Re: ANSI X3H4 meeting next week Robert Neches <email@example.com>
To: Gio Wiederhold <firstname.lastname@example.org>
Cc: email@example.com, Marianne Siroker <firstname.lastname@example.org>,
email@example.com, SRKB@isi.edu, INTERLINGUA@isi.edu,
KR-ADVISORY@isi.edu, GINSBERG@t.stanford.edu, SKPEREZ@mcimail.com,
Subject: Re: ANSI X3H4 meeting next week
In-reply-to: Your message of Fri, 04 Oct 91 11:31:02 -0700.
Date: Fri, 04 Oct 91 14:16:08 PDT
From: Robert Neches <firstname.lastname@example.org>
>For our mechanical systems I would prefer that a single module NOT consider
>multiple ontologies (near-)simultanously. Few humans are good a keeping two
>sets of books distinct, even when motivated by the IRS.
>On the other hand, layering, where one module reasons about the ontology of
>another module, seems fairly managable (as .. this person makes that
>statement, because his/her background is such and so --) seems to managable
>if we can abstract concepts as background, scope, contex, and reasoning power.
Perhaps I should have made myself clearer about time frames. I believe that
individual modules which can maintain multiple conflicting ontologies are
ultimately necessary. However, I think I'll retired in my rocking chair before
we have them working. In views on the near term, in fact, I think I'm actually
more conservative than you: I view the kind of layering you describe as
something to work up to rather than something we're clearly ready for.