Joint meeting of ANSI X3H4 and X3H7
sowa <sowa@turing.pacss.binghamton.edu>
Date: Sun, 14 Mar 1993 09:00:39 -0800
Message-id: <9303141654.AA01839@turing.pacss.binghamton.edu>
Comment: List name: SRKB-LIST (do not use email address as name)
Originator: srkb-list@isi.edu
Errors-To: neches@ISI.EDU
Reply-To: <sowa@turing.pacss.binghamton.edu>
Sender: srkb-list@ISI.EDU
Version: 5.5 -- Copyright (c) 1991/92, Anastasios Kotsikonas
From: sowa <sowa@turing.pacss.binghamton.edu>
To: Multiple recipients of list <srkb-list@ISI.EDU>
Subject: Joint meeting of ANSI X3H4 and X3H7
Last week, the ANSI X3H4 Committee on IRDS (Information Resource
Dictionary Systems) had a joint meeting with X3H7 on Object Information
Management. The meeting was very fruitful and another joint meeting is
planned for July 20 in Chicago. Following are some of the highlights:
1. The X3H7 Committee on OIM has the charter of looking at various
object models and trying to relate them. Their first step is to
draw up a matrix of features of the various O-O models. Their
general conclusion is that trying to develop a single model that
will support all existing and future O-O languages and systems
is probably impossible, and a model that supports only the
least common denominator of all existing O-O systems would be
too weak to be useful.
2. Some of the X3H7 members have a strong background in logic, and
they would like to use some version of logic for defining and
interrelating object models. As usual, however, there are others
who have the common fears that logic may be too difficult or too
unreadable for the average programmer to use effectively.
3. One of the X3H7 members, Haim Kilov from Bellcore, has been using
Z (pronounced "ZED") for defining object models. He has found the
current version of Z to be inadequate for representing O-O features.
There is a new version of Object Z that might be suitable. Haim is
sympathetic to the logic-based approach advocated by X3H4 and the
Knowledge Sharing Effort, but he doesn't want to get involved in
"religious wars" between competing systems.
4. On Wednesday, March 10, Sandra Perez,
Roger Burkhart, and I gave presentations on the IRDS Conceptual Schema
and the proposed normative language based on conceptual graphs. We
emphasized the agreement to adopt a common semantic base for KIF and
conceptual graphs (see the my note to these lists on Feb. 10, which
reported on a meeting with Richard Fikes, Jim Fulton, Mike Genesereth,
Bob Neches, and me). The presentations were favorably received, and
we spent another hour discussing some of the details.
5. In my part of the presentation, I showed how conceptual graphs could
be used to represent encapsulated objects and messages to and from
objects. I emphasized the point that I was not trying to develop
yet another object model, but to demonstrate several points:
a) A graphic representation, such as conceptual graphs, can be as
readable as any of the popular diagrams, such as E-R, NIAM, etc.,
but it has the full expressive power of predicate calculus.
b) The CG context mechanism, which is definable in KIF, is a
natural way of defining the O-O encapsulations.
c) The metalanguage capabilities of KIF and CGs appear to be powerful
enough to define various systems of logic that are being used in
AI, and we believe that they are general enough to define the same
kinds of capabilities that are being used in Object Z and other
new logics for O-O systems, including the new F-logic.
These points were well received, but they want more details to
demonstrate that the KIF-CG coalition can indeed support all of
their needs.
The general conclusion was that this approach is very promising, and
if we can demonstrate that the KIF-CG work can satisfy X3H7, we may be
able to bring the KSE work into the O-O world. That would go a very
long way towards bringing it into the mainstream of data processing.
John Sowa