Developing an agenda for institutional e-print archives

From: Chris Rusbridge <c.rusbridge_at_COMPSERV.GLA.AC.UK>
Date: Fri, 13 Jul 2001 15:56:52 +0100

Catherine, I was quite enthused by the e-Prints meeting held yesterday in
London. At the end of the day we were challenged to find ways this
technology and ethos can be taken forward.

The first challenge I felt was to those present. How could any library
administrator leave that meeting without a personal commitment to ensure an
institutional e-print server was up and running within 6 months (if not by
next turn)? How could anyone of us leave the meeting without a personal
commitment to ensure that all our papers would be available in a free
e-print server within the same sort of period? And if there is no e-print
server in our institution or discipline, that commitment should include
badgering our library administrators to create one. This is consistent with
the main theme at the end: fill those empty archives. But we need a little
more southern Baptist fervour at this point: brothers and sisters, have you
seen the light (;-)?

We were also challenged to find ways that JISC could support this movement.
We do have something of a 'chicken and egg' situation here. Or have I got my
metaphor wrong; is it the first telephone? (Who do you call? Who calls you?)
The motivation to deposit papers comes partly from the increased exposure of
those papers. In the OAI model which underlies e-prints, there is a logical
separation between the data providers (those empty archives...) and the
service providers which make that content accessible to those who know
nothing of the individual archives. It appears there are very few service
providers; the ARC service in the US is not likely to be the first call for
many interested in our material, and its (sole) existence is not going to be
a very powerful motivator to a depositor.

So I would like to urge JISC to support the creation of a variety of service
providers. These could be regional, central, subject, domain or genre-based.
The first does not seem to me to make a great deal of sense in a wired world
(although there could be management motivations from existing regional
structures, eg MANs). Subject services however could map neatly onto the RDN
or LTSN, and the central server onto the Resource Finder. I understand the
DNER expects to use OAI (rather than e-prints) in domain areas such as
images, etc.

As well as filling those empty archives, we should think about exploiting
the material we already have. There are a number of EJ and pre-print-type
resources some of which owe their existence to JISC. CogPrints and WoPEc
presumably are already OAI-compatible; other pre-print-like projects include
EducatiOnline, Formations and DeLiberations. There are also e-magazines,
like Ariadne and its spin-offs, and more formal EJs like Internet
Archaeology, JILT and Sociological Research Online (the latter had an
associated E-press project which could perhaps be persuaded to add
OAI-compatibility to its deliverables). Others include JIME, Journal of
Buddhist Ethics etc, created without JISC support.

It cannot be a large job in each of these cases to convert the resource to
be OAI-compatible and register with an appropriate service provider; this
would presumably increase the exposure of their contents. Could JISC offer
say 5-10K per provider, subject to a 1-page bidding process, to carry out
this conversion?

A related possibility might be to convince SPARC, perhaps via its European
chapter, and/or the Scottish initiative, to make all their journals
OAI-compatible. There's nothing in OAI itself that says the content has to
be delivered free, although in the spirit of the subversive proposal we hope
it will be. And there have to be some possibilities in the EASY project
(where a few publishers have agreed to offer articles at a price roughly
commensurate with ILL costs).

JISC, HEFCs and the Research Councils fund a great deal of publishing and a
larger number of reports. JISC should require all publications derived from
its funding to be deposited in a free e-prints service (eg the Projects
Output Database, under development for so long). If the POD has been
abandoned, then perhaps JISC should simply set up a pre-prints service of
its own; failing that, require deposit somewhere as suggested above. JISC
should also attempt to persuade the other funders to do the same. A job
perhaps for the new JISC Research Committee?

Finally, could JISC fund the existing developers to add some improvements?
An easy way for depositors to import bibliographic details from reference
databases would be nice, as would a way for administrators to easily derive
PDF from Word, LaTeX or PS source material. I'm sure there are lots of other
ideas. I would very much like to hear them!

Could JISC also get someone to do some thinking on the long term
preservation issues implicit in these archives. It's one things to set such
considerations aside for now, but they cannot be ignored for long. The PDF
converter mentioned above is a stop-gap proposal. Another short term option
might be a slightly more long-lived format easily derivable from PowerPoint
templates, which would make this a more attractive format. A more long term
solution could be for example Word templates which automatically produce XML
in an agreed format for journal articles (see for example the Glasgow CDocS
system produced from the JISC-funded Electronic Records Management project).

Chris Rusbridge
Director of Information Services, University of Glasgow
phone 0141 330 2516   fax 0141 330 5620 
Received on Wed Jan 03 2001 - 19:17:43 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:46:10 GMT