Re: WWW space needs a card catalog for navigation...

fritz@rodin.wustl.edu (Fritz Lehmann)
Date: Fri, 13 May 94 06:25:39 CDT
From: fritz@rodin.wustl.edu (Fritz Lehmann)
Message-id: <9405131125.AA26181@rodin.wustl.edu>
Newsgroups: comp.infosystems.www,comp.infosystems,soc.libraries.talk
Subject: Re: WWW space needs a card catalog for navigation...
References: <nicka.1118588387B@156.151.6.30> <howden.94.768320113@lis.unt.edu> <2qggg6$ifh@spud.Hyperion.COM> <CpHCLu.CJ3@lut.ac.uk>
Organization: Center for Optimization and Semantic Control, Washington University
Apparently-To: wille@mathematik.th-darmstadt.de
Apparently-To: srkb@ISI.EDU
Apparently-To: cg@cs.umn.edu
Sender: srkb-owner@cs.umbc.edu
Precedence: bulk

     In the usenet Newsgroups: comp.infosystems.www,comp. 
infosystems, and soc.libraries.talk , 
M.D.Grimleigh1@student.lut.ac.uk wrote:
---------begin quote---------------------------------------
In article <2qggg6$ifh@spud.Hyperion.COM>
koreth@spud.Hyperion.COM (Steven Grimm)
writes:
>The first step, of course, would be to make the classification codes
>available via the Web.  Anyone know of an online list of Dewey decimal
>codes?  Or is that not what librarians like to use nowadays?

Standing by for flames on this but *I* think that Dewey is not
suitable for on-line browsing.  I've been having a look at
Ranganathan's Colon system and I suspect that that may be far
more useful for this kind of purpose, each thread of interest
(trying to avoid librarian jargon here for non lib. types) is
recognisably identified, rather than being buried in a long
number which will take quite a complex algorithm to
de-synthesize and return what aspects the classifier actually
considered in allocating the code.

Over to the Cat. and Class. experts....

 <-*-><-*-><-*-><-*-><-*-><-*-><-*-><-*-><-*-><-*-><-*-><-*-><-*->
 |   ______    ____   Milton Grimleigh, c/o Dept. of Information |
 |  /  /  /   / __    & Library Studies, Loughborough University,|
 | /  /  / * /___/ *  Loughborough, LE11 3TU, ENGLAND            |
 |  M.D.Grimleigh1@lut.ac.uk  mgrimlei@nyx.cs.du.edu             |
 | Where is the knowledge we have lost in information?  T.S.Eliot|
 <-*-><-*-><-*-><-*-><-*-><-*-><-*-><-*-><-*-><-*-><-*-><-*-><-*->
------------end quote--------------------------------

     I agree completely -- in theory, Ranganathan's "faceted"
classification system COLON surpassed Dewey/Library of Congress/Bliss
many years ago.  The notion of faceted knowledge (in COLON the facets
are Personality, Matter, Energy, Space and Time) is essential.  Different
aspects (facets) of a document are relevant for different purposes.
Where should a book on "Eighteenth-Century Japanese Cast Lead Statuary" go?
Under Japan?  Art? Casting? Lead?  The 18th C.?  Different purposes
("points of view") emphasize different aspects of a document.  Even
Ranganathan's theory is not enough -- we need to recognize multiple
classifications of knowledge for different purposes.  I have argued
within Artificial Intelligence that a mere vector of facets is not enough
to capture the full hierarchy of possible descriptions. (Faceted semantic
networks or order-sorted relational structures are needed.)  Further,
there must be a formal theory by which different "ontologies" (taxonomies)
covering the same domain can be integrated and communicate.  These
ideas are involved in the current Knowledge Sharing Effort in A.I.  It
would be foolish for WWW to wed itself irrecoverably to outmoded "tree"
classifications like Dewey or L.C.

     Already, the UMLS Metathesaurus for medical information (at
Nat. Lib. of Medicine in US) has gone far beyond the monolithic tree
stage.  Information on competing systems of knowledge organization 
appears in publications of the International Society for Knowledge 
Organization, (ISKO), Woogstrasse 36a, D-60431 Frankfurt, Germany.
Information on the Knowledge Sharing Effort may be accessed by WWW
from a server at Stanford or by gopher at hpp.stanford.edu.

     There is a triadic relation among A. a particular document, object,
passage, description or proposition (the "extension"), B. a category
which conveys significant information, within a classification
scheme (the "intension"), and C. the purpose (or persons with
purposes) served by such a categorization.  New theories now being 
developed will take account of all three arguments of this relation.

                          Yours truly,   Fritz Lehmann
GRANDAI Software, 4282 Sandburg Way, Irvine, California 92715, U.S.A.
Tel.: (714)-733-0566  Fax: (714)-733-0506  fritz@rodin.wustl.edu
====================================================================