Summary of ANSI IRDS and ISO SWG-MF meetings in December

sowa@watson.ibm.com
Date: Fri, 29 Jan 1993 19:08:33 -0800
Message-id: <199301300305.AA24068@venera.isi.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@watson.ibm.com>
Sender: srkb-list@ISI.EDU
Version: 5.5 -- Copyright (c) 1991/92, Anastasios Kotsikonas
From: sowa@watson.ibm.com
To: Multiple recipients of list <srkb-list@ISI.EDU>
Subject: Summary of ANSI IRDS and ISO SWG-MF meetings in December

During the week of December 7 to 11, there were two overlapping ANSI and
ISO meetings that are relevant to the Knowledge Sharing Effort:

 1. A meeting of the ANSI X3H4 Committee on Information Resource
    Dictionary Systems (IRDS) in New Orleans.

 2. A meeting of the ISO Special Working Group on Modeling Facilities
    in Namur, Belgium.

I wasn't able to attend the X3H4 meeting, but from other participants, I
heard that it went very well.  The technical report on the IRDS Conceptual
Schema, which the X3H4.6 Task Group had been writing during the past year,
was approved by the full X3H4 committee.  It will be published by ANSI in
early 1993.

The X3H4 TR recommends a logic-based approach for unifying different
representations.  As "the initial normative language," it proposes
conceptual graphs, but says that any other notation with equivalent
semantics could also be used as a normative language.  During the week
of January 18, 1993, Richard Fikes and Mike Genesereth from Stanford,
Jim Fulton from Boeing, Bob Neches from ISI, and I met to discuss the
relationships between KIF, conceptual graphs, and SUMM (the Semantic
Unification Meta-Model).  Next week, I'll send a more detailed report
on that meeting, but as a brief summary, we agreed to propose both
KIF and CGs as semantically equivalent normative languages and with
SUMM as an ontology of metalanguage primitives for defining modeling
languages such as SQL, Express, NIAM, and E-R diagrams.

Besides approving the technical report, the X3H4 Committee also voted
unanimously for an SD-3 (Standing Document #3 for a New Work Item) to
develop the approach recommended in the technical report as a Draft
Proposed Standard for the IRDS Conceptual Schema.  If the SD-3 is approved
by the full ANSI board, the recommendations in the TR will be continued
in collaboration with ISO, other ANSI committees, and the Knowledge
Sharing Effort.  Since the object-oriented languages must also be brought
into this coalition, the X3H4 committee on IRDS will be meeting with the
X3H7 committee on object management in March 1993.  There are a number
of issues that have to be resolved, especially the relationship between
the logic-based approach of KIF, CGs, and SUMM and the procedural approach
of most O-O languages.

At the ISO meeting in Namur, the US delegates included members of ANSI
X3H4 on IRDS, X3T2 on Communications, and Bob Neches of the Knowledge
Sharing Effort.  The topics were a continuation of the themes discussed
at Renesse, the Netherlands, in March 1992.  As before, the main
confrontation was between Bill Olle from the UK and the rest of the
world.  Olle kept trying to narrow the scope to "data modeling" using
SQL as the principal language.  Everybody else wanted a broader approach
to "conceptual modeling" using a logic-based approach.  Olle kept
insisting that the approach should be practical instead of research.
Since the term "research" is rather vague, I suggested that we use the
term "implemented", which is more objective and easier to verify.  By
the criterion of implementation, conceptual graphs, KIF, and many other
logic-based approaches would be included, while much of the current SQL2
standard (the extension to SQL approved in 1992) and the "object-oriented"
SQL3 would be excluded.

We're hoping that Olle and the SQL diehards may be satisfied or at least
mollified by a logic-based approach with at least one normative language
to be expressed in SQL-like syntax.  The next ISO SC21 (Subcomittee 21)
meeting will be in June in Yokohama, Japan.  By then, the group must
complete a report with recommendations to SC21.  If SC21 approves, the
group will continue working towards a standard, in conjunction with ANSI
and other national bodies -- e.g. BSI (British Standards Institute),
DIN (Deutsches Institut fuer Normung), AFNOR (Association Francaise de
Normalisation), etc.

John Sowa