Re: US Federal Research Public Access Act of 2006

From: Stevan Harnad <>
Date: Fri, 26 May 2006 14:04:51 +0100

    [The following is an anonymized exchange, clarifying two
    misunderstandings about the FRPAA: (1) Is the FRPAA proposing to
    mandate duplicate publishing? and (2) Why do peer-reviewed publishers
    need to be independent third parties]

(1) Self-Archiving of Published Articles, Not Double Publishing

It is important to understand exactly what the FRPAA is proposing to mandate:

    "4(b) Each Federal research public access policy shall provide for...
    (1) Submission to the Federal agency of an electronic version of the
    author's final draft manuscript of original research papers that
    have been accepted for publication in a peer-reviewed journal and
    result from research supported, in whole or in part, from funding
    by the federal government... (4) free public online access to such
    final peer-reviewed manuscripts... as soon as practicable, but not
    later than six months after publication in peer-reviewed journals...
    (6)(A) in a stable digital repository maintained by the Federal
    agency; or... any repository meeting conditions determined favorable
    by the Federal agency, including free public access, interoperability,
    and long-term preservation.

It is not mandating double publication. It is mandating the Open Access
Self-Archiving of *published* journal articles (the author's final,
corrected, refereed draft) so that those potential users worldwide who
cannot afford access to the journal *in which it has already been
published* can access and use it.

ANON (below) seems to think FRPAA is mandating double publication, or even an
alternative to publication. Not at all: A supplement ot publication.

Please note also that 93% of journals already officially endorse immediate
author self-archiving of the articles they publish; so the (up
to) 6-month delay FRPAA need only apply to 7% of articles -- and there too,
the solution is to deposit the article and its metadata immediately upon
acceptance for publication, and send the eprint by email individually to any
eprint requesters: The requesters see the metadata and the OA repository
offers a semi-automatic email-eprint button, to which the author can respond
with one click in each case.

> The majority of the outstanding and highly acclaimed journals
> belong to countries in EU or US. Publishing articles in such journals
> is often linked to the researcher's career, etc. Now, most
> journals are not in a position to accept a paper which is already
> published in some form.

FRPAA is not requiring the author to submit a published paper to a
journal: it is requiring them to make the final, corrected draft of each
paper they publish Open Access within 6 months of publication.

> If a researcher has to publish an article that is almost similar
> to the journal article (see, the text of Bill ), why on earth would a
> refereed journal publish the almost same content in their publication?

Where did you get the idea that the Bill requires authors to try to
pre-publish (where?) an article, and then try to submit it to a journal?

> So what are the implications of OA for a researcher or S&T as a whole?

A researcher should publish his paper in the most suitable journal that
will accept it, exactly as always. Then, once it is accepted, he should
immediately deposit the final draft in his Institutional (or Central)
Repository. (Access should be set as Open Access immediately wherever
possible, and as Closed Access for up to six months where judged
necessary.) Self-archiving is not duplicate publication; nor is it
prior publication. It is supplementary access-provision for *published*

> Is there any solution? Should researchers be mandated to publish in a
> form other than article?

There is no problem, hence no need for a solution. Researchers should
publish as they always did, and in addition, self-archiving their
publication (author's final accepted, corrected draft).

> See Section 4 of the FRPAA

Nothing in section 4 implies what you seem to be worrying about. It is all
speaking about an article already accepted for publication by a journal of
the author's choice, as it always was.


(2) The Publisher (Peer-Review Service Provider/Certifier) Has to Be an
Independent Third Part (Not Author/Institution or Government/Funder)

> Thank you for the clarification - it resolves half of my doubts.
> The moot point is that OA intends to provide access to national scholarly
> research through internet

OA intends to ensure that there is free internet access to all
research, from all nations, published in any of the world's 24,000
peer-reviewed journals.

> OA (not FRPAA per se) implicates two parties: the researcher and the
> research institutions both funded by taxpayers money. Then why does it
> have to be dependent on a third party - a commercial journal publisher,
> a professional society and so on... ?

Because peer review requires adjudication and certification by a
qualified reputable, answerable independent party, not the researcher's
own institution or funder.

The reason publication involves a third party -- besides the government
funder and the author/institution -- is that research results are not
merely published as-is. They have to be peer-reviewed to make sure they
are new, correct, and complete, and meet the established quality standards
of a given journal. Journals vary in quality, and this is in general known
to users, so they calibrate their use and their confidence on the basis
of the established quality standards of the hierarchy of journals -- a
few very high quality ones on top, then less rigorous journals, etc. all
the way down to what is virtually a vanity press at the very bottom.

The peer review is part of the self-corrective nature of science and
scholarship. The journal editor chooses qualified experts -- from the
world over in the case of a rigorous international journal -- and those
referees are answerable to the editor, who is also supposed to be a
reputable, qualified expert; the author too is answerable to the
editor, in consultation with the referees, to ensure that the
journal's established quality standards have been met.

Human nature being what it is, this quality-control system is far from
infallible: mistakes sometimes are not caught by peer review, and valid
work is sometimes rejected from a higher-level journal and appears in a
lower-level one than the one it really merited. But Open Access is not
about reforming peer review; it is about maximizing access to its results:
the peer-reviewed, accepted, published articles.

There are 24,000 peer-reviewed journals in the journal quality
hierarchy, across all fields. Each of them has its own independent
editor(s) and editorial board. Referees and authors are answerable to
that editor and editorial board, and the editor and board are in turn
answerable to the research community as a whole for their quality

And there is your answer about why publication requires a "third party",
other than either the author/institution or the funder: Because the
peer-review process must be independent. "In-house publishing" in which
the author's own institution vets what it publishes is rightly
considered vanity press, and its quality standards are not trusted,
because they are answerable only to themselves. If national journals are
lower in the quality hierarchy than international ones, then
self-published institutional or governmental journals are lower still,
if they are considered "peer-reviewed" at all. To publish the highest
quality, a journal must be refereed by the highest-quality specialists,
and these are not, in general, either institutional appointees or
governmental ones, vetting their own output. They are neutral third
parties, controlling quality for the research community as a whole.

So OA is not about changing any of that -- but merely about maximising
access to (and hence the usage and impact of) its quality-controlled
outputs: the refereed journal articles.

> It is desirable that if government mandates a government agency to
> make certain provisions in their research system, then a third party
> which is governed by a different rule of law should not be involved.

The government can attach conditions to the money it disburses,
conditions on those who receive it. Here this consists of researchers
and their institutions. Publishers do not receive government money,
hence no conditions can be imposed on them. Moreover, publishers
(peer-review service providers and certifiers) need to be independent
third parties, otherwise it is not really peer review. The qualified
experts for a given piece of research might be anywhere on the planet.
Moreover, they are busy, and they referee for free. So they are far more
likely to hew to the call of the reputable editor of a reputable,
independent journal with a known track-record for quality, than to
someone from the author's own institution or funder. And a neutral,
independent editor with a journal's established quality-standards to
maintain, rather than someone at the author's institution or funder with
an interest in getting institution/funder research published, is far
more likely to maintain rigorous, objective, independent quality standards,
both in choosing referees and in seeing to it that their expert
recommendations are followed (including rejection rates of as high as
95% for the most selective journals).

So the government has every right to see to it that publicly funded
research is accessible to the widest possible number of potential users
and appliers of that research (to the benefit of the public that funded
it), and this includes making sure that it receives the most rigorous
quality control possible, but it does not include taking over the
quality control itself, which would simply compromise the peer review
system. That must be done by an independent third party.

> Does it mean OA/ OA legislation should seek a different publication
> system for OA?

Definitely not. OA legislation is not about altering the publication
system but about altering the means of gaining access to its output:
published, peer-reviewed journal articles.

Stevan Harnad

A complete Hypermail archive of the ongoing discussion of providing
open access to the peer-reviewed research literature online (1998-2005)
is available at:
        To join or leave the Forum or change your subscription address:
        Post discussion to:

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:

    BOAI-1 ("green"): Publish your article in a suitable toll-access journal
    BOAI-2 ("gold"): Publish your article in a open-access journal if/when
            a suitable one exists.
    in BOTH cases self-archive a supplementary version of your article
            in your institutional repository.
Received on Fri May 26 2006 - 14:23:46 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:48:21 GMT