Information World Review article on RCUK

From: Stevan Harnad <harnad_at_ecs.soton.ac.uk>
Date: Thu, 13 Jul 2006 16:24:12 +0100

To Mark Chillingworth:

Just a couple of suggestions for your next article in Information
World Review:

(1) I think your current article

http://www.iwr.co.uk/information-world-review/news/2160288/rcukfails-stamp-open

occasionally conflates OA and OA publishing: The RCUK mandates have
nothing to do with OA publishing. They are mandates to self-archive
published articles (largely articles published in non-OA journals)
in order to maximize their usage and impact.

(2) The RCUK's failure to specify WHEN to self-archive is not just
the failure to pinpoint a cut-off date. The cut-off date should only
apply to the time when the deposited article must be made Open Access
(it can be Closed Access until then, if an embargo period is necessary),
whereas the deposit itself should be mandated to take place immediately
upon acceptance for publication. (This is an absolutely critical point,
and on it hinges the success of the RCUK mandates.)

(3) Your article also elided a bit the other question of WHERE to
deposit: The primary locus for deposit should be the researcher's own
Institutional Repository (IR). Central repositories can harvest it from
there on the appointed date, once it access is set to Open Access.

The reason this is so important is that the success of OA
self-archiving depends on all researchers doing it, systematically and
reliably. Virtually all UK researchers are at universities or research
institutions that already have an Institutional Repository. (For the
few that are not, SHERPA will have a generic repository they can use
instead.)

All papers should be deposited in the author's own Institutional
Repository (IR) immediately upon acceptance, because that milestone is
the onset of the growth tip for many research areas, the time when the
immediate uptake and usage is most important and productive; that is
when research grows and spreads.

If all papers are systematically and without exception deposited
in their own IRs immediately, the IR software makes it possible for
individual would-be users worldwide to generate a semi-automatic email
to the author to request an individual email copy of any article they
need that is still in Closed Access. This covers research needs until
any embargo period is over and the deposit is made OA. It is not OA,
but for the purposes of research progress, it is almost OA.

If the time and location of each deposit is instead arbitrary, variable
or unspecified, then we are and remain far further from the 100% OA
that is optimal and inevitable for research, within reach, and already
long overdue.

WHEN to deposit: immediately upon acceptance for publication
(access can be set immediately as OA or as Closed Access for the
allowable length of an embargo period, 6 months max.)

WHERE to deposit: in the author's own Institutional Repository
(can be harvested by central repositories if desired, later, at
the end of the allowable embargo period, 6 months max.)

Limits on the allowable length of the embargo period should be specified
too (preferably not more than 6 months after acceptance for
publication).

Some comments on your text:

> RCUK fails to date-stamp open access
>
> Further study commissioned by the Research Councils
> By Mark Chillingworth 13 Jul 2006
>
> A long-awaited position paper from the Research Councils UK (RCUK) has
> failed to clarify its position on free access to research information
> funded by British tax payers.
>
> Released by the Research Councils UK Executive Group, the paper makes no
> judgement on which publishing method authors should use

The RCUK policy is not about "publishing method" but about deposit and
access-provision requirements for articles published by any "method",
whether in an OA journal or a non-OA journal.

> and also fails
> to provide a clear time limit for when Research Council funded research
> should be made publicly available. â^À^ÜPublicly funded research must be
> made available and accessible for public examination as rapidly as
> practical,â^À^Ý it states.

The right compromise is to require deposit immediately upon acceptance
for publication in all articles, no exceptions. The metadata (author, date,
title, journal) are immediately visible webwide. but the article text
itself can, if necessary, be set as "Closed Access" for a limited period
(not exceeding 6 months).

> Instead of a clear time limit, the RCUK has instigated a new study on
> author-pays publishing and self-archiving.

Author-pays publishing has absolutely *nothing* to do with
self-archiving mandates. Self-archiving has already been going on for
over 15 years, and many studies have shown that its effect is to double
research usage and impact. The only further study that needs to be done is
on the effects of self-archiving mandates, but self-archiving mandates
can only be studied if they are tried. Hence the RCUK mandates
themselves will provide the evidence on the effect (if any) of
self-archiving mandates on publishing.

> RCUK has thrown its weight
> behind institutional repositories, however, saying: â^À^ÜFunded
> researchers should, where required to do so, deposit the outputs from
> research councils funded research in an acceptable repository.â^À^Ý

This is vague on both WHEN and WHERE.

> Rather than tie its colours to open access or traditional publishers,
> the position paper states that: â^À^ÜIt is for authors^À^Ù institutions
> to decide whether they are prepared to use funds for any page charges or
> other publishing fees.â^À^Ý

This is completely irrelevant to the RCUK policy, which is about OA
self-archiving mandates, not about OA publishing (let alone OA
publishing mandates!).

> Institutional repository and open-access campaigner Stevan Harnad has
> criticised the RCUK^À^Ùs stance, â^À^ÜThe mandates are still needlessly
> wishy-washy about one important thing: when the deposit must take
> place,â^À^Ý he told R. â^À^ÜNone of this is specific enough to be a
> clear, effective mandate.â

And about another important thing: WHERE the deposit must take place. It
should be in the author's own IR, immediately upon acceptance for
publication. The only allowable delay should be in the date when access
to the deposited full text is made OA. Preferably that too should be
immediate, but, if necessary, an embargo of no more than 6 months can
be allowed.

Stevan Harnad

PS If possible, I would be grateful if you would describe me either
as the moderator of the American Scientist Open Access Forum or as
Canada Research Chair in Cognitive Sciences at UQaM or as Professor of
Cognitive Sciences at the University of Southampton. It is true that
I am a long-time campaigner for IRs and OA, but I have been doing this
by commissioning and providing the specs for IR software (GNU EPrints),
publishing empirical evidence of the OA citation advantage, supervising
doctoral students who create software for generating citation links and
citation metrics, and advising on OA policy for research institutions,
funders and assessors. (I keep being characterised as a kind of
"cheer-leader" for OA, but I think what I'm actually doing is perhaps
just a tad more substantive than that...)
Received on Thu Jul 13 2006 - 17:00:15 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:48:26 GMT