Ebs Hilf's clarification is very helpful. I think we agree on the
fundamental points, and disagreements are very minor.
On Wed, 7 Aug 2002, Eberhard R. Hilf wrote:
> Here follows a rather lengthy discussion partially in broken english;
> maybe you can wait for the summarizing until the dust settled.
After the dust settles comes amnesia and indifference! I think it's
better to strike while the iron is hot, in skywriting. But as we are
agreed, there is no need for a lengthy summary. Only one strategic point
needs highlighting, and I don't think it is one on which we disagree:
Ebs (rightly) takes the author/institution self-archiving of all
research for granted, as optimal and inevitable. The problem is that
although it is indeed optimal and indeed inevitable, it is not yet
actual! This is true even in Ebs's field of physics, where self-archiving
is the most advanced: Even in physics, the rate of growth of
self-archiving is stlll only linear, so that it will be another decade
before all current physics research (in that year) is being
self-archived. In other disciplines the progress toward the
optimal/inevitable outcome of open access is even slower.
So whereas Ebs is certainly right to describe and promote his vision for
future ways we might ENHANCE the current system, rather than merely
duplicating it online and with open access, the open-access is still a
necessary precondition for the enhancements he envisions! And the reason I
was insisting on clarity about what is actual and what is hypothetical,
especially insofar as peer review is concerned, is that it is (among
other things) a-priori worries about altering or abandoning the current
classical peer-review system that are holding back many researchers and
their institutions from self-archiving. Hence it is extremely important
to make it crystal clear to them that classical peer review is fully
preserved under self-archiving and open-access, and that the new
medium simply holds out the possibility of providing ADDITIONAL forms
of quality-assessment and tagging, over and above the classical one,
not instead of it.
http://www.eprints.org/self-faq/#Peer-review-reform
http://www.eprints.org/self-faq/#5.Certification
http://www.eprints.org/self-faq/#6.Evaluation
http://www.eprints.org/self-faq/#7.Peer
"Publication at LANL as involving peer review"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/0333.html
"A Note of Caution About 'Reforming the System'"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/1169.html
"Peer Review Reform Hypothesis-Testing"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/0479.html
> ERH:
> Most sophisticated and established (you asked for
> examples) is the system of CERN: they have several levels of
> certification inside CERN, tagged by 'individual scientist, group, etc.
> up to the final level of CERN-paper. (By chance it seems that their last
> level is of higher quality than the refereeing of almost any refereed
> journal (rejection rate zero for such CERN papers)
I know about, admire, and have often commended the CERN internal review
system. But note that it already exists, and is built on top of an
established self-archiving practice! Moreover, CERN is a huge
multi-institutional, multi-national entity with no obvious homologue
at the level of a single university. It is probably true to say that
many if not most of the competent potential referees of CERN papers
worldwide are already affiliated with CERN in some way. It would be
splendid if universities had the resources to emulate CERN's rigorour
internal vetting system, but I rather doubt that they have the requisite
local expertise in all their research areas.
As to other disciplines, I doubt there are many CERN-homologues there
either; with most disciplines the problem is still getting them
self-archiving at all, let alone at CERN-scale!
http://documents.cern.ch/AGE/current/fullAgenda.php?ida=a01193
http://documents.cern.ch/archive/electronic/other/agenda/a01193/a01193s3t2/transparencies/Baron.pdf
http://makeashorterlink.com/?G28C41571
> The confusion here in the discussion comes possibly from my choosing the
> word 'certification' and putting it into a sequence of label numbers.
> But this is purely technical. I argue for more fully exploiting and
> making explicit for an archive what they present to the reader and make
> use fully of the possibilities.
By all means, let full and informed use be made of all OAI metadata
tags, including AUTHOR, INSTITUTION, and ARCHIVE tags, but let us
not confuse those with certification, which is currently performed by
JOURNALNAME. The reputation of CERN for internal vetting standards is
of course known and used, and hence as much associated with CERN as the
INSTITUTION tag as with PRL as the JOURNALNAME tag, and hence potentially
just as helpful in navigation, search and retrieval.
And as long we are on the subject, author/paper citation and download
impact measures will also prove helpful in navigation, search and
retrieval, yet those too are merely digitometric aids and not
new forms of certification:
http://citebase.eprints.org/
> ERH: We should not just try to mirror the existing services of traditional
> journals but to exploit the full richness of the e-prints...
Agreed, but before we can explore their richness, we must have their
fullness (20,000 journals-full, 2,000,000 papers per annum), whereas most
archives are at the moment still near-empty! As long as your visions
of future riches are clearly cast as preserving peer-review in full,
they will help to fill archives, whereas anything that appears to
compromise peer review will only help keep them near-empty.
> > > Further reading:
> > > E.R.Hilf and H.-J.Waetjen:
> > > Scientific Refereeing in a Distributed World
> > > http://isn-oldenburg.de/~hilf/vortraege/cern01/
>
> ERH: That is a clear formulation of your personal view. And here is mine:
> You try to tell us that only Journal-name and their refereeing is giving
> certification, peer review and quality levels. The reality is different:
> Numerous research institutions and research groups have their own
> paper-lines, for historic reasons called 'Preprint' often
> (e.g., the BROWN BAG SERIES of Caltec Astrophysics Laboratory or the yellow
> line of the T.University of Darmstadt, or the green one of the Small
> Systems group of Oldenburg) where only those papers are added and
> distributed that have undergone a refereeing of a topical expert.
In physics, not only on-line self-archiving practices but also prior
on-paper preprint practices -- and reputations -- were considerably
more advanced than in other fields. The open-access eprint archives
of course inherited all of the pre-existing quality reputations and
the predictive power of their preprint-series tags.
In recommending these admirable practices to other disciplines you
are recommending two things: (1) new forms of self-vetting, such as
institutional preprint series, some of which will no doubt develop
track-records for quality standards that will have signal value, as in
physics, and (2) self-archiving. It is extremely important to distinguish
the two, especially for disciplines without a preprint tradition, lest
they wrongly conclude (as they have so often done) that self-archiving
somehow entails that preprint self-vetting replaces journal peer review.
This has to be spelled out. (Believe me!)
> For historic and technical reasons, topical journals with their blind
> refereeing were the only quality filter in the past.
>
> With the eprint age, we as scientific community have to improve the
> professional information of our colleagues by exploiting and implementing
> all means of filtering that will assist in finding the material we
> seek.
But first, we have to get our research -- both pre-refereeing preprints
and post-refereeing postprints, but most critically the latter -- online
and openly accessible, by self-archiving it. That is the primary mission;
and the potential enhancements must clearly be characterized as merely
additions, not alternatives, to those refereed postprints to which we
are all yearning for open access.
Stevan Harnad
NOTE: A complete archive of the ongoing discussion of providing free
access to the refereed journal literature online is available at the
American Scientist September Forum (98 & 99 & 00 & 01):
http://amsci-forum.amsci.org/archives/American-Scientist-Open-Access-Forum.html
or
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/index.html
Discussion can be posted to:
american-scientist-open-access-forum_at_amsci.org
See also the Budapest Open Access Initiative:
http://www.soros.org/openaccess
and the Free Online Scholarship Movement:
http://www.earlham.edu/~peters/fos/timeline.htm
Received on Wed Aug 07 2002 - 14:40:23 BST