Re: EPrints, DSpace or ESpace?

From: Leslie Carr <>
Date: Mon, 17 Feb 2003 01:02:32 +0000

[This message contains some long quotes. Please bear with me!]

At 10:20 16/02/2003 -0500, Dempsey,Lorcan commented on Les Carr's comment:
> > The "Open" in OAIS comes from the fact that the standard is open (the
> > archives may be closed), whereas OAI and BOAI assume open distribution
> > of metadata and open access to texts (respectively).


>This is misleading. The "open" in OAI is explained in the OAI FAQ on the OAI
>website as follows:
>What do you mean by "Open"?
>Our intention is "open" from the architectural perspective - defining and
>promoting machine interfaces that facilitate the availability of content
>from a variety of providers. Openness does not mean "free" or "unlimited"
>access to the information repositories that conform to the OAI-PMH. Such
>terms are often used too casually and ignore the fact that monetary cost is
>not the only type of restriction on use of information - any advocate of
>"free" information recognize that it is eminently reasonable to restrict
>denial of service attacks or defamatory misuse of information.
>This is available from
>The protocol is agnostic about the business or service environment in which
>it is used. The RDN for example uses OAI to gather metadata
>from its contributing partners in a closed way.

Absolutely. I apologise for the ambiguity that remained in my comments. I
deliberately used "access" for BOAI and the weaker "dissemination" for OAI,
but failed to communicate my meaning. In OAI, the access is "open" not in
the "economic" sense, but in the "architectural" sense - OAI-PMH is
designed for disseminating metadata between system- and organisational-
boundaries. In that sense, it increases "openness", because unrelated
systems MAY (if permitted), participate in metadata dissemination and
processing activities (hence the OAI terms "data provider" and "service
provider"). Nothing about OAI addresses whether the DATA itself (e.g.
documents) may be shared and processed.

I think I was correct about the O in OAIS (see section 1.1 of the
standard document), but more to the point, the architecture diagrams
which are contained in the standard seem to lack any mechanism for sharing
between Archives or Archives and other distributed computational entities
(harvesters, agents, semantic web spiders etc). In fact, I now see (in
the tutorial you refer to below) that federated archives are supported by
the standard (I guess they are a simply a different class of consumer),
so that is a step in the right direction.

I'd like to emphasise that the mission of the OAI is to "facilitate the
efficient dissemination of content" (quote from FAQ). Interestingly
enough, the enormously sucessful High Energy Physics archive became
the problem that OAI was established to solve - it was an ISOLATED
archive. Building an archive is a great step forward if you don't
have a stable environment from which to share your data or documents,
but a planetful of individual, isolated, unco-operative archives
(part of the so-called deep web) is a solution to one kind of problem
(scholarly storage), but the foundation of another kind of problem
(scholarly communication).

There are at least two of these kinds of archives available right now,
and we think it's hugely important to maintain the momentum in encouraging
a planetful of scientists and scholars to use one or other of them!!

> > It is worth noting that the scenarios given in OAIS are without
> > exception
> > data archives - enormous collections of database records comprising
> > government forms or scientific measurements. In contrast, scholarly
> > papers are documents, not data; their purpose is communication rather
> > than processing. It is perhaps unsurprising that the users of these
> > documents require something different from their archives, accounting
> > for Stevan's emphasis on immediacy and access.
>This is again misleading. If you look at the following tutorial on the OAIS
>website by Don Sawyer and Lou Reich (dated October 2002)
>you will see several examples of document-related scenarios.

My comments applied to the set of illustrative scenarios that were
published in the standards document itself. Thanks for drawing my
attention to this helpful tutorial which expands on the role of the

> > Perhaps there is an unavoidable tension here - for a librarian, an
> > article
> > about Cognitive Science can only be an object to be curated, whereas
> > for a
> > Cognitive Scientist it is a message to be interpreted and used.
>Well ... I would argue that this is also again misleading. Curation and use
>are intimately connected: libraries engage in curatorial practices to
>support use. A librarian wants to make sure that what was written yesterday
>is available for you to use today. A librarian wants to make sure that an
>article you write today is available for somebody else to read tomorrow. I
>doubt whether you really only want to read today's articles, or to have your
>own work unavailable to somebody else tomorrow.

I think my comment is contentious, but I don't think it is misleading -
I think that it is a statement about the disparate requirements of two
sets of stakeholders. You (and I both) believe that curation by one
party serves the use by the other. I was reflecting on the position
which we temporarily find ourselves in - arguments about theoretical
best practises for curation unintentionally inhibiting the practical
improvements for use which we are trying to achieve.

Les Carr
Southampton University
Received on Mon Feb 17 2003 - 01:02:32 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:46:52 GMT