Yahoo, metadata

[Karen G. Schneider]

(To the tune of "Thumbelina")
Metadata, metadata, tiny little thing;
Metadata dance, metadata sing...

Seriously, Stephen Thomas raises excellent issues, none of which I argue
with.  A cursory review of Bib Formats (y'all still take cataloging in
library school, right?) confirms we DO have existing practices--e.g.
Intercat, etc.  I am a brave woman who has sat through a MARBI meeting to
hear catalogers carefully discuss the innards of the MARC fields related to
electronic resources, and let me tell you they are fast-moving and reten--I
mean, attentive folks.  Bless those catalogers.  However, we do NOT have
existing money or guidance.  As Stephen Thomas didn't say, "show me the
records!"  (What HAS gotten into me today?)  

I do think that if we took the labor used to make all the redundant and--in
relation to how we organize other information--somewhat primitive finding
aids floating around, we would have a pool of resources for cataloging
electronic resources.  In some ways I think we're on the cusp of a place we
have been to before in our profession--when the Anglo American Cataloging
Rules were developed, and all those disparate methods were (pretty much)
standardized into the guidance that allows us to have such fantastic tools
such as OCLC.  (This is not to be critical of some of the interesting
projects cited yesterday, many of which may fall into the important
Brainiac's Playground category, or of useful aggregates of information,
such as the IMDB recently cited.)  But for some reason when we talk about
cooperative cataloging, we think about books, not electronic resources.
Why? Or perhaps the operative question (since we all kinda KNOW why) is
what now?  

Howsoever... to clarify something... the MARC record IS metadata!   Every
original record in OCLC, RLIN or WLN is a new metadata record. It isn't
designed to be attached to the thing it describes, mostly, because such a
thing was impossible when these massive cooperative catalogs were designed,
not because that's ideal.  You can't merge an electronic record with a
paper book.  You couldn't even make a paper catalog card attach itself to
the book.  At the time, you couldn't even THINK that way.  But you CAN use
a record to describe a website, and it can attach and/or point to what it
describes.  I don't have the expertise to say whether it is better for the
metadata to be separate, connected, redundant, or be able to sing; I just
dunno.  

At any rate, every new record entered in a bib utility such as OCLC for an
electronic resource is another potential link from a web-based OPAC.
Current Cites recently lauded librarian/LITA member Amanda Xu for an
article she wrote on a radical new tool for accessing information--the
OPAC! Is this the revolution of the fin de siecle?  I'm all for it!  What
do we want?  MORE CATALOGING!  When do we want it?  NOW!  Access for the
masses!  

Re some other questions I got yesterday.  (Oh oh, watch out, I have a
MASTERS in LIBRARY SCIENCE... beware!  "Nurse Internet" strikes!)  Someone
asked, which standard?  How would I know?  But anyway, I know that there is
a MARC/Dublin Core cross-walk. So are we really talking standards, or
technology for accessing the metadata? Someone asked me what standard PICS
uses.  That's what's interesting about PICS... it doesn't describe
metadata; it's a specification for enabling standardized description of
content.  With the RSACi ratings for sex, violence and nudity, the TIFAP
website is:

<META http-equiv="PICS-Label" content='(PICS-1.1 
"http://www.rsac.org/ratingsv01.html" l gen true comment 
"RSACi North America Server" by "kgs at bluehighways.com" for 
"http://www.bluehighways.com/tifap/" on "1997.08.30T12:22-0800" 
r (n 0 s 0 v 0 l 1))'>

n=nudity; s=sex; v=violence; l=language.  But this is arbitrary (as all
descriptive tools, from alphabetical order to LCSH, are arbitrary), and is
related to the RSACi ratings (see http://www.rsac.org)--not to PICS itself,
which *does not describe content.*  There's no reason why there couldn't be
a MARC standard interpreted for the PICS label (though its fields would be
much longer).  An interesting thing about PICS labels is that they do NOT
have to be imbedded in the document they describe (though they can be).
(This also means there is no limit to the number of PICS labels that can be
used to describe a resource, at least as far as I know; the TIFAP main page
has two built-in, and presumably any other label bureau could create
another label for it.)  The retrieval tool determines which PICS label is
operative--so even if there were a zillion PICS labels available for any
one resource, if your OPAC specified the XYZ labels, that is what would be
operative.  

Like OCLC records, PICS could include Dewey and LC, use standard language,
and include all kinds of information, some of which (as I have suggested)
could perhaps be automatically generated (again, because the medium allows
it; books can't do that because they're made of paper).   OCLC, as has been
discussed on this list, has done some interesting stuff involving automatic
assignment of metadata.  The fact that these projects are still not quite
ready for prime time is not meaningful, because I wouldn't expect such
experimental technology to work right for the first umpty-ump iterations.
That's why it's experimental.  The tweedy folks at OCLC are doing good
stuff for us, even if they don't know how to market it.

I bet, furthermore, PICS labels could be *generated* from MARC records.
Would we want to do that, in lieu of or in conjunction with more
traditional cataloging of Internet resources?  Why would that be useful?
Heck, again, I dunno.  I'm not Seymour Lubetsky; I don't even play him on
TV.  Maybe, with our existing cataloging tools, we have always had all the
tools we ever needed, except for money, direction and momentum.  

______________________________________________
Karen G. Schneider |  kgs at bluehighways.com
Director, US EPA Region 2 Library  |  Contractor, GCI
Councilor-at-Large, American Library Association
The Internet Filter Assessment Project:   
 http://www.bluehighways.com/tifap/
Author, Forthcoming: A Practical Guide to Internet Filters
(Neal Schuman, 1997 ISBN 1-55570-322-4)
Information is hard work  -------------------------------------------