing agreement with Elsevier explicitly restricted me to my institution’s repository. I was debarred from sending my version of the paper to UK PubMed Central.

This regulation hinders access to the literature and therefore undermines the whole purpose of open access. PubMed, the first port of call for anyone searching the biomedical literature, frequently links to publisher’s site but never to institutional repositories. An avid reader seeking out my recent paper would easily be directed to the journal site where, without the benefit of a subscription, they would have to pay Elsevier’s $31.50 charge for 24 hours access (not even for a permanent copy!*). Alternatively, the poor reader would have to guess that there might be a copy at Imperial and would then have to try to find our repository.

I hope it is obvious that fragmentation of the literature into institutional repositories cannot be a desirable long-term solution to open access. I would urge RCUK to look closely at this aspect of their draft policy and to push as hard as possible for subject-based repositories.

I’ll stop there. I’m sorry to harp on about this and have already gone on at length. But needs must.


*Thanks to Alicia Wise of Elsevier for pointing out that the 24 hr access allows you to download a permanent PDF. Apologies for any confusion caused.

Update (09:00, 22 Mar 2012): There is a report of the negative reaction of the Publishers’ Association to the RCUK draft proposals in the Times Higher Education today — hat-tip to Prof. Martin Humphries at Manchester University (@HumphriesPrsnl). This is to me a good sign that RCUK is pushing in the right direction.

This entry was posted in Open Access, Scientific Life and tagged , , , . Bookmark the permalink.

37 Responses to Elsevier, the Research Works Act and Open Access: where to now?

  1. Benoit says:

    Yeah, I’m more committed to Open Access than I previously was, and will certainly strive to publish as much as possible in OA journals, or at least pay for OA options. Sadly, the CNS (Cell Nature Science) dominance is still prevalent, and it’s hard to NOT send manuscripts to those journals (esp with current grant climate!). Hopefully tides will turn and there will be a slow but inevitable shift, but it won’t happen overnight.

    • Stephen says:

      As far as I understand it Benoit, all three of those journals allow open access options (see Sherpa/RoMEO), though taking up those options may incur a hefty charge: $5000 at Cell and $3000 at Nature. Science permits deposition of the author-formatted version in PubMed Central 6 months after publication (not sure if there is any charge for that).

      I wonder how many people are put off by the issue of cost (a problem that funders should address) or simply by ignorance of the options available? It can be difficult to dig out this information. The Sherpa site linked to above, for example, did not provide any information on open access options available at Science.

      • Frank says:

        Stephen – minor correction: Nature does not have a paid OA option, but they will deposit the article into PubMedCentral for you with a 6-month embargo.

        My first port of call to check on journal policies is always the Wellcome Trust list of their top 200 journals‘ policies. MRC has a downloadable spreadsheet of frequently-used journals which also includes prices.

        If you go to the Romeo section of Sherpa you can search for any journal to find more information about its OA options.

        • Stephen says:

          Thanks for the the comment Frank, but Sherpa/Romeo lists Nature as a journal with “Author Pays Hybrid Model”.

          So, is their information incorrect?

          One of the problems for authors seeking Open Access options is the difficulty in finding these things out!

          • Frank Norman says:

            Stephen – I agree it does get confusing. The link you posted is to a page within Sherpa/Romeo titled “Publishers with Paid Options for Open Access”. It is a list of publishers, not of journals, and only of publishers that provide paid OA options, (which is the reason Science is not listed there).

            Rather misleadingly, the entry in that list for “Nature” links to a page about some of the Nature Publishing Group (NPG) journals, the ones they call their “academic journals”, that is the non-Nature-branded titles that they publish. These do indeed have paid OA options.

            The Nature-branded titles, and Nature itself, do not have paid OA options. Their rejection rate is so high that the fees would need to be set very high, so NPG say. (The exception is Nature Communications, which was explicitly started as a hybrid OA journal).

            I see that in the main Romeo section for Nature it links to that page of NPG academic journals too, which is rather misleading.

          • Stephen says:

            Thanks so much for that Frank. I’m glad someone is on top of this (even the misleading information). Horribly confusing!

      • Grant says:

        “I wonder how many people are put off by the issue of cost”

        I am. I (currently!*) work as an independent computational biology consultant. I have several things that I could in principle try work towards publishing but I can’t justify paying several thousand dollars for the ‘privilege’.

        (* I key an eye open for interesting jobs.)

  2. Pingback: Ninth Level Ireland » Blog Archive » Elsevier, the Research Works Act and Open Access: where to now?

  3. Pingback: Elsevier, the Research Works Act and Open Access: where to now? | Library Corner | Scoop.it

  4. RvD says:

    I have followed the whole Elsevier saga with interest, but as an outsider.

    How is it, in this age of open-source software and other free, community supported models that the scientific community still feels it necessary to publish the old-fashioned way? What are the obstacles to making the changes necessary to taking your rights as an author back?

    • Stephen says:

      That’s a very good question RvD, and one that I explored in a post last September.

      One of the main obstacles to establishing new models of scientific publishing is the inherent conservatism of the scientific community. We are, as Benoit alludes to above, wedded to our glamour mags, which have an established hierarchy that people use a shortcuts for the evaluation of the quality of science and scientists. Such evaluations are important since they can determine the outcome of applications for grants and promotion.

      There have been moves to break away — the PLoS stable of journals is one such — and the Wellcome Trust, has incorporated into its open access policy a statement that “affirms the principle that it is the intrinsic merit of the work, and not the title of the journal in which an author’s work is published, that should be considered in making funding decisions.” (see this recent post).

      If more funders would come out with such statements, and scientists themselves were prepared to judge their peers on more than just the impact factors of the journals they published in, we would be freer to develop newer, cheaper and more accessible publishing models.

      • It really is all about the journal hierarchy and it’s ‘proxy value’ as a perceived quality indicator.

        When you hear Benoit – who works at a top level Univ in a US system where funding is still relatively generous (if only in comparison to the deeply dire UK situation) – say that about feeling the need to go to the glamrags, then you know that it is trebly so down in the systems where funding is scarcer and the failure rate approaches 90%.

        One UK (Russell Group) lecturer I know told me that in the last decade something like 90% of all Faculty hires at their place had a recent paper in one of the glamrags on the CV that got them hired. It was widely believed (and I hear the same from other places) that that was because the appointment committees thought that having such papers was the best ‘signifier’ that the hire would be able to raise research grant funding.

        Even if committees and funders SAY that it is ‘science not journal’ – and I applaud them for doing so – people look at stuff like that and draw their own conclusions. And it will take an awful lot of water under the bridge to change that.

        • Stephen says:

          Perhaps I am being unrealistic in my hopes and expectations…

          But maybe it is better to try to uncouple the issue of open access from journal ranking (however determined)? After all, the glamour mags that Benoit refers to all have open access publishing options so there should (with proper funding and enforcement) be no excuse for authors not to make sure their work is freely available to readers.

  5. Alicia Wise says:

    Hi Stephen,

    Just to continue the comment thread about open access options, I would add that Elsevier has a variety of these and also an array of public access options. For example you mention some frustration at not being able to deposit in UKPMC, but we do have arrangements with a number of funding bodies that specifically enable posting in UKPMC (http://www.elsevier.com/wps/find/authorsview.authors/fundingbodyagreements). Also, you express some concern about linking from PubMed to journal pay-per-view sites. We offer an array of low cost public access options and it is a flourishing urban myth that people unaffiliated with a subscribing institution must pay $31.50 for 24 hours access! You can see examples of some public access options here: http://www.elsevier.com/wps/find/intro.cws_home/access_initiatives

    With kind wishes,

    Alicia Wise

    • Stephen says:

      Many thanks for your comment Alicia. You are quite right to point out that Elsevier has options that allow deposition in UKPMC. That this didn’t work in the case described in my post is unfortunately due to the unworkable nature of the present funding arrangements of the research council involved (the BBSRC). From the revised RCUK proposals, it looks as if their funding mechanisms may remain problematic, but we shall see.

      I’m afraid I have to take issue over your assertion that it is a myth that people without subscriptions have to pay $31.50 for short-term access. I tested the system from home (where I have no subscription without a VPN connection) using my own paper. From PubMed I got to the web-page for the journal, Structure and it was here that I was informed of the $31.50 charge for 24hr access. As far as I can remember, there were no obvious pointers to the alternative routes that you mention (but I could be wrong about that – I can’t test the system from work).

      But I have now looked at the other avenues mentioned in your link. I found that patientINFORM appears to be US-only so would be no good to a UK-based reader. If this is incorrect, the declaration in the heading of the web-page (“patientINFORM is a program that brings together the publishers of the world’s leading medical journals and the U.S.’s most trusted health organisations “) needs to be changed.

      I then tried DeepDyve but again, no luck. The journal Structure is not listed and I could not find the paper.

      So, although I appreciate these are genuine efforts by Elsevier to broaden access (to some groups — not every interested reader/taxpayer is also a patient), problems remain that are real and not mythical.

      • Alicia Wise says:

        Hi Stephen,

        I do think it is a myth that anyone without an affiliation to a subscribing library MUST pay $31.50 for 24 hours access. However, I also understand it is not always easy to know what the access alternatives are if you follow a link and are presented with only this option. Interestingly, I was talking with NIH colleagues only last week about link feeds and link resolvers – I was out of my technical depth, but clearly these conversations need to continue!

        Your various comments on our public access options are useful – thanks for continuing the conversation. patientINFORM is not only available in the US, but the patient organizations who initially partnered with publishers to set up this service were US-based. I can see why you might get the impression that this content is not available outside the US, but I can assure you this is not the case. There is at least one UK-based patient group participating, and I know more international partners are in the pipeline. The benefits for the patient organizations include free access to subscription based journals from participating publishers so they can keep up with the scholarly literature. And when they identify key articles they feel are particularly relevant to their constituents they can produce a reader-friendly summary and link to the full-text of the articles in ways that ensure users do not encounter paywalls. It’s a terrific initiative, and if you know of a patient advocacy group that might wish to participate I would encourage them to contact the team via http://www.patientinform.org/

        Re DeepDyve thanks for the suggestion of including the title Structure, and I will pass this on to our DeepDyve team.

        Re transactional sales – to clarify, when you do hit our PPV system and are asked to pay $31.50 you get a PDF which you can download and keep forever, and in addition you get 24 hour access to the HTML version which has added functionality. I know there are various purchase options/discounts for people who use PPV regularly. There are also various document delivery providers we work with, and possibly one of them could have provided an article from Structure to you quickly.

        To conclude, thank you for recognizing Elsevier’s genuine efforts to broaden access. We would not claim that everyone has access at the moment which is why we have a vision for a future time in which there is universal access. Our universal access programme reflects a commitment to the systematic identification of remaining access gaps and to filling these access gaps in sustainable ways (which include the methods we’ve been describing, open access, etc.)

        With kind wishes,


        • Stephen says:

          Dear Alicia

          Thanks again for the comment and for taking my comments on board. It would be good to see the info at patientINFORM clarified and to have some reassurance that the titles made available at deepDyve will be updated. It does seem odd that such a well-established journal as Structure would not already be on the list. 

          Also it would be helpful if links to these options (perhaps with an explanatory note) were included on the ScienceDirect page that you arrive at when trying to get to the full-text. Good information is important for promoting access. 

          Looking again at that page and re-reading the T&C about a PDF purchase I see that $31.50 would indeed, as you say, get me a PDF that I could print and keep. But given that the emphasis is on 24 hr access to the full-text and that PDFs can be time-limited, it is possible for mis-understanding to arise. Perhaps the message could be reworded to make it clear that you would be buying an electronic version that is not time-limited? Knocking the price down from the exorbitant $31.50 price point would also help!

  6. A very clear discussion of the issues involved and objective analysis of the problems yet to be solve. I have supported Open Access for a long time and I am pleased to see the support for it grow in the last few months.

    The discussion on where we go from here is just starting, I think and an open discussion between academics, funding bodies and publishers is the way forward.

  7. Nick Barnes says:

    Can you enlarge on exactly how you were prevented from depositing your paper in UKPMC? You say “I discovered that my publishing agreement that Elsevier explicitly restricted me to my institution’s repository. I was debarred from sending my version of the paper to UK PubMed Central.”
    I’m contributing to a critique of the RCUK draft policy, and I’d like it to ensure that this sort of control by publishers is impossible (i.e. that a journal which restricts the venues of green OA publication should not be considered ‘Open Access Policy Compliant’). The whole point of OA is that people (including the author) can do *what they want* with a paper, including depositing it in a repository of their choice.

    • Stephen says:

      Hi Nick.

      Happy to oblige. Elsevier’s terms and conditions applying to work funded by the BBSRC normally assumes that the funder will have paid the open access fee, in which case there is no problem: Elsevier will immediately deposit the paper in PMC.

      However, in my case, since there were no funds forthcoming (for the reasons I have given), I was restricted to depositing the ‘Accepted Author Manuscript’ with my institutional repository. The relevant paragraph from the terms and conditions is (with my emphasis in bold):

      There is no change to Elsevier’s author posting policy that allows authors to post revised personal versions of manuscripts (those that reflect changes made in the peer review) on their own websites and the sites of their institutions, provided a link to the journal is included. Posting directly to PMC or other sites outside an author’s institution continues to be prohibited, as does any further republishing or redistribution of Elsevier copyright-protected content and society copyright-protected content published by Elsevier. This new policy enables BBSRC-funded authors to comply with the BBSRC policy without having to violate their publishing agreements with Elsevier.

      So I was able to comply with the BBSRC’s open access policy but only via a repository that is not satisfactorily accessible. Even if you have paid the OA fee, I don’t think you can do what you want; Elsevier remains the copyright holder. However, if the condition that papers should be published under a CC-BY licence (as provided in RCUK’s new proposals), that should change things. No?

      • Nick Barnes says:

        Yes. Elsevier’s current green “open access” isn’t true open access at all, it’s merely a very limited gratis distribution license (which doesn’t extend any rights, for instance, to third parties). If the RCUK draft succeeds in forcing true “libre” BBB-compliant open access on the publishers, that situation will be remedied.

        • Alicia Wise says:

          Hi there,

          BBSRC expects the grant holder to pay a gold open access publishing fee from their grant, and to include these costs in grant applications. See http://www.bbsrc.ac.uk/web/FILES/Policies/access_research_outputs_update.pdf which states “where journals charge authors a fee for publishing open access articles, this should be identified as a cost via the fEC payment on a”BBSRC grant; staff in BBSRC-sponsored institutes should pay publishing fees from their strategic programme grants.”

          I imagine it could be rather difficult for a researcher to predict in advance how many papers will result from their research, where these might be published, and how much to budget. I would be interested to learn more from BBSRC grant recipients.

          If the researcher does not have sufficient funds in their grant it is my understanding that BBSRC may expect either the researcher or their institution to pay the gold open access publishing fee(s).

          These practical challenges perhaps help to explain why compliance with the BBSRC open access policy by authors who publish with Elsevier is less than 5% despite an agreement and very good will on all sides.

          With kind wishes,


  8. Pingback: Elsevier, the Research Works Act and Open Access: where to now? | Reciprocal Space | csid

  9. Benoit says:

    Cell doesn’t have OA option either. Some of their other journals have the option, but not Cell itself.

    And yeah, My lab still needs the glam mag pubs, but that’s a whole other story. It doesn’t matter if they are OA (but sadly they’re not), just that we get judged on what journal we publish in. Perhaps eLife will be as good as they hope, in which case I’ll be happy to send papers there.

    • Stephen says:

      Thanks Benoit – but are you sure?

      The situation is indeed confusing (see Frank’s informative comment above) but when I looked on RoMEO, it seemed to indicate that Cell allows a paid sponsorship option for open access (type Cell into the search box on this page).

      But maybe that only applies to NIH and HHMI funded labs in the US. No other funders are listed on Elsevier’s site.

      No wonder I’m confused…

  10. Niamh Brennan says:

    Great post, Stephen! Couple of comments:

    1. PMC should harvest the metadata associated with articles like yours – or your institutional repository should export the metadata to PMC. The actual paper stays in your institutional repository but it would be discoverable through its metadata in PMC. The technology and protocols are there – that’s what the Open Access Initiative Protocol for Metadata Harvesting (and OAI-PMH-compliant repositories like Imperial’s) is all about. The functionality is sadly under-utilised, but if academic authors (and funders) demanded it… (Examples of OA metadata harvesting in action: Open Access EC FP7 /ERC publications – OpenAIRE http://www.openaire.eu; Open Access publications in Irish Universities: RIAN: http://rian.ie; Open Access Research Theses: DART-Europe http://www.dart-europe.eu). I believe arXiv is an example of a subject specialist repository which can already harvest from institutional repositories. PMC & UKPMC have specific requirements regarding deposition of content – but presumably that could be worked out.

    2. If your paper is available on open access in your local repository it will be accessible to anyone via a simple search of Google and Google Scholar. But I’m sure you are aware of that.

    • Stephen says:

      Thanks for such an informative comment Niamh — the uploading of metadata would provide a useful bridge and make UK/PMC a much more useful resource. I wonder if a requirement for authors to upload that data might the incorporated into the revised version of the RCUK OA proposals (and those of any funder).

      I tried the google scholar search and it came up trumps! My paper was listed at the top; the link given is to the journal site but there is also a link (on the right) to the PDF.

      Google Scholar search

      Clever old Google. I shall definitely make more use of it in future trawls of the literature.

  11. Neil Stewart says:

    Further to Niamh’s excellent points about material in IRs being findable…

    As well as Google’s search engines, IR material is also harvested by specialised search engines such as BASE and OAIster (two search engines for open access material, very useful if e.g. you’re at a university with little in the way of journal subscriptions). IR data is also often used to populate other parts of institutions’ web presences-for example here at City we have IR data being used to populate a search of our publications, as well as feeding Library guides and Departmental publication pages. Different institutions will do things differently, of course- for example at LSE they use IR data to populate their well-used Experts profiles.

    I agree that IRs need to do more to aggregate their data (and have written about that here), but I would argue there are significant personal benefits in depositing locally (as well as in a subject archive)- and that’s before you look at the benefits for your home institution!

    • Stephen says:

      Thanks Neil – I confess I hadn’t heard of BASE or OAIster. From a first glance at their front pages, I’d say I could wish for a tad more information on what they contain but I shouldn’t issue any judgement on the basis of a few minutes exploration.

      It is good to know that people are thinking and working on how to aggregate the fragmented IRs. I’d be interested to know if the folks running my own university’s repository are doing anything to share their metadata and resources. I’ll ask.

      • Neil Stewart says:

        Librarians use (and advise others to use) the likes of BASE and OAIster when conducting literature reviews, because they often include material not available in e.g. WoS or Scopus. This includes less “prestigious” journal titles (which would be included in Google Scholar) and grey literature (e.g. working papers, discussion papers, reports, theses) that doesn’t get into GS.

        Incidentally, to add another point about IR material findability- we find that nearly 50% of our traffic comes from web searching, with hits to our IR resulting from over 400 search terms. Bigger repositories than ours find this figure rises to 70%+. IR material really does get found!

  12. nadg says:

    It is a frustrating situation, but there are other options for a sort-of open access. Many university libraries allow some access to the public to read papers on their terminals, and there are websites like http://www.reddit.com/r/scholar/ too. Either of which may technically be a minor violation of copyrights, if not handled correctly, but I feel that is a moot point on a small scale.


    PubMed & PubMed Central are wonderful resources, but not nearly as resourceful or wonderful as they easily could be.

    (1) PMC & UKPMC should of course be harvesting or linking institutional repository (IR) versions of papers, not just PMC/UKPMC-deposited and publisher-hosted papers.

    (2) Funders should be mandating IR deposit and PMC harvesting rather than direct PMC deposit. By thus making funder mandates and institutional mandates convergent and collaborative instead of divergent and competitive, this will motivate and facilitate adoption and compliance with institutional mandates: institutions are the universal providers of all research output, funded and unfunded.

    (3) IRs should mandate immediate deposit irrespective of publisher OA policy: If authors wish to honor publisher OA embargoes, they can set access to the deposit as Closed Access during the embargo and rely on providing almost-OA via the IR’s email eprint request button

    (4) Funder mandates should require deposit by the fundee — the one bound by the mandate — rather than by the publisher, who is not bound by the mandate, and indeed in conflict of interest with it.

    (5) Publishers (partly to protect from rival publisher free-loading, partly to discourage funder mandates, and partly out of simple misunderstanding of network capability) are much more likely to endorse immediate institutional self-archiving than institution-external deposit. This is yet another reason funders should mandate institutional deposit and metadata harvesting instead of direct institution-external deposit.

  14. Pingback: UK Government Goes For Broke on Open Access | Reciprocal Space

  15. Pingback: Is it a good thing that Elsevier bought Mendeley? — paidContent

  16. Pingback: Is it a good thing that Elsevier bought Mendeley? | BaciNews