Results of ISO CSMF Mtg. in Southampton, UK (Anthony K. Sarris)
Date: Mon, 15 Aug 94 13:55:22 PDT
Message-id: <>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
From: (Anthony K. Sarris)
Subject: Results of ISO CSMF Mtg. in Southampton, UK
Precedence: bulk
Thought your mail list might find this of interest: progress is being made
on an international standard for conceptual schema modelling facilities:

Tony Sarris

Ontek Corporation
22941 Mill Creek Drive
Laguna Hills, CA, USA 92653
Phone: 714/768-0301
Fax: 714/768-0851


17 people from 8 national bodies participated in the Conceptual Schema
Modelling Facilities (CSMF) rapporteur group meeting held July 18-28, 1994
in Southampton, UK as part of the International Organization for
Standardization (ISO) SC21 Open Systems Interconnection, Data Management
and Open Distributed Processing, Working Group 3 - Database meetings.

This was the first real opportunity to explore technical issues in detail,
and even with some rather heated discussions over various technical points,
considerable progress was made. There is a high degree of consensus on
basic technical objectives, principles and requirements. There are still
some differences of opinion over definitions of certain terms, although by
the end of the meeting the group had reached compromise agreements that
will allow us to go forward in this area. There is even agreement on the
basic aspects of an architecture, although the specifics of that
architecture are still being formulated (there are several ideas in this
area, with most being variations of each other, but with one national body
having a considerably different architectural view based on a different
(more limited) view of CSMFs and their role and usage in an information
engineering life-cycle). The group agreed to progress the results of this
initial work in a single document described as a 'pre(liminary)-Working
Draft' of the ISO CSMF standard. Since it is not an official Working Draft
(WD) yet (and isn't scheduled to be until July 1995), this version of the
document is only being circulated within the CSMF RG and the SC21 Working
Group 3 - Database community. The official WD version will go to ISO SC21,
and will also be distributed to liaisons in other ISO SCs such as
SC7,11,14,22 and 30, as well as other standards development organizations.

There was some initial discussion of the Common Logic Foundation (CLF) and
the semantic constructs that would be a part of the CSMF Normative Schema.
Here again, there is broad agreement on these notions and on a way to
proceed with processing contributions in these areas. One national body had
some concerns about a syntax based on logic, but agreed to wait and see
whether contributions in this area would overcome its concerns. The major
issue concerning the semantic constructs is how to keep the initial set for
standardization down to a reasonably small number (a 'minimal set'), while
at the same time ensuring that those initial constructs are sufficient to
meet the requirements of basic conceptual modeling. This includes
requirements for integration/subsumption of existing enterprise modeling
methods that are focused on specific areas of modeling (e.g., data,
information, process, state/event, data flow, etc.).

The US is intending to bring its contributions on CLF and semantic
constructs to the next ISO CSMF RG meeting, scheduled for January 1995 in
Albuquerque, New Mexico. The next two meetings of the US TAG (ANSI X3T2
Information Interchange and Interpretation) will focus primarily on edits
to the pre-WD document and proposed US contributions on CLF and semantic
constructs. Regarding CLF: the US is looking to Sowa, Genesereth and Fulton
to prepare a unified position on CLF based on Conceptual Graphs (CGs),
Knowledge Interchange Format (KIF) and the STEP Semantic Unification
Meta-Model (SUMM). In the area of semantic constructs: several national
bodies have provided a 'sneak preview' of their positions. Japan will
almost certainly base its contributions on the object-oriented constructs
underlying the Japanese Data Modelling Facility (JDMF). JDMF is intended to
bridge E-A-R and object-oriented approaches, but the objects subsume the
entities. France will also likely propose constructs with an o-o flavor,
based on work with process/state/event modeling. The Netherlands indicated
a more general set of constructs based on fundamental ontology. This should
be fairly consistent with the emerging US position. Based on previous
contributions, positions and discussions at the Southampton meeting, other
national bodies including and Canada and Australia should also be
supportive of an approach with broad converage based on commonly used
relations. The UK is participating in the work but has not made a firm
commitment to continue. UK experts at the Southampton meeting were
supportive of a general approach that reflects the fundamental notions of a
conceptual schema, as described in ISO TR9007.

The next meeting of the US TAG (ANSI X3T2) is scheduled for September 12-16
in Seattle/Bellevue, WA. There will be another X3T2 meeting in early-mid
November to finalize US input for the January ISO CSMF meeting. The initial
outputs of the Sept. meeting will be circulated internationally to CSMF
participants to get feedback and to allow for response to that feedback at
the Nov. X3T2 meeting. Assuming that consensus and overall progress
continue on the track established in Southampton, there should be no
problem meeting the July 1995 schedule for a Working Draft CSMF standard.
Given that the standard is progressing quite rapidly now, those who wish to
contribute to the work are encouraged to get involved in their respective
national body efforts ASAP. For information on the US effort within ANSI
X3T2, please feel free to contact either me or Dr. John Sharp, X3T2
Chairperson (