World Wide Web Digital Library System

Kevin C. Marsh iai at neosoft.com
Mon Mar 4 17:23:31 EST 1996


Dr. Heinrich C. Kuhn wrote:
>...I don't quite know, what you think this "World Central WWW 
>Digital Library Site" of the future should be like. The Net
>has hitherto worked well without too much centralism. A way
>of doing "combined" searches in several "catalogues" of 
>ressources might be sufficient.

This is easily done if those catalogues comply with some common interface
standard such as Z39.50.

>...For some >types of information it seems reasonable to "catalogue" them
>regionally (e.g. for indices on hotels and restaurants, for
>some types of yellow pages, etc.). For other types of infor-
>mation however distributions according to subject areas might
>be more appropriate - like today for texts published on paper:
>NLM does at good job at indexing, "cataloguing" medical texts
>and the CAS at indexing, "cataloguing" chemistry texts - from
>(at least potentially) all over the world. The same holds true
>for what special libraries do for monographs.

Exactly!  It seems to me that what we need at this point is not a new
infrastructure of global, regional, and local directories, but rather a
proliferation of standards-compliant information collections that are
capable of being interconnected in a variety of ways based on the needs of
user groups or individual users.  Intead of creating a directory of
Chemistry sites on the Web, for example, you could simply configure a Web
page to launch a query against the contents of those sites.  An individual
researcher could copy and modify that Web page to search only the sites they
prefer, or to add more sites that match their research interests.

>   Today's situation with regional and local and world wide
>specialised indexing and cataloguing of lots of "conventional"
>and more and more "non-conventional" material is not a simple
>situation, and it probably will become an even more compex 
>situation in the future. Global, regional, local endeavours:
>all of them probably will be needed to cope.

I agree, but information professionals can reduce that complexity by
creating, studying, advocating, and implementing appropriate standards.
Distributed, standards-compliant collections would be more responsive to
"local" requirements and be more sustainable than a central registry system
IMHO.  They are also easier to implement, and can be created on a
collection-by-collection basis.  (I hope to publicly release two collections
this week!)


Kevin C. Marsh, Executive Director
Information Access Institute
IAI at neosoft.com     http://www.neosoft.com/~iai



More information about the Web4lib mailing list