Re: On Distinguishing Open Access Self-Archiving from Open Access Journal Publishing

From: Stevan Harnad <harnad_at_ecs.soton.ac.uk>
Date: Sat, 30 Oct 2004 20:30:15 +0100 (BST)

On Fri, 29 Oct 2004, Prof. Tom Wilson wrote:

> > You might have saved some effort by looking at the following,
> > which charts the growth of about 220 archives, 144 of them GNU Eprints:
> >
> > http://archives.eprints.org/eprints.php?page=all
>
> Unfortunately, I did not find the archives part of the site to be readily
> discoverable from the basic www.eprints.org site - and when I do look at it,
> its organization is far from helpful for my purpose. I was interested in which
> universities in the UK are implementing institutional archives - discovering
> that from this list is hardly made easy.

Prof. Wilson does not find the explicit link from the Eprints site to

   "Registry of OA Eprint Archives"
    http://archives.eprints.org/

readily discoverable. But it is prominent in the top banner of the eprints.org
home page!

    http://www.eprints.org/

And it is not at all clear why from

    http://archives.eprints.org/index.php?action=browse

Prof. Wilson could not discover the universities in the UK implementing
institutional archives:

     Country

    * United States (58)
    * United Kingdom (33)
    * Canada (20)
    * Germany (15)
    * France (15)
    * Sweden (13)
    * Netherlands (12)
    ... etc.

> > Interesting report, but not clear why you focussed on university-wide
> > archives
> > rather than departmental ones (since departments are parts of universities,
> > the eprints software is deliberately modular, and university self-archiving
> > policies are best implemented at a departmental level!):
> >
> > http://software.eprints.org/handbook/departments.php
>
> I could hardly disagree more - a self-archiving policy is hardly a convincing
> policy if it operates only at the departmental level. The success of any
> information policy is highly dependent upon being driven from the top and until
> there is university-wide commitment to a policy, together with the strategies
> to drive that policy throughout the organization, commitment is likely to be
> piecemeal and unsatisfactory.

It is interesting to note the degree of confidence underlying Prof. Wilson
disagreement. One is tempted to ask "compared to what?" he is making
this confident judgement: Is it based on his empirical evidence on
the relative frequency of institution-wide vs. department-based
self-archiving policies, and their relative success in generating
self-archived articles? Is it based on his analysis and findings on
the relative efficacy of self-archiving policies implemented at the
central university level versus the departmental level? Is it based on
his analysis and understanding of what self-archiving actually amounts
to, in technical and practical terms, including the creation of the
archives and the implementation of the policy?

These are all matters worth looking into, but I see very little evidence
of Prof. Wilson's actually having looked into them. He seems to be making
judgements here on the basis of a priori notions about "any information
policy." Prof. Wilson's notions may prove right, but such notions first
have to be tested, because self-archiving and open-access (OA) are
something rather new, and may in fact not be best thought of as just
another "information policy". In fact, OA provision may be more closely
related to procedures like the UK Research Assessment Exercise (RAE)
than to an "information policy"; and, as we know, the RAE returns
are implemented departmentally, not centrally.

But although I am by now alas far more experienced in these matters than
Professor Wilson appears to be (he still has a bit of savvy to pick up
about how to discover web pages, and how to discover links on web pages,
not to mention a lot of existing data on self-archiving and its current
implementations, successes and failures, worldwide!), I would say that
it is far too early to state things with the conviction expressed by
Professor Wilson. He may well be right that a university-level commitment is
necessary for a university self-archiving policy, but he may be missing
the fact that to persuade a university of the virtues of committing
itself to a self-archiving policy, it may first require successful examples
from departments that actually go ahead and do it.

In his report (from which all further indented passages below that are
in quotation marks are drawn):

   "Open archives and open journals"
   http://www.informationr.net/tdw/publ/papers/2004OArchives.html

I had written that Prof. Wilson seems to

>sh> (1) confuse self-archiving with self-publishing
>sh> (publishing is done in peer-reviewed journals, open access to that
>sh> publication is provided by self-archiving it) and to
> >
>sh> (2) confuse open-access self-archiving ("green") with open-access
>sh> publishing ("gold")
>sh> (universities are not *publishing* the articles their authors
>sh> self-archive, they are merely maximising their access and impact be
>sh> making them open-access).

to which Prof. Wilson replied:

> I am not in the slightest confused - nor do I think my report conveys any such
> confusion.

Ad (1) (confusing self-archiving with self-publishing):

> "Academics publish and the problem with the concept of an archive
> is that it is generally perceived as a mode of preservation, not a
> mode of publishing."

Self-archiving of published journal articles is *not* a mode publishing,
it is a mode of supplementing access to the publication. The publication
is done (as always) in peer-reviewed journals.

(Nor, by the way, is the self-archiving of published journal articles
a mode of preservation of journal articles; it is an author-provided
OA supplement, for those would-be users whose institutions cannot afford
to access the journal's toll-access version; it is the journal's version
of the articles that needs the preservation, not the author's supplement.
This is not to say that the supplements cannot and will not be safely
kept too; just that it is a misnomer and a misdirection of effort to
imagine that self-archiving is done for the sake of solving the
journal article's long-term digital preservation problem, rather than for the
sake of supplementing access so as to maximise impact, now.)

    http://www.jisc.ac.uk/index.cfm?name=programme_404

Ad (2) (confusing OA self-archiving with OA publishing):

> "In any event universities, collectively, in any country, are
> strong enough to act as publishers -- indeed, many of them do act as
> publishers and I am baffled by the failure of the Higher Education
> Funding Councils in the UK, the Joint Information Systems Committee
> (JISC) of those Councils and university heads in general to grasp
> the opportunity for collective, open-access publishing that the
> Web offers."

Self-archiving of published journal articles is not a mode publishing,
it is a mode of supplementing access to the publication. The publication
is done (as always) in peer-reviewed journals.

There are also OA journals now, that make all their contents OA; for
these journals (only), authors need not supplement by providing their
own self-archived OA version. However, only 5% of journals are such OA
journals. So 95% of journal articles do need self-archiving if they are
to be made OA. The author's institution, in hosting the OA Eprint Archive
with its self-archived articles is not thereby becoming an open-access
publisher.

If universities wish to become open-access publishers, they can of course
do so, but that has nothing whatsoever to do with the self-archiving of
the articles that their own researchers already publish in existing journals.

> "The scale of the effort needed, spread over more than 100
> institutions would be achievable for virtually any field of research
> and their effort could be supported by the research councils, as a
> condition of making the award of a grant, requiring researchers to
> publish in, say, the relevant, open access, 'British Journal of...' ,
> before they published anywhere else."

This rather radical proposal by Prof. Wilson seems to be that:

    (i) rather than mandating that UK researchers must self-archive
    supplementary versions of their journal articles so as to make them
    OA (as the UK Select Committee has wisely recommended),

    http://www.publications.parliament.uk/pa/cm200304/cmselect/cmsctech/399/39903.htm

    (ii) the UK should instead mandate that UK researchers must publish
    in OA journals (5% of existing journals), rather than the journals
    they now choose to publish in (as the UK Select Committee has wisely
    *not* recommended).

I think this is abundant evidence that Prof. Wilson is indeed confusing
self-archiving with self-publishing, as well as OA self-archiving with
OA journal publishing -- confusing them both with respect to means and
with respect to ends.

It remains to ask *why* Prof. Wilson would make such a radical proposal
as the above -- to the effect that the UK should create and fund a new
fleet of UK OA journals and require UK researchers to publish in them,
rather than just requiring them to self-archive the articles they already
publish in the existing journals of their choice? Surely the radicality
of his proposal is not just a consequence of the importance and urgency
of OA, for OA can already be provided by far less radical means. But
Prof. Wilson has already dismissed those less radical means. Why? Let us
look for the reason [the interpolations are mine]:

> "Is self-archiving working? We have heard much about the Open
> Archives Initiative and of the virtues of self-archiving over other
> methods of open access publishing [sic], but what is going on? After
> a certain amount of labour [vide supra] I discovered a list of 10 institutions
> using the e-Prints software from Southampton... I've ignored separate
> departmental... archives [vide supra], but concentrated on those that
> aim to cover an entire institution or... more than one institution....
>
> "I've also shown the percentage contributed to the total by the
> archive at the University of Southampton, as that contribution is
> highly significant, accounting over all for more than 70% of the
> total. As the originator of the software, this is not surprising,
> since the institution will have to make some effort to provide a
> basis for encouraging others to use that software."

Before going on, I have to point out that Prof. Wilson is making
the wrong inference here: Southampton did not create (and give away)
its open source software in order to encourage others to "use that
software" but in order to encourage others to self-archive!

    http://www.arl.org/sparc/pubs/enews/aug01.html#6
    http://www.dlib.org/dlib/october00/10inbrief.html#HARNAD

Moreover, Southampton is not designing and implementing self-archiving
policy in order to promote the eprints software but in order to promote
self-archiving and thereby open access.

    http://software.eprints.org/handbook/departments.php
    http://www.eprints.org/signup/sign.php

It might hence be a good idea, after all, to give a little more thought
and attention to the Southampton departmental self-archiving policy model,
and the reasons it has proved successful, rather than dismissing it as
a ploy for promoting Southampton software!

> "For the rest, however, institutional archive appears not to have
> been a very great success (the English mastery of understatement
> is at play here)."

I couldn't agree more! Eprint Archives are created, but they mostly lie empty,
for lack of an institutional or departmental self-archiving policy or
mandate. That is why we are working so hard on promoting both the policy and
the mandate. Is Prof. Wilson aware of the sizeable ongoing collaborative
scientometric project measuring and documenting the degree to which
OA enhances research impact? Those accumulating data, and their
dissemination, we hope, will help encourage both self-archiving and
self-archiving mandates:

    http://opcit.eprints.org/oacitation-biblio.html
    http://citebase.eprints.org/isi_study/
    http://www.crsc.uqam.ca/lab/chawki/OA_NOA_biologie.gif
    http://www.crsc.uqam.ca/lab/chawki/sociologie.htm
    http://physnet.uni-oldenburg.de/~stamer/isi_study/

> "So what has happened to this movement? Clearly, with considerable
> motivation, Southampton is doing something right - but what? How is
> it encouraging participation from the academic staff?...

See: http://software.eprints.org/handbook/departments.php
    
The movement is alive and well, and growing, especially with the prospect of a
self-archiving mandate, already recommended (but not yet implemented!) in the
UK and US, and currently also being considered by France, Germany, Australia,
Canada, Norway, Switzerland, Japan, India and other countries.

    http://www.eprints.org/signup/sign.php

> "...An examination of the Faculty contributions shows... a discipline
> effect, rather than an overall university policy - out of the 844
> items contributed from the Faculty of Engineering, etc., 665 (79%)
> are from the School of Ocean & Earth Science, and the Southampton
> Oceanography Centre contributes another 763. There's no information
> that could tell us why academics in the earth and ocean sciences
> are so keen to contribute. The zero returns from the School of
> Humanities and the School of Social Sciences are rather telling -
> how are academics in these areas to be persuaded?"

The answer is that only one Southampton department so far has adopted
an official self-archiving policy -- Electronics and Computer Science
(the creators of the software). And Oceanography is working in
conjunction with the TARDIS project, which is dedicated to designing
university self-archiving policies, under the associated JISC FAIR
programme, which is also devoted to the same issue, as is EPRINTS-UK,
and SHERPA

    http://tardis.eprints.org
    http://www.jisc.ac.uk/index.cfm?name=programme_fair
    http://www.rdn.ac.uk/projects/eprints-uk/
    http://www.sherpa.ac.uk/

A good deal of research and development is going on in this area,
research whose findings might have informed Prof. Wilson's own
efforts in this area. (For the data on the OA impact advantage
in the Humanities, stay tuned!)

> "By any measure it can hardly be claimed that the concept of open
> archiving has taken off in British universities...we seem to have
> possibly thirteen or fourteen universities in the UK involved in
> archiving - 11% or 12% of the total."

I agree entirely that self-archiving has not only not taken off in British
universities, but that it has not yet taken off anywhere: Only about 20% of
the annual journal literature is being self-archived today (but the
growing evidence of the OA impact advantage, and the prospect of
self-archiving mandates, is around the corner!).

The correct figures for the UK, though (see list near top of this
message), are that it is second in the world in number of Eprint Archives
(33), second only to the US (58), but with only 73 research universities,
that almost certainly means the UK is leading the world in the relative
*number* of Eprint Archives. (Filling them is another matter, and
there the heroes are as yet few, anywhere: the half dozen who have signed
http://www.eprints.org/signup/fulllist.php plus CalTech and a few other
archives around the world.)

But now we come to the crux of Prof. Wilson's argument:

> "...those institutions that [have archives] appear to be having
> difficulty in getting academics to contribute, [1] perhaps because they
> are putting insufficient effort into the process, but also, [2] perhaps,
> because the whole idea of self-archiving in institutional archives is
> based upon false assumptions about the behaviour of academic authors."

This pair of candidate reasons why we do not yet have 100% OA
self-archiving probably does cover the two major possibilities, but
unfortunately Prof. Wilson presses on from this point on the assumption
(unexplained, and without any supporting evidence, either empirical or
logical) that the correct reason is in fact [2]: self-archiving is based
on "false assumptions".

Which false assumptions? And be careful, in replying, that the falsity of those
assumptions should not be such that it would have by the same token
made us discard the "publish-or-perish" mandate too, as incompatible
with the behaviour of academic authors! For self-archiving to maximise
research impact is merely a natural, online-age extension of the existing
and universal academic mandate, which requires that you must make your
scholarly or scientific research findings public -- by publishing them,
rather than merely putting them in a desk-drawer -- so other scholars
and scientists can use and build upon them. The self-archiving mandate
merely updates this existing mandate for the online era, taking into
account one highly pertinent new fact: that if the impact of refereed
research remains restricted to only those users whose institutions can
afford the access-tolls for the journal in which the article appears
then, today, that just amounts to putting it into a bigger desk-drawer,
now that the online medium has made it possible instead to make it openly
accessible to all would-be users.

False assumption? that researchers, their employers and funders are capable
of understanding and acting upon that new information?

I think Prof. Wilson's first candidate explanation for the sluggish growth
of OA self-archiving so far -- [1] that it is because institutions are
putting insufficient effort into the process -- is the correct one,
and the remedy is more effort!

> > Please look at the proposed self-archiving mandates and that may help
> > sort these things out more coherently: Universities provide OAI-compliant,
> > interoperable archives. Mandatory self-archiving policies ensure the
> > archives are filled, and researchers, their institutions, their funders
> > and research as a whole is the beneficiary of the maximised research
> > access and impact that this affords.
>
> You miss the point - deliberately, perhaps? My point is that self-archiving
> demands a change in user behaviour and, as Leslie Carr points out, involves "An
> enormous uphill struggle to engage the academic and research communities with
> the issues in order to change their personal and institutional practices." The
> history of technological developments is riddled with projects that are
> technology-led and in opposition to the established routines and behaviour of
> target groups. There is a danger that self-archiving could be perceived by its
> intended target group, academic authors, as similarly at odds with their normal
> practice.

This is far too abstract: Let us talk about what is actually at issue
here, in specific, concrete, practical terms. Here is a sample datum:
34,000 biomedical researchers worldwide showed themselves to be willing to
perform the few keystrokes required to sign an Open Letter to the effect
that they would no longer publish in, referee for or use journals that did
not make all of their contents Open Access within 6 months of publication.

    http://www.plos.org/about/letter.html

I would say that this boycott threat constitutes evidence that the
desire for OA itself is not at odds with normal academic practice. But
now it gets interesting:

When the deadline went by and the journals declined to do what the authors
had demanded, what did the authors do? They just kept on publishing in,
refereeing for and using those same journals. What choice did they have?

Yet with not many more keystrokes than it took to sign the Open Letter,
they could each have self-archived one of their articles. Just those
keystrokes would have given us 34,000 more of what they were petitioning
for in the first place!

Was this a technological barrier? Hardly, since these were all
keystroke-capable academics. Was it for lack of archives to self-archive
in? Fine, let's just ask about that subset of institutions or departments
that already have (near-empty archives): Were there technology barriers
there too, to performing those few keystrokes?

No, the barriers are not technological: They were merely inertial. The
cure for those inertial barriers to a desideratum that academics
themselves recognize as a desideratum is the carrot/stick of
publish-or-perish policy (publication being likewise a desideratum whose
value academics recognize, yet they still have to be carrot-sticked
into actually doing it!). Research impact is already a criterion
for employment, promotion and funding. OA is being demonstrated
quantitatively to enhance research impact. It is just one short policy
step to formulating a formal requirement where there is already such abundant
evidence for causality and benefit:

     Harnad, S., Carr, L., Brody, T. & Oppenheim, C. (2003) Mandated
     online RAE CVs Linked to University Eprint Archives: Improving the
     UK Research Assessment Exercise whilst making it cheaper and easier.
     Ariadne 35 (April 2003). http://www.ariadne.ac.uk/issue35/harnad/

> I very much doubt that, without effective strategies to motivate academic
> authors, insitutional commitment to self-archiving mandates is likely to remain
> a paper commitment. I would be interested to hear from institutions as to the
> nature of whatever successful strategies they have evolved.

You are quite right. Effective strategies to motivate academic authors
to self-archive are needed. But the components are already in place:
the evidence of the impact-enhancing power of OA, the carrot/stick
of publish/perish performance evaluation, increasingly weighted by
publication impact rather than mere publication counts, the ease of
creating OA archives, and of performing the few keystrokes per paper
needed to fill them, and the prospect of not being funded or promoted
if you do not maximise the impact of your publications.

(But if you think it was hard to persuade academics to perform the few
extra keystrokes it takes to self-archive a paper, wait till you try instead
to persuade funders to create 24,000 new journals to OA, and then persuade the
academics to give up the journals of their choice and publish instead in
those new journals! -- All instead of performing those few extra keystrokes!)

I note that at the bottom of Prof. Wilson's paper it says:

    "How to cite this paper

    Wilson, T.D. Open archives and open
    journals. , 10(1), 2004. 37-48 [Available at
    http://informationr.net/tdw/publ/papers/2004OArchives.html]

Was the journal-name inadvertently omitted here, or is Prof. Wilson
also advocating unrefereed self-publication?

Stevan Harnad

AMERICAN SCIENTIST OPEN ACCESS FORUM:
A complete Hypermail archive of the ongoing discussion of providing
open access to the peer-reviewed research literature online (1998-2004)
is available at:
    http://www.cogsci.soton.ac.uk/~harnad/Hypermail/Amsci/index.html
        To join or leave the Forum or change your subscription address:
http://amsci-forum.amsci.org/archives/American-Scientist-Open-Access-Forum.html
        Post discussion to:
    american-scientist-open-access-forum_at_amsci.org

UNIVERSITIES: If you have adopted or plan to adopt an institutional
policy of providing Open Access to your own research article output,
please describe your policy at:
        http://www.eprints.org/signup/sign.php

UNIFIED DUAL OPEN-ACCESS-PROVISION POLICY:
    BOAI-2 ("gold"): Publish your article in a suitable open-access
            journal whenever one exists.
            http://www.earlham.edu/~peters/fos/boaifaq.htm#journals
    BOAI-1 ("green"): Otherwise, publish your article in a suitable
            toll-access journal and also self-archive it.
            http://www.eprints.org/self-faq/
    http://www.soros.org/openaccess/read.shtml
Received on Sat Oct 30 2004 - 20:30:15 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:39 GMT