[GOAL] Re: HEFCE Consultation on limiting submission to future REF to Open Access papers

[Stevan Harnad]

On Mon, Oct 7, 2013 at 4:52 PM, Graham Triggs <grahamtriggs at gmail.com>wrote:


> "The funding bodies propose the following criteria for open access"
> "We propose that outputs fulfilling the following definition must meet
> these criteria to be eligible for submission to the post-2014 REF...
> including the option of a percentage-based approach to compliance that
> would allow HEIs flexibility in preparing a submission to the post-2014 REF"
>
> Seems very clear that it is describing a criterion, not a mechanism.
>

No, *eligibility for submission* does not mean *criterion of assessment*.

Try it on this one in order to grasp the distinction:

"For eligibility, papers must henceforth be submitted in digital form, not
as hard copy."

 As the UKCRC/CPHC response itself states, "publication time" is extremely
> vague and variable, and can diverge by years from submission date.
>
> Publication time may significantly diverge from submission date, but it is
> not extremely vague - it's usually precise to a day, and if not a month.
> And more importantly, unlike the submission and acceptance dates, it is a
> matter of public record.
>

No, publication time for most journals corresponds to the date that appears
on the issue (which may be quarterly, or might specify a month). But that
calendar date (if any) often does not coincide with when the issue actually
appears, which can be much -- sometimes much, much -- later than the
calendar date (if any) (as many authors know). Nor does the author always
even get notified when his article actually appears, in real time.

In short, "publication date" is not only arbitrarily late if access
matters, but far too unpredictable to base a systematic policy of
time-based deposit (or compliance verification) upon. (Think about it.)

The only reliable, date-stamped marker for the author and the author's
institution, and hence for HEFCE and RCUK  (besides the submission date) is
the date of acceptance, on which the author will have a dated email or
letter informing him of acceptance.

Institutions, eager as always to meet REF procedures to the letter, will
(easily) find a systematic internal way to verify the deposit date
coincides with acceptance date.

Yes, but they are not (generally) a matter of public record. Which makes it
> very difficult to enforce a criterion based on them.
>

What *is* a matter of  "public record," as just noted, is a vague, late
date some arbitrary, varying and indeterminate time *after* the calendar
date (if any) of the journal issue.

And for a reliable, systematic mechanism for compliance with the HEFCE and
RCUK mandates what is needed is something with which the institution can
ensure timely compliance. That something is the author's dated acceptance
letter. (That also marks the point in the author's work-flow when it is
known that this is the accepted final draft, and can now be deposited.)

publisher constraints are on the version that is subject to a CTA, which is
> generally the author's final draft.
>

More publishers embargo their proprietary version-of-record than the
author's final draft.

And as to the embargoes on the author's final draft, I've already said that
the mandate's immediate-deposit clause (plus the repository's facilitated
eprint-request Button) moot the embargo: All accepted drafts can be
deposited immediately upon acceptance; at least 60% of them can be made
immediately OA; the remaining c. 40% can be made restricted access, and
users and authors can rely on the Button to provide Almost-OA during the
embargo.

The important thing is that the immediate-deposit clause makes it possible
for all institutions and all funders, worldwide, to ensure that all papers
are deposited immediately upon acceptance. It is a "keystroke mandate."

That done, the rest of the transition to 100% Green OA will be surprisingly
and refreshingly fast and easy. It is only this keystroke-barrier that
stands between us and 100% OA today.

A portion of the information - pagination, issue, publication date - will
> not be available at time of acceptance. And if it was really that painless
> for academics, a lot more would already be doing it.
>

Who cares?

The paper will be immediately accessible and the page information will be
available later.

Arxiv would only get you the pre-print, whereas the Hefce proposals call
> for it to be "made available as the final peer-reviewed text".
>

Arxiv authors do not deposit -- and have not been depositing, since the
beginning -- only their unrefereed drafts: they can make 5 updates, and
most include the final, refereed draft too. (Often in physics, though, the
refereeing does not introduce any substantive changes, I am told, so the
unrefereed preprint is the same as the refereed postprint.)


> Hefce does "propose that embargo periods are aligned with the Research
> Councils’ open access policy, and those endorsed by Government6. [Hefce]
> propose that the REF main panels will follow the embargo period set by the
> appropriate Research Council".
>

Who cares?

The RCUK will work out the issue of embargoes. (Right now it is, in its
wisdom, proposing to *ignore* them for at least 2 more years!)

Meanwhile, HEFCE/REF will ensure that all papers are deposited immediately.


> If a publisher chooses to not allow immediate deposit / adopts embargo
> periods that exceed that set by the appropriate Research Council, then it
> will constrain choice of journal.
>

No it won't. Authors will still publish in their journal of choice, and
deposit immediately upon acceptance (as dictated by HEFCE) and embargoes
will take care of themselves. (RCUK certainly won't be able to take care of
them!)

But one of the (several) reasons for author (and institution) rebellion
against the current Finch/RCUK policy will fall quietly by the wayside:
Authors will preserve their right to publish in whatever journal they
choose (and not have to pay to publish either, if they don't wish to).

Thanks to HEFCE/REF.

UKCRC/CPHC are absolutely correct to state a risk of constraint on journal
> choice, if there isn't a guarantee from the publishers that they will offer
> deposit terms in accordance with the Hefce proposals for at least the
> entire duration of the REF period.
>

For HEFCE/REF all that's needed is immediate deposit. Complying with
publisher embargoes is not a HEFCE/REF matter; nor is the length of what
RCUK ultimately deems to be the allowable embargo.

Meanwhile, the immediate-deposit clause (and the Button) moot the embargo
matter altogether.

So let's be neither pedants nor pessimists about all this: It's all heading
in a sensible direction, at last.

Stevan Harnad
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ecs.soton.ac.uk/pipermail/goal/attachments/20131007/9f52198c/attachment-0001.html