ontolog-forum
[Top] [All Lists]

Re: [ontolog-forum] Curation view of Ontologies

To: "[ontolog-forum]" <ontolog-forum@xxxxxxxxxxxxxxxx>
From: "John F. Sowa" <sowa@xxxxxxxxxxx>
Date: Mon, 23 Jun 2008 00:20:32 -0400
Message-id: <485F2490.1040500@xxxxxxxxxxx>
Pat and Ken,    (01)

Ranganathan's system is not an ontology, since it does not
contain any axioms.  But it is a very detailed, flexible,
and extensible classification system.    (02)

PC> Do you think that a topic classification system suitable
 > for libraries can actually serve to encode information in
 > a form suitable for automated reasoning?    (03)

Any ontology must include a classification system as a subset.
If an ontology is going to be usable for the full range of
topics that are covered by the WWW, it must include a system
of classification as broad as Ragnanathan's.    (04)

PC> Do you have any examples of how such a classification has
 > been used with an automated reasoner?    (05)

Look at Cyc.  It has a very broad classification system that
supports multiple inheritance.  Any ontology that is intended
to support interoperability among any range of systems that
access and use the WWW, must be as extensible as Cyc and/or
Ranganathan's system.  (I'm not recommending either one
exactly as is, but I do suggest that we study both.)    (06)

KL> My thought has always been I build the ontologies to relate
 > the various classification facets (or the values applied to
 > the facets) without having to assume a governing set of facets
 > and values for all uses.    (07)

There is no reason why you should define more than you need for
your application.  But if two or more groups want to make their
systems work together, it must be possible to embed all of them
within a broader system that can accommodate all the categories
of each.  Therefore, it's desirable to have a broader framework
in mind, even for a small, narrowly defined project.    (08)

An important reason for studying a system such as Ranganathan's
is that he had spent many years trying to work out categories
that can accommodate all the books in all the libraries of India.
So he has a system of categories that spans an enormous range,
which probably includes a large part of what is being covered
by the WWW.    (09)

I don't recommend that we adopt R's system exactly as is.  But
we should consider it or something like it as an example of how
to accommodate multiple systems interoperating across the WWW.    (010)

John    (011)


_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/  
Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/  
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/ 
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx    (012)

<Prev in Thread] Current Thread [Next in Thread>