Re: Central vs. Distributed Archives

From: Stevan Harnad <>
Date: Tue, 29 Jun 1999 18:09:31 +0100

On Tue, 29 Jun 1999, J.W.T.Smith wrote:

> I don't see what is 'papyrocentric' about... the idea of an item
> gaining some kudos by being in a certain archive...
> A similar situation occurs when a journal
> gains kudos from being indexed in a specific online biblographic database.
> No paper involved here.

Forget about indexing in databases. (If the primary journal publishers
need to think about what their new niche will be in the online world of
free, full-text self-archiving by authors, the secondaries and
tertiaries will unfortunately have more serious worries!)

The kudos comes from (P) the prestige (peer-review rigour, quality,
impact factor) of the journal that accepts the paper and (I) the impact
that it makes on research, in the form of further work citing and
building upon it. The potential impact will be made incomparably
greater by free online access for one and all.

Where is the papyrocentric thinking? In the thought that the paper's
"locus" on the Web is the source of the kudos (as the source of a paper
paper's kudos was the paper journal in which it appeared).

The accepting journal's imprimatur will shrink to a quality control
metadata tag, like a brand-name; the locus (virtual or real) of the
bytes will be of no consequence whatsoever.

Papyrocentric too is the idea that there is something to compete for in
being the "locus" of a paper. Nothing to sell, nothing to compete for.

> The subject specific archive seems an unnecessary complication. Applying
> Occam's razor it seems we can chop it off and the system can run happily
> without it.

The Los Alamos Archive has demonstrated that (at least in Physics), the
"centralized" end of the candle managed to free the literature before
the distributed end did. Occam says: Hedge your bets and do both:
Deposit in your local server AND the global one.

    Harnad, S. (1998) On-Line Journals and Financial Fire-Walls. Nature
    395: 127-128.

    "All authors should continue to entrust their work to the paper
    journals of their choice. But if, in addition, they were to
    publicly archive their pre-refereeing preprints and then their
    post-refereeing reprints on-line on their Home Servers, for free
    for all, then the de facto practises of the reader community would
    take care of the rest (irrespective of their reservations about
    bed/bath/beach reading); library serial cancellations, the collapse
    of the paper cardhouse, publisher perestroika, and a free for all,
    e-only serial corpus financed by author-end page charges would soon
    follow suit.

    "A centralised variant of this subversion scenario,, has already passed the point of no return in
    Physics and some allied disciplines in the form of Paul Ginsparg's
    (1994, 1996) U.S. NSF- (National Science Foundation) and DOE-
    (Department of Energy) supported Physics Eprint Archive at Los
    Alamos National Laboratory; as history will confirm, he
    single-handedly set the world Learned Community on its inexorable
    course toward the optimal and the inevitable in August 1991."

> js> Why don't you drop the word 'journal' then? Why not use 'validator' or
> js> some other word that indicates the role and doesn't carry over
> js> connotations from the old "papyrocentric" model?
> >
>sh> Suit yourself. But I think "Physical Review Letters" will continue to
>sh> prefer to call itself by its current familiar and trusted brand name --
>sh > and why on earth shouldn't it?
> I'm not saying we shouldn't have "Physical Review Letters" (or any other
> title) just that in the new model we should stop calling it a 'journal'.

Suit yourself. Maybe we should stop calling the contents "articles" too.
But what's the point?

> The problem with the word 'journal' is that it carries connotations from
> the "papyrocentric" world. For example - the idea that an item can only be
> in one 'journal'.

And a good connotation too! We have already gone round this one before:

Referees are a scarce and overworked resource. There is no justification
for asking anyone to referee an already-refereed, already-accepted
paper yet again, for acceptance yet again, elsewhere.

See the two reasonable sources of kudos above: (P) is acceptance by a
peer-reviewed Journal; (I) (and more important) is "acceptance" by
one's peers through the paper's impact on their reading, research and
citations. No more need for infinite rounds of peer reviewing and
re-reviewing. Otherwise it's like going back to school for more and more
exams instead of getting on with it! We haven't the time or the manpower
for such an orgy of endless assessment (even in the UK!).

> This does not need to be the case in a net-based model.
> Your descriptions of your model seem to contain a "papyrocentric"
> influence since there still seems to be a close relationship between an
> item and the 'journal' that validates it.

There is nothing papyrocentric about quality control and certification.
Even eggs go through that process...

> There is no reason why an item
> could not be validated by more than one validator - especially if it
> crosses current subject boundaries.

There is indeed. Vide supra...

Stevan Harnad
Professor of Cognitive Science
Department of Electronics and phone: +44 2380 592-582
Computer Science fax: +44 2380 592-865
University of Southampton
Highfield, Southampton
Received on Wed Feb 10 1999 - 19:17:43 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:45:33 GMT