Re: Central vs. Distributed Archives

From: Greg Kuperberg <greg_at_MATH.UCDAVIS.EDU>
Date: Fri, 3 Nov 2000 08:48:07 -0800

On Fri, Nov 03, 2000 at 08:24:44AM +0000, Stevan Harnad wrote:
> But why restrict efforts to centralized ones only? The whole point of
> OAI interoperability is that it should no longer make any difference
> whether a refereed paper is archived in a central archive or a
> distributed archive or both! (The only alternative we want to avoid is
> "neither"!)

It is not really a neutral statement to declare that it no longer
matters whether a paper is in a central archive or a distributed one.
Each archive is in a way an entrenched interest. Each archive maintainer
has put a lot of work into his or her project, and therefore wouldn't
want it assimilated into a larger archive without a very good reason.
So saying that it no longer matters whether it is centralized or
distributed is like saying that it no longer matters whether states
answer to Washington.

This is overconfidence. The biggest reason that it is overconfidence
is that it defers the permanence question. But there are other reasons
as well. One is that one of the most useful features of the arXiv
(and similar services such as CogPrints) is immediate notification of
new results. Another is non-redundancy: the arXiv almost completely
eliminates the disarray of having many copies of a paper which may
or may not be different versions. The OAI standard does not address,
and perhaps cannot address, either of these important advantages of a
centralized system.

A more balanced point of view would be to recognize that while a
standards-based distributed system may be much better than anarchy,
it doesn't finish the job.

I also note that interoperability keeps getting reinvented. Precedent
suggests that if OAI succeeds, it will fade into a transparent layer,
and that beyond it people will see incompatability at a new level and
invent another standard. HTTP is already an interoperability standard,
originally invented for the purpose of distributing research documents.
And there are already HTTP-based search engines, including CiteSeer,
which searches only for research papers. So it's important explain how
OAI would go beyond HTTP+CiteSeer.
  /\  Greg Kuperberg (UC Davis)
 /  \
 \  / Visit the Math ArXiv Front at
  \/  * All the math that's fit to e-print *
Received on Mon Jan 24 2000 - 19:17:43 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:45:56 GMT