What universities can do for open access
harnad at ecs.soton.ac.uk
Mon Nov 17 07:07:24 EST 2003
Peter Suber's list of suggestions in reply to Philip Pecorino's query is
excellent. I reproduce it in full, adding a few comments only on the minority
of points where I would demur:
> At 03:44 PM 11/12/2003 -0500, Philip Pecorino wrote:
> >What would it mean for a university or other academic institution to
> >support or "endorse" Open Access?
> I've been working on a long answer to this question. Here's the current
> draft. I welcome further comments. Note that it's only about
> universities, not learned societies, publishers, foundations, or other
> What faculty can do
> * Submit future articles to OA journals, when there are appropriate OA
> journals in their fields.
> * Deposit preprints in the open-access, OAI-compliant university archive.
> --If there isn't one already, another item below recommends making one.
> * Negotiate with journals either (1) to retain copyright and transfer only
> right of first print and electronic publication, or (2) to transfer
> copyright but retain right of postprint archiving.
No need to negotiate at all with the many journals that already support
And I note for reflection that the physicists in particular have been
systematically self-archiving since 1991 without any prior copyright
negotiations (or subsequent publisher objections). Twelve years and a
quarter million papers later, I think there is a lesson to be learned
there by those of us who have lost those same 12 years of potential
impact for our own work for various non-reasons, the belief that they
need to negotiate the right to self-archive their own papers being one
of them: http://www.eprints.org/self-faq/#1.Preservation
> --If faculty succeed, then they should deposit their postprint in the
> university archive alongside the preprint.
> --If the journal refuses, then faculty should ask for permission to deposit
> the postprint in the university archive. If it refuses, then they should
> ask for permission to put it on their personal web site. If it refuses,
> then they should post the corrigenda (differences between the preprint and
> postprint) to the archive.
Or, better, they should post their postprint (final revised, refereed
draft, not the publisher's PDF) without asking anyone anything (as the
physicists did and do) and wait to decide what if anything to do if
they are ever asked to remove it by the publisher. (Not a single one of the
250,000 physics eprints has been removed at the publisher's request in
12 years -- nor, so far as I know, has there been a single publisher
request to do so.)
> * Deposite data files in the university archive along with the articles
> built on them. If possible, cite the data files in the articles so that
> readers know where to find them.
> * Negotiate with conventional journals to try the Walker-Prosser method of
> experimenting with OA.
It is not clear how this fits into this sequence of options. The
Walker-Prosser method is for the author to pay the journal to self-archive
their paper for them.
When and why is it recommended that authors do that?
> * Consider launching OA journals in their areas of their expertise.
> * When asked to referee a paper or serve on the editorial board for an OA
> journal, accept the invitation.
> * When asked to referee a paper or serve on the editorial board for a
> toll-access journal, consider declining and explaining why.
I would definitely not recommend that researchers decline to edit or
referee for the 23,400 toll-access journals, and to agree to do so only
for the 600 open-access journals! A useful thing editors and referees
could do, however, would be to recommend and make it a condition for
offering them their services that the 45% of "white" journals become
"green" (officially self-archiving-friendly) and that the 35% that are
"blue" (preprint self-archiving) join the 20% that are "green" (postprint
self-archiving) -- which most of them probably are already, because what
they really mean by "preprint" is anything before the publisher's PDF.
The above, editors and referees can and should demand. In addition,
they might *recommend* that the journal consider becoming "gold"
(an open-access journal). But as that economic model is new and not
yet tested, it would be unreasonable for editors and referees to make
journals' taking that risk a condition for their services at this
time. Nor is it necessary to attain 100% open access.
For the record, if I were the editor of an established toll-access
journal at this time -- as I was for 25 years, until a year ago
http://www.ecs.soton.ac.uk/~harnad/Temp/bbs.valedict.html -- I
would definitely *not* recommend conversion to the open-access
model at this time. I would merely insist the journal become
green, and then wait and see what, if anything, happens:
For new journal start-ups, on the other hand, the open-access model may
counterbalance the risks (of not being able to make ends meet) with the
advantages (of a simpler, cheaper, and more impactful start-up in an
increasingly open-access-conscious time).
> --Faculty needn't donate their time and labor to journals that lock up
> their content behind access barriers where it is less useful to the
> profession. The university should support faculty who make this otherwise
> career-jeopardizing decision. Faculty don't need to boycott priced
> journals, but they don't need to assist them either.
The numbers -- 23,400 vs. 600 -- make this an impracticable decision at this
time. Nor is it necessary for 100% open access.
> * Faculty who already serve as editors of toll-access journals should start
> an in-house discussion about converting to OA or "declaring independence"
> (as this term is used by SPARC).
I would only suggest insistence on green at this time, definitely not insistence
> * When applying for research grants, ask the foundation for funds to pay
> the processing fees charged by OA journals.
AND recommend they mandate open access for funded research (by either road, green
> * Work with fellow members of professional societies to make sure they
> understand OA. Persuade the organization to make its own journals OA,
> endorse OA for other journals in the field, and support OA eprint archiving
> by all scholars in the field.
Persuade them to make their journals green, definitely not insisting on gold at
this time (though it can be recommended).
> --If the society launches a disciplinary eprint archive for the field,
> faculty could offer to have the university host it, just as as arXiv is
> hosted by Cornell.
Better still, all faculty should insist that their own institutions create and
mandate the filling of their own departmental eprint archives with all their
> * Create a online index or database of the OA sources in their field.
> --This could also be done by a professional association in the field.
> * Write opinion pieces (articles, op-eds, letters to the editor, discussion
> forum postings) advancing the cause of OA.
> --Or, at a minimum, don't let myths about OA circulate without challenge,
> e.g. that OA violates copyright, dispenses with peer review, or presupposes
> that journals have no expenses.
or needs to go straight for the gold today... ;>)
> * Educate the next generation of scientists and scholars about OA.
> What librarians can do
> * Launch an open-access, OAI-compliant institutional eprint archive, for
> both texts and data.
> --There are a handful of open-source packages for doing this. The three
> most important are eprints (from Southampton University), DSpace (from
> MIT), and CDSWare (from CERN).
And offer proxy self-archiving services for researchers who for some reason
cannot arrange to do it for themselves, e.g., as St. Andrews University does:
> * Join SPARC, a consortium of academic libraries crusading for OA.
> * Help OA journals launched at the university become known to other
> libraries, indexing services, potential funders, potential authors, and
> potential readers.
> * Include OA journals in the library catalog.
> * Offer to assure the long-term preservation of some specific body of OA
> --OA journals suffer from the perception that they cannot assure long-term
> preservation. Libraries can come to their rescue and negate this
> perception. For example, the National LIbrary of the Netherlands recently
> (9/17/03) agreed to do this for all 120+ BioMed Central journals.
> * Add metadata to OA articles and books that lack it.
> --OA content is much more useful when it is properly annotated with
> metadata. University librarians could start by helping their own faculty
> annotate their own OA works. But if they have time (or university funding)
> left over, then could help the cause by annotating other OA content as a
> public service.
> * Undertake digitization, access, and preservation projects not only for
> faculty, but for local groups, e.g. non-profits, community organizations,
> museums, galleries, libraries, Native Americans.
> * Educate faculty and administrators about the scholarly communication crisis.
and about access/impact loss, and what they can do to remedy it --
self-archiving their own work being the single most significant thing
that is within everyone's reach and would guarantee universal open access
overnight! Use these powerpoints to educate faculty and administrators:
> * Negotiate with vendors of priced electronic content (journals and
> databases) for full access by walk-in patrons.
> --An article in the 9/03 issue of Scientific American suggests that only a
> minority of libraries already do this.
> * Help design impact measurements that take advantage of the many new kinds
> of usage data available for OA sources.
And use the tools that have already been designed to do just that:
> --The OA world needs this and it seems that only librarians can deliver
> it. We need measures other than the standard impact factor. We need
> measures that are article-based (as opposed to journal or institution
> based), that don't oversimplify, that can be automated, and that take full
> advantage of the plethora of data available for OA resources unavailable
> for traditional print resources.
> --Librarians can also help pressure existing indices and impact measures to
> cover OA sources.
> What administrators can do
> * Adopt a policy: In hiring, promotion, and tenure, due weight will be
> given to all peer-reviewed publications, regardless of price or medium. Or
> even more strongly, articles in OA journals and articles on deposit in the
> university's OA archive, with a working URL in the resume, will count more
> heavily in the P&T review than other articles. (This policy would not
> affect books or other kinds of output at all.)
> --Let current and prospective employees know about this policy.
I regret to have to say that this recommended policy sounds neither
practical nor just. (Moreover, it is unnecessary.) In hiring, promotion,
and tenure, due weight should be given *only* to the indicators of quality
and impact, as is already the case. No special weight should be given --
between two works of equal quality and impact -- to the one that appears
in an open-access journal or has been self-archived! That is an arbitrary
intrusion of non-quality/impact criteria where they do not belong. (To
see this, imagine an asymmetric case: one candidate's work is *lower*
in quality/impact, but it is open-access: How much extra "weight" should be
assigned to it for being open-access, thereby offsetting the higher
quality/impact of the toll-access work of another candidate?)
No, a far more sensible and justifiable policy is one that simply mandates
open access-provision for all faculty research output (by either the
gold or the green road) in order to maximize impact: a simple and natural
extension of the existing "publish or perish" rule, not calling for artificially
scaling actual quality/impact measures down or up. Because of the demonstrated
impact-enhancing effect of open-access, the result of such a policy will be
genuine enhanced impact, without the need for any artificial weighting change.
> * If the policy above takes one of the weaker rather than one of the
> stronger forms, then find another way to encourage or compel faculty to
> fill the OA eprint archive.
Encouraging is one thing, mandating is another. Changing research
quality-evaluation criteria to reflect non-quality factors is yet another.
> --This is where most universities with an institutional archive fall
> down. They build it but then do nothing to fill it.
> --One way is to pay for a digital librarian (whole or fractional FTE) to
> help faculty put their past publications into digital form, deposit them in
> the university archive, and enter the relevant metadata. Many OA-friendly
> faculty are simply too busy to do this for themselves.
> * Adopt a new policy: faculty who publish articles must either (1) retain
> copyright, and transfer only the right of first print and electronic
> publication, or (2) transfer copyright but retain the right of postprint
Simpler way to put it (as the "must" above is already a mandating policy):
Mandate that faculty research output most be open-access (and leave it
to the faculty to decide what road to take to achieve this)!
> --The University of Kansas has language that other universities could
> borrow or adapt for this purpose. I believe that Kansas requires its
> faculty to insert the language into all copyright transfer agreements with
> * Adopt a new policy: when faculty can't get the funds to pay the
> dissemination fee charged by an OA journal from their research grant, then
> the university will pay the fee.
> --If the university is worried about a runaway expense, it could cap the
> number of dollars or articles per faculty member, and raise the cap over
> time as the spread of OA causes the library to realize larger and larger
> * Support, even reward, faculty who launch OA journals.
> --For example: give them released time, technical support, server space,
> secretarial time, P&T credit, publicity, strokes.
> --Related: give due recognition to faculty who serve as editors or
> referees for OA journals, at least if this recognition is given for similar
> service on important traditional journals. Most OA journals, because they
> are new, haven't acquired the prestige of established, conventional
> journals. Universities should support faculty who help bring about a
> superior publishing alternative, not just those who bring prestige to
> themselves and the university.
> * Consider buying an institutional membership in BMC and PLoS.
> * Sign the BOAI as an institution.
> * There should be some outreach to the community. For example (as
> mentioned above), the university should invite community groups to use its
> OA archive. It should offer to digitize, host, and preserve content for
> some non-profit organizations in the region.
> * Public universities should explain to the citizens of their state, to
> state legislators, to state newspapers, why the new OA policies are
> maximizing the return on tax dollars, and how they put the university in
> the vanguard of enlightened institutions. Private institutions can make
> the same argument to donors, parents, and students.
> * If a university adopts a coordinated plan to promote access, through its
> faculty, librarians, and administration, then it should launch a central
> web site for the plan, and perhaps a newsletter, to explain its many
> facets, monitor progress, publicize the rationale, and show which elements
> are still to come.
> * On funding this grand plan: Many parts of the plan are either costless
> or result in net savings. Many others will bring waves of good publicity,
> which will help the bottom line through improved recruitment and retention,
> soft money, or alumni loyalty.
> Peter Suber
> Research Professor of Philosophy, Earlham College
> Open Access Project Director, Public Knowledge
> Editor, Open Access News blog
> Author, SPARC Open Access Newsletter
> peter.suber at earlham.edu
More information about the Jrnlnote