SGML & Ontologies (etc.)

Daniel D Suthers <suthers+@pitt.edu>

Mail folder: SRKB Mail
Next message: Bill Rich: "SGML and CG's"
Previous message: John F. Sowa: "Re: Doug Lenat's THE BIG PICTURE"
In-reply-to: Sam Hunting: "Re: Doug Lenat's THE BIG PICTURE"
References: Sam Hunting: "Re: Doug Lenat's THE BIG PICTURE"
Reply: Jeff Kenyon: "Re: SGML & Ontologies (etc.)"

Message-id: <Ajq6TKy8XZwyQ6ZH8O@pitt.edu>
Date: Fri,  9 Jun 1995 11:25:10 -0400 (EDT)
From: Daniel D Suthers <suthers+@pitt.edu>
To: srkb@cs.umbc.edu
Subject: SGML & Ontologies (etc.) 
CC: Sam Hunting <sgmlsh@world.std.com>
In-reply-to: <Pine.3.89.9506081623.A20906-0100000@world.std.com>
References: <Pine.3.89.9506081623.A20906-0100000@world.std.com>
Sender: owner-srkb@cs.umbc.edu
Precedence: bulk

Regarding SGML and the ontology discussion ...

The SRKB community is most concerned with *semantic* commonalities in
document sharing, while SGML has addressed primarily *syntactic* level
isues. 

The two are of course compatible. The SGML community has recognized the
need for shared semantics, but for most of them "semantics" is how you
format the abstract document -- hence the DSSSL effort. For folks like
us who are interested in supporting "intelligent" functionality,
something else is needed. 

We have new funding based on a proposal to (among other things) use
KIF-based ontologies available on a WWW server to provide the semantic
basis for documents (also available on a server), just as an SGML DTD
(a.a.o.a.s.) provides the syntactic basis for the documents. (The
primary emphasis of the proposal is on combining collaborative learning
and coached apprenticeship educational approaches, and making the
supporting technology more widely available over the net via generic WWW
tools as well as via our own specialized software.) 

We have not decided whether we will actually use KIF.  The decision will
be based less on expressiveness and related issues, and more on things
like whether people in the educational technology community give a damn
about the kinds of things KIF can do for you, whether there are
ontologies available that do them, or whether the cost of constructing
and publicizing such ontologies is sufficiently low, and whether the
whole approach (KIF and ontologies) appear to have enough of a future as
a shared resource to be worth putting the investment into using them. 

================================================================
Dan Suthers         	| Learning Research & Development Center
suthers+@pitt.edu   	| University of Pittsburgh
(412) 624-7036 voice	| 3939 O'Hara Street
(412) 624-9149 fax	| Pittsburgh, PA 15260
(412) 363-3992 home	| http://www.pitt.edu/~suthers/
================================================================