Open Access Archivangelism

Estimating Open Access Mandate Effectiveness: I. The MELIBEA Score

Philippe Vincent-Lamarre, Jade Boivin, Yassine Gargouri, Vincent Lariviere, Stevan Harnad

ABSTRACT: MELIBEA is a Spanish database that uses a composite formula with eight weighted conditions to estimate the effectiveness of Open Access mandates (registered in ROARMAP). We analyzed 68 mandated institutions for publication years 2011-2013 to determine how well the MELIBEA score and its individual conditions predict what percentage of published articles indexed by Web of Knowledge is deposited in each institution's OA repository, and when. We found a small but significant positive correlation (0.18) between MELIBEA score and deposit percentage. We also found that for three of the eight MELIBEA conditions (deposit timing, internal use, and opt-outs), one value of each was strongly associated with deposit percentage or deposit latency (immediate deposit required, deposit required for performance evaluation, unconditional opt-out allowed for the OA requirement but no opt-out for deposit requirement). When we updated the initial values and weights of the MELIBEA formula for mandate effectiveness to reflect the empirical association we had found, the score's predictive power doubled (.36). There are not yet enough OA mandates to test further mandate conditions that might contribute to mandate effectiveness, but these findings already suggest that it would be useful for future mandates to adopt these three conditions so as to maximize their effectiveness, and thereby the growth of OA.

In response to the HEFCE Open Access Policy for REF2020, the Publishers Association has made the following counter-proposal, which is as predictable as the succession of night upon day:

1. Don't require deposit upon the date of acceptance, because publisher embargoes are timed to begin on the date of publication, not the date of acceptance.

2. Don't require deposit of the author's final draft; wait for the publisher's version of record.

3. Don't require deposit at all: let us take care of it, after our embargo has elapsed: then you don't have to worry about figuring out the date of publication.

4. Please send us a justification for wanting to do otherwise.

Now I find these four points so outrageous and presumptuous -- and so transparently self-interested, running roughshod over the interests of research, researchers, their institutions, their funders, and the tax-paying public -- that they don't need to be answered at all (and of course they need not and should not be heeded in any way).

But past experience has shown that the research community (like Schultz's Charlie Brown and the annual football that Lucy always manages to pull out from under him every time) somehow always manages to fall for publisher FUD. The FUD is invariably formulated to appear as if the publishers are just trying to help, and make things simpler and easier for the research community. But what keeps being missed by the naive research community is that the complications and difficulties that the publishers' proposals to "help" with are invariably complications and difficulties that are being imposed by the publishers themselves, in the form of embargoes and restrictions!

In the present case it's all about ensuring the one thing that is important to publishers about open access and open-access mandates: that publishers' embargoes on Green OA are faithfully obeyed (unless, even better, the author pays for immediate Gold OA). Nothing to do with ensuring OA, and as soon as possible.

The short answers to these four pieces of publisher FUD are, accordingly:

1. No thank you. We want deposit to be done as soon as the paper is refereed and accepted, because (i) that is when the paper is ready to be used by researchers, (ii) many publishers do not embargo OA, and (iii) even if there is a publisher OA embargo and an author wishes to comply, the author can provide individual copies to individual users with the repository's request-copy-Button immediately upon deposit during the embargo -- as long as the final draft is deposited immediately upon acceptance.

2. No thank you. The acceptance date is clear and certain for all papers, and it is the earliest access point; the date of publication is later, sometimes much, much later, and its timing is variable and indeterminate (cover date often does not correspond to date the issue appears). And besides coming too late, the publisher's version of record has more (publisher) restrictions on it.

3. No thank you. We would rather retain control of providing Button-access, and then OA, to our own final drafts; and we can track the embargo for ourselves, thank you very much.

4. No thank you. Please send us your justification for asking research institutions and funders to justify to publishers when they require their researchers to deposit their final drafts: We are not talking about OA or embargoes here but about deposit.

That said, here is some tit-for-tat for the Publishers Association's letter to HEFCE:

PA: "Following the publication of HEFCE’s statement of policy in March 2014 The Publishers Association and its members have been consulting with other stakeholders as to how best we in the publishing community can assist researchers and institutions to comply with the policy, particularly with regards to the deposit requirements.

With regards (and all due respect), the deposit requirements have nothing to do with publishers (or with publisher embargoes on OA).

PA: "In the course of these conversations it has become clear that this aspect of the policy is a source of widespread concern. Therefore I am writing to ask if HEFCE would give serious consideration to reviewing and revising certain elements of it."

No doubt there is publisher concern about the deposit requirement: That concern is precisely because deposits (unlike OA embargoes) cannot be controlled by publishers.

PA: "For us, the problematic section is Section 18 which states that “the output must have been deposited as soon after the point of acceptance as possible and no later than three months after this date (as given in the acceptance letter or email from the publications to the author)”.
Section 19 then goes on to require that the output must have been deposited as the author’s accepted and final peer-reviewed text, which may later be updated by the version of record.

Never mind the version of record. It's another matter. We're talking here about the author's final draft, and the time it is to be deposited.

PA: "For the avoidance of doubt, The PA and members are of course in no way opposed to the principle of deposit per se; rather we are concerned with the timing and form of the mandatory deposit."

Well, it's good to know that the PA are in no way opposed to something that is none of their business, and over which they have no control. But then why mention it at all?

PA: "In paragraph 29 of the Consultation Document on the Policy (July 2013) HEFCE stated that:
“We also wish to make the process of compliance as simple as possible for authors and HEIs and have received advice that the point of acceptance would be more suitable.” The PA’s response to the Consultation (October 2013) spoke to this point by saying “it is unclear where this advice has come from and what the justification for it is. To ensure compliance with embargo periods, which commence from the date of publication, it is logical to coincide the deposit or linking of papers with the publication date.”"

Is the PA owed a justification for a deposit policy that is none of its business and out of its control?

And why are publisher embargoes even being mentioned in this context? Yes, publisher OA embargoes are timed from date of publication (which varies wildly from paper to paper). But the issue here is deposit -- not publisher OA embargoes or their timing.

There is no link (logical or otherwise) between the deposit date, publication date, and embargo end-date.

PA: "Our subsequent discussions have confirmed us in our view that it would be far preferable for HEIs, researchers and publishers that the timing of the deposit of papers be coincident with the date of the publication of the version of record. This is for two principal reasons:
It would allow for the clearer management of embargo periods. Since these begin at the point of publication, deposit in institutional repository at this point removes the potential for any confusion arising between the availability of the accepted author manuscript and the version of record. Repositories will need to know the date of publication in order to respect the embargo period but they cannot know this from the date of acceptance."

As stated, OA embargo periods are reckoned from the (variable) publication date, not from deposit date. They have nothing to do with deposit date.

And OA is OA whether it is OA to the author's final draft or OA to the publisher's version of record (which has extra usage restrictions).

The HEFCE/REF mandate is to deposit the final draft upon acceptance and to provide OA after the allowable embargo at the latest. No later version is stipulated.

And the only version that is available until publication (and can hence be made OA or can be provided via the Button during an OA embargo) is the author's version.

PA: "It would reduce the level of costs to HEIs, many of which have expressed concerns about the financial and administrative burden of ingesting author manuscripts in their repositories. We understand that for many institutions this will require a number of operations (for example, subsequent up-dating of data) to be performed manually which could be fulfilled automatically if deposit of the author manuscript is made upon publication. Not all HEIs have repositories capable of dealing with this requirement at present."

It is terribly good of publishers to worry about institutions' operating expenses! There is a very useful way in which publishers could help lower these expenses: charge less for journals, or drop OA embargoes.

But no help is needed with deposit, thank you very much. Authors are quite capable of depositing for themselves, as well as reckoning OA embargo intervals.

PA: "Some stakeholders have made proposals for mitigating what we would see as the adverse effects of the proposed policy: for example encouraging publishers to provide author manuscripts and metadata in a standardised format through the Jisc “Publishers Router”.
However, this would not address the central issue of the clear potential for confusion around the start date of the embargo period, which often will not be known at the point of acceptance. For many publishers, this proposed mitigation would imply an additional cost, which seems somewhat unreasonable to ask publishers to bear when we believe that deposit at the point of publication would eliminate these costs and concerns without compromising the policy’s goals to advance open access."

To repeat: determining the date of publication and the date of the end of the embargo has nothing whatsoever to do with date of deposit.

It is clear that from the publishers' point of view, the ideal outcome (if there is to be Green OA at all) is that it should be the publisher (who knows the date of publication, who owns the version of record, and who imposes the OA embargo) who controls the deposit, its timing, and the timing of the embargo.

No, thank you. This is not about depositing as late as possible, and providing OA as late as possible, but about exactly the reverse.

PA: "Ultimately we believe that any of these mitigating or alternative solutions are poor substitutes for amending the underlying policy requirement.
It would greatly help our understanding of the HEFCE stance if you were able to share with us the justification for the policy – as referred to in last year’s consultation – and your perspective on the costs."

Publishers have no right to ask institutions and funders for "justification" of deposit policies if the deposits are not OA.

But if you want a general idea: it is to provide Open Access -- or at least Button access -- to refereed research as soon as the refereed final draft is ready.

PA: "We would be grateful if you would give consideration to amending this aspect of open access policy. We believe it will act as a hindrance to the widespread take up of open access in the UK research community, whereas a requirement to deposit in an institutional repository on publication would be a policy which would command support from a much wider range of sector stakeholders."

Good of publishers to be so concerned about compliance with the HEFCE deposit mandate, but, no, ceding control to publishers over deposit timing is not the way to ensure compliance. Making compliance a prerequisite for REF eligibility is.

Happily, HEFCE was unfazed by this FUD. See how they have advised the Publishers Association (rather more diplomatically than I have put it above) that, no thanks, we will stay the course.

Stevan Harnad