> The model that has worked for libraries for standard materials has been
> successfully implemented in OCLC. I would be interested in any
> thoughts on how this model could be improved upon for Web resources.
There is an underlying assumption behind this discussion that I would call
the 'cataloging the Web' frame of reference. Like the major search services
Yahoo!, AltaVista, Excite, etc, this perspective views the entire Web as its
domain, and the criteria for implementation and success are structured
around this assumed information universe. I think that Martin's reference to
OCLC's research in this area is valid, and OCLC can point to some
interesting and innovative developments here. It will be especially
interesting and exciting to see if the 'OCLC approach' to resource control
will provide a viable alternative to the commercial search services, both
for libraries and for end-users who are now making use of the commercial
services.
I think this Cataloging the Web perspective is fine, and is a valid
exercise in itself, but there is another frame of reference that is at least
as important. This is the 'specific domain' metadata environment, which may
refer to an institution like a university, or it may be a region, like the
Tasmania Online site given in an earlier posting. Mixing these two realms of
conceptual discussion- the Cataloging the Web, on the one hand, and the
specific metadata domain, on the other- is the source of much confusion in
metadata discussion, in my opinion, and I think the source of possible
misunderstanding here. There will be some relationship between the two
domains, and it is certainly worthwhile to speculate how this will develop.
This general vs. specific approach is one that OCLC knows well, since
there is an ongoing, healthy tension between the view that OCLC can satisfy
ALL the bibliographic control needs for all libraries, and the view that
every library has specific bibliographic control needs for its particular
user community. I think this healthy tension will play itself out in the
developing metadata culture. There is little question that institutions like
universities and libraries have their own specific metadata needs, and I
believe that many of us are anticipating that libraries will develop their
own metadata administration and infrastructure, with their own servers,
spiders/harvesters, search engines, portals, and user interfaces. Metadata
will be configured and developed for the particular search engine
environment set up for the library, as with other
institutions/organizations. Most of us are expecting an entire industry
developing around this institutional approach to metadata, and a number
companies are developing suites to work with metadata. Rather than looking
at what AltaVista or Excite are doing with metadata (that will play itself
out in the commercial marketplace), we should be looking at what companies
like Blue Angel, Microsoft, Open Text, etc etc, are doing in this area, for
these are the working tools that we will be using.
********************************************
Roger Brisson
Digital Access Librarian,
Selector for German Language & Literature
Pennsylvania State University
Coordinator, AAU/ARL German Resources Project
814-865-1858 (voice)
814-863-7293 (fax)
[log in to unmask] (email)
http://www.personal.psu.edu/faculty/rob1
********************************************
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|