On Wed, 9 Mar 2005, adam hodgkin wrote:
> Cockerill [makes] some good points and Harnad may be missing something.
> It sometimes seems that there can be too much reliance on the fact that
> authors of scientific research articles are not paid by the publishers
> for publishing their papers, and is Harnad relying too much on this
> criterion to define the field of OA concern?
I don't think so: Matt and I have exactly the same target literature
in mind: the 2.5 million annual articles published in the planet's
24,000 peer-reviewed scholarly and scientific journals. To be sure,
those articles are not the only author give-aways on the planet (*some*
authors give away their books, magazine articles, unpublished works,
poems, novels, plays, music, video, software). But I would say that the
peer-reviewed research literature is the only one for *all* of which this
is true without exception -- in fact, the only literature that even comes
close (to being 100% author give-aways). Moreover, it is the literature
for which the urgency of OA is greatest, as it is all written for research
impact, and anything that blocks research access, blocks research impact.
The difference between give-away and non-give-away work is the single
most important Post-Gutenberg distinction. Without it, one hasn't a hope
of understanding OA:
http://www.ecs.soton.ac.uk/~harnad/Tp/resolution.htm#1.1
> I suspect that Cockerill
> and Harnad agree that the major science publications which have
> journalist-written reports and news are not expected to be OA (the
> front half of the magazines).
Correct.
> But the research reports that Science,
> Nature and the Lancet carry should on the Cockerill and Harnad view be
> OA -- but not I suggest for the simple reason that the authors are not
> being paid.
Not just for that simple reason: Also for the reason that they are
all written for research impact:
http://www.ecs.soton.ac.uk/~harnad/Tp/resolution.htm#1.2
> If the unpaid nature of the authorship were the key issue, would
> Harnad consider it an acceptable way of reforming the system for such
> authors to be paid a fee on the appearance of their article, which was
> thereafter Toll Access? Presumably not.
Certainly not. The problem OA is intended to remedy is not loss of author
revenue, since these authors don't *seek* revenue. The problem OA is
intended to remedy is loss of author impact, for which the only remedy
is to make sure that every would-be user can access the research.
> It seems to me that Cockerill's point is that publishing the results
> of research is itself a part of the research process, and the
> publication itself should be viewed as such, openly, with the maximum
> opportunity for review, analysis, reuse and improvement;
But that is all abstractness and ideology. The practical problem is
far more straightforward: 2.5 million articles are published yearly
in 24,000 peer-reviewed journals. Currently, those articles are only
accessible to those users whose institutions can afford to subscribe to
the journals in which they are published. If there were no other
would-be users than these, there would be no OA problem! But the fact
is that most institutions cannot afford most journals.
http://www.arl.org/stats/index.html
This means that most articles are inaccessible to N > 0 of their
potential users. (No one knows how many, but the dramatically and
consistently higher impact of self-archived articles suggests that the
N is substantial.)
http://opcit.eprints.org/oacitation-biblio.html
The remedy? Supplement the publisher's official version (accessible to all
users whose insitutions can afford it) with the author's self-archived
version (for all users whose institutions cannot afford it).
Nothing to do with "publishing the results of research is itself a part
of the research process" (which is merely an empty ideological cliche).
Not does it have anything to do with "the publication itself [being]
viewed, openly, with the maximum opportunity for review, analysis,
reuse and improvement" -- another porous platitude:
Peer-review reform is not at issue here: *Access* to the peer-reviewed
literature -- *such as it is* -- is.
http://www.nature.com/nature/webmatters/invisible/invisible.html
> and Harnad
> cannot rely on the point that most of a typical article is text, in
> many cases the most important parts of a publication are not text, but
> images, data and those other elements of a research report, eg
> references, that matter to those in the field.
So what is the point? The images and references are all in the
self-archived supplement too. So are the data (if they are part of the
published article: and if they are not, they should be self-archived
anyway! see below).
> There is a big lesson
> to be learned from the experience of the human (and other genomes)
> where the toll access mode of publication was resoundingly and
> decisively defeated four or five years ago.
To repeat what I said to Matt: That was data self-archiving. Highly
commendable, but another matter, and not the one at issue
here. (Peer-reviewed journal article content is.)
http://www.psigate.ac.uk/ebank/
The analogy with the human genome project (or with open-source software)
is incorrect and misleading, and it does not get any better with
repetition!
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/2967.html
> It didnt even get off the
> ground because it was so clearly going to impede scientific progress
> if all the databases (and their annotations) were closed or subject to
> proprietary control.
Open access to scientific data is a splendid, highly desirable thing --
but not what is at issue here! (Peer-reviewed journal article content is.)
http://www.ecs.soton.ac.uk/~harnad/Temp/data-archiving.htm
> Science is most efficiently pursued when its an open and collaborative
> process and that is where the OA argument strikes hardest.
When it is aimed at a concrete, practical target, not an abstract (and
empty) ideological one.
Stevan Harnad
Received on Wed Mar 09 2005 - 18:42:39 GMT