Organizing Web Information

[Rhyno Art]

>With the advent of desktop publishing over the past 13 years (recall that laser
>printers were barely accessible to mortals prior to 1982) before the explosion
>in accessibility to the internet, the traditional library model has not been
>able to document the proliferation of self-published work.  If the paper
>library hasn't had to deal with this kind of publication as a peer of the book,
>then why should the digital library?
> 
>-marc
>

The difference may be that Alta Vista and other similar search engines could 
indeed be a partial solution for keeping up with the massive amount of output 
generated by the Internet and, to echo an earlier post, I think that there would
be great value in having access to such services using Z39.50. Cataloguing and 
classifying the "wheat" is very worthwhile and projects that attempt to do this 
deserve widespread support, but there are advantages in using a tool that may 
have the best chance of catching many resources as soon as they appear, 
particularly if they do so in the "6 month" or whatever time frame that the 
documents have the most value. Z39.50 may even offer a toolkit for 
specifying the types of searches that most Internet search engines handle with 
the Type 102 Ranked List Query Spec (see 
<http://potomac.ncsl.nist.gov/~over/rlq/RLQ_spec_Feb96_ZIG.html>). Accessing 
these gigantic databases using a standard protocol would be one way to 
deal with Internet resources that simply can't be catalogued in a timely
manner. We have patrons that take words and phrases they see in an OPAC display 
and paste them into an Alta Vista search, and vice versa. It is possible that 
we could construct Web interfaces to our catalogues that could do things like 
use a bibliographic record as a source of keywords for running against an 
Internet search engine, use our understanding of indexing techniques to 
automatically add qualifiers to searches in order to increase the odds of 
identifying useful resources, and attempt to offer the Internet as a solution 
when the local collection, even with locally catalogued "stable" Internet 
materials, simply may not have what a patron is looking for. To a certain 
extent, this can be done now with enough cgi-bin hacking and studying the URLs 
generated by Internet search engines, but I think that Z39.50 could go a long 
way towards helping us take advantage of Internet search engines on our own 
terms. I think that another challenge we face in addition to improving the level
of indexing and classificaton on the Internet is to get one of the major players
in the Internet indexing world to regard Z39.50 compliance as a strategic 
direction. 

Just my $0.02...

art
---
Art Rhyno, Systems Librarian
Leddy Library, University of Windsor
Internet: arhyno at server.uwindsor.ca
Tel: (519) 253-4232, EXT. 3163
FAX: (519) 973-7076
WWW: <http://www.uwindsor.ca/library/leddy/people/art.html>