Re: SGML & Ontologies (etc.)

Jeff Kenyon <jkenyon@advtech.uswest.com>

Mail folder: SRKB Mail
Next message: Adam Farquhar: "Re: All agreed, but what's out there now?"
Previous message: Bill Rich: "SGML and CG's"
In-reply-to: Daniel D Suthers: "SGML & Ontologies (etc.) "

From: Jeff Kenyon <jkenyon@advtech.uswest.com>
Message-id: <9506091800.AA07606@tonto.advtech.uswest.com>
Subject: Re: SGML & Ontologies (etc.)
To: srkb@cs.umbc.edu
Date: Fri, 9 Jun 1995 12:00:50 -0600 (MDT)
In-reply-to: <Ajq6TKy8XZwyQ6ZH8O@pitt.edu> from "Daniel D Suthers" at Jun 9, 95 11:25:10 am
X-Mailer: ELM [version 2.4 PL23]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1044      
Sender: owner-srkb@cs.umbc.edu
Precedence: bulk

Daniel D Suthers writes:
> 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. 

While historically SGML has been primarily used for syntactic tagging
(due to the user communities that have adopted it), it's important to
realize that it's a metalanguage.  A SGML DTD can define syntactic
containers like paragraphs, chapters, tables, figures, or
whatever...or it can define semantic containers like objects, faults,
symptoms, tests, repairs, or whatever.  Those containers can be as
detailed as you'd care to make them.

At this point, I think it'd be a mistake to write off SGML as useless
for support of intelligent functionality.  As metalanguages go, it's
pretty powerful.  

Jeff

-- 
Jeff Kenyon (jkenyon@uswest.com)
Member of Technical Staff, Human Factors/Knowledge Base Engineering
U S WEST Communications, Boulder/Denver, Colorado
http://web2.advtech.uswest.com/~jkenyon/index.html