Basic Semantic Repository & ISO 11179 & Z39.50 ?

Fritz Lehmann <flehmann@orion.oac.uci.edu>
Date: Fri, 21 Jul 1995 02:05:36 -0700
From: Fritz Lehmann <flehmann@orion.oac.uci.edu>
Message-id: <199507210905.AA11758@orion.oac.uci.edu>
To: bsr@premenos.com, cg@cs.umn.edu, edi-new@tegsun.harvard.edu,
        fritz@rodin.wustl.edu, srkb@cs.umbc.edu, z3950iw@nervm.nerdc.ufl.edu
Subject: Basic Semantic Repository & ISO 11179 & Z39.50 ?
Sender: owner-srkb@cs.umbc.edu
Precedence: bulk


     Klaus-Dieter Naujok's question on his EDIFACT/BSR (Basic Semantic
Repository) list bsr@premenos.com reminds me to ask another.  My question
is about three (four?) international standards for "semantic repositories" (or
possibly "ontologies" in the recent AI sense):

     What is the exact relationship between the (X3L8) ISO 11179 draft Data
Metamodel standard and the Basic Semantic Repository (BSR) for EDI (Electronic
Data Interchange/Electronic Commerce)?  Are they separate efforts,
or are they one and the same?  If separate, are they guaranteed to
be in substantial harmony?  And are they coordinated in any way with the
emerging ISO Z39.50 "semantic repositories" for automated data and document
transfers? -- These include semantic categories ("attribute sets" -- of
data elements): BIB1 for the semantics of documents, and STAS, a somewhat
loosely unified set of a couple of thousand scientific and technical data
elements taken from Chemical Abstracts and other sources. 

     I'm impressed with the ISO 11179 draft I've seen; it recognizes the
practical necessity of using a principled (and somewhat lofty) semantic
definition of basic concepts needed to define more complicated meanings
of particular data elements and value-codes.  I do have some suggestions
to correct what I see as structural shortcomings of the "content" chapter
of ISO 11179.  Where (on email or at meetings) should such suggestions be
made and discussed?

     I also am impressed with certain aspects of the Z39.50 initiative.
This first emerged from the efforts of libraries to share information
on documents, books, etc., and it benefits from many years of developing
the elaborate "MARC" data format for cataloging documents.  The library
community has developed a lot of doctrines over the years on establishing
specific semantics for data elements, including their notion of an
"authority", which relates to the EDI ideas of X12 Element 66  or the
EDIFACT 1131/3055 code-authority pairs.  The STAS Z39.50 "attribute set"
duplicates many scientific measurement units and other things from the X12
and EDIFACT test/measurement elements and value-codes.

     Furthermore (if anybody knows), is there any coordination of the
semantic standards just mentioned and the CSMF (Conceptual Schema Modelling
Format) Standard's intended "semantic" component?

     This question is crossposted to the EDI-L, EDI-NEW, SRKB (shared
ontologies), CG (conceptual graphs) and Z39.50 lists, in the hope of
getting some answers from someone aware of ISO 11179, BSR and/or Z39.50.


                          Yours truly,   Fritz Lehmann
GRANDAI Software, 4282 Sandburg Way, Irvine, CA 92715, U.S.A.
Tel:(714)-856-0671               email: fritz@rodin.wustl.edu
=============================================================