[OAI-eprints] Re: Public Access to Science Act (Sabo Bill, H.R. 2613)

[Stevan Harnad]

    [I liked Peter Suber's paper on "The taxpayer argument for open
    access." https://mx2.arl.org/Lists/SPARC-OANews/Message/97.html 
    The following was written earlier, after I saw his draft, but Peter
    asked me to wait till his paper appeared before posting mine. Our
    positions agree, and differ only slightly in emphasis.]

            Public Access vs. Public Domain

                Stevan Harnad

(1) HEALTH RESEARCH VS. ALL OTHER KINDS OF RESEARCH

The tax-paying patient-rights argument for toll-free access to
research is a good one for health-related research, but it does not
generalize to other research -- and hence risks inducing the (incorrect)
conclusion that health-related funded-research is the only special case
that needs toll-free access! It is accordingly important to immunize
against all such narrow interpretations of the case for toll-free access
from the outset, by coupling the health/tax-payer rationale for toll-free
access with the other, more general toll-free-access rationales,
pre-emptively.

The most general rationale is that toll-free access to research is best
for research itself. Blocking research access blocks research uptake and
usage, thereby blocking the pace, progress, productivity, and impact of
research, in every field. It has to be made very clear that open access
is as much about hadrons (and Hadrian!) as it is about health! Health
just serves to drive the point home, because the connection between
research progress and our health is so transparent, vital and close to
where we all live. But if the rest of scientific and scholarly research
too is worth doing (and funding), and not just health-related research,
then it *all* needs toll-free access just as much as health research does.

(2) TAX-PAYER ACCESS VS. RESEARCHER ACCESS

The tax-payer argument for toll-free access to research in general
(rather than just to health research in particular) should not be cast
solely, or even primarily, in the form of toll-free research-access
for tax-payers! It should be cast primarily in the form of toll-free
research-access for researchers:

Researchers are funded by tax-payers to conduct research because of the
potential benefits of research progress to tax-payers. Those benefits do
not consist of the tax-payer's freedom to read the research results! Most
tax-payers will have no interest in reading research results. Their
interest is in the potential *application* of research results --
technological, medical, commercial and cultural -- to the improvement of
their lives. 

The primary readers and users of research are researchers, and the
benefits to tax-payers of giving researchers toll-free access to one
another's research are exactly the same as the benefits to tax-payers
of funding the conduct of the research itself: Research is interactive,
collaborative, collective, cumulative. Maximizing researchers' access
to research results maximizes research progress and productivity and
thereby maximizes the probability of eventual applications, and hence
of eventual direct benefits to the tax-payer.

Access-denial (the current subscription/license toll-based status quo)
has the exact opposite effect. It blocks researchers' access to one
another's research. Researchers can only access and use the research for
which their own institution can afford the access tolls. No institution
can afford access to all or even most published research. Most
institutions can afford only a small and shrinking fraction of it.
Hence all research is effectively inaccessible to most researchers. All
that potential research usage and progress is currently being lost,
daily. That is why it is in tax-payers' (and research funders') interests
to make all research accessible toll-free to all potential users
(worldwide).

(3) PUBLISHING IN OPEN-ACCESS JOURNALS
    (THE 5% SOLUTION)
    VS. 
    SELF-ARCHIVING TOLL-ACCESS PUBLICATIONS IN OPEN-ACCESS ARCHIVES
    (THE 95% SOLUTION)

But the appeal and the emphasis should not be unduly focussed on the
tax-payer either. Researchers themselves are the ones that most need to
be targeted. It is critically important to stress that whereas, with help
from the tax-payer and from funding agencies, it is possible to
help cover the costs of publishing in open-access journals, there
are still very few open-access journals in existence today (500
journals, according to http://www.doaj.org/ ), relative to the amount
of research that is published annually (24,000 journals according to
http://www.ulrichsweb.com/ulrichsweb/analysis/ ). This demographic fact
and its consequences must be stated very explicitly and they must 
also be understood clearly:

If the 5%/95% figure is not stated and understood, it is certain that
most people will draw the erroneous and counterproductive conclusion that
publishing research in open-access journals is the fastest, most direct
(perhaps even the only) way for researchers to provide toll-free access
to their research today.

But this is far from being the case. Open-access journals exist for
only 5% of annual research output today. Hence publishing in (and
covering the costs of) open-access journals is only the 5% solution
for refereed research publication and access today. The other 95% of
research can only be published in toll-access journals today. Hence
researchers self-archiving their own toll-access journal publications
in their own institutions' open-access archives is the solution for the
remaining 95%. This too must be strongly encouraged by the tax-payer and
the research-funder, not just the 5% solution. The complementarity as
well as the relative scope of the two means of attaining open access
must be clearly understood if we are to reach the optimal and inevitable
outcome of toll-free access to 100% of the refereed research literature
now rather than decades hence.

(4) "PUBLISH (WITH MAXIMIZED ACCESS/IMPACT) OR PERISH"

Last, there is a limit to how much the government, the research-funder,
and the tax-payer can do through funding and legislation. The traditional
institutional carrot/stick mandate that researchers must "publish or
perish" must now be extended to mandating that they "publish with
maximized access/impact" (i.e., open access) so as to help bring
it home to researchers that it is in their own interest to make their
research accessible toll-free for all potential users. The optimal dual
strategy is hence:

    (i) Wherever a suitable open-access journal for you to publish your
    research in exists today (c. 5%), publish it there, today.

    (ii) For the rest of your research (c. 95%), for which a suitable
    open-access journal does not exist today, publish it in your preferred
    toll-access journal, but also self-archive it in your institutional
    toll-free access archive, today.

http://www.ariadne.ac.uk/issue35/harnad/

(5) PUBLIC ACCESS VS. PUBLIC DOMAIN

The Public Access to Science Act (Sabo Bill, HR 2613) (PASA) rightly
invokes the taxpayer argument in support of open access to (funded)
research. However, the *means* by which PASA proposes to provide open
access is far more radical and confrontational than necessary -- and the
reason is that the Bill's proponents clearly had only the 5% solution in
mind in proposing it, as if that were the only possible solution.

All that is needed to serve all the needs of researchers, research,
and tax-payers is open access -- that is, immediate, permanent, free,
online access to the full-text of every research journal article. To
attain this, it is not necessary that all authors and publishers renounce
copyright protection for all those texts. It is only necessary that
*someone* provide open access to them! Open-access journals do that,
but they represent less than 5% of the 24,000 journals there are. What
about the rest?

All that the PASA need mandate is that all journal articles based on
funded research must be made openly accessible. To comply with this,
neither the author nor the publisher needs to renounce copyright
protection for those articles. The journal must simple allow the
author to self-archive them, in his own institutional open-access
archive -- as 55% of the journals sampled already do officially (and
many others will agree to do if asked on a per-article basis):
http://www.lboro.ac.uk/departments/ls/disresearch/romeo/Romeo%20Publisher%20Policies.htm
And even for the few journals that might refuse, the author has a legal
alternative that is almost as simple, and effectively provides the open
access anyway: self-archive the unrefereed preprint plus the corrections.
http://www.eprints.org/self-faq/#publisher-forbids

PASA could help bring the 55% figure to the 95% needed (together with the
open-access journals' 5%) to make the entire research journal literature
100% open-access -- by simply mandating open access itself.

By needlessly mandating more -- the renunciation of copyright (forcing
authors not only to seek a journal that will accept those terms, instead
of the journal they might have preferred, but also putting their texts at
risk of being plagiarized or corrupted) -- PASA would needlessly elicit
opposition from publishers and authors alike.

The tax-payer is best served if the access to (and hence the usage
and impact of) tax-funded research is maximized by requiring that its
full-texts must be made openly accessible online to all would-be users,
everywhere. There is no need whatsoever for those texts to be put into
the public domain in order to obtain that universal benefit.

Stevan Harnad

NOTE: A complete archive of the ongoing discussion of providing open
access to the peer-reviewed research literature online is available at
the American Scientist September Forum (98 & 99 & 00 & 01 & 02 & 03):

    http://amsci-forum.amsci.org/archives/september98-forum.html
                            or
    http://www.cogsci.soton.ac.uk/~harnad/Hypermail/Amsci/index.html

Discussion can be posted to: september98-forum at amsci-forum.amsci.org