Re: CCAT Conceptual Catalogue and Ontologies
fritz@rodin.wustl.edu (Fritz Lehmann)
Date: Mon, 5 Sep 94 20:09:47 CDT
From: fritz@rodin.wustl.edu (Fritz Lehmann)
Message-id: <9409060109.AA10971@rodin.wustl.edu>
To: M.J.Johnson@qmw.ac.uk
Subject: Re: CCAT Conceptual Catalogue and Ontologies
Cc: cg@cs.umn.edu, srkb@cs.umbc.edu
Sender: owner-srkb@cs.umbc.edu
Precedence: bulk
Dear Mark Johnson,
I will add you to the CCAT "conceptual catalogue" list. You
could re-send your expression of interest to the list CG@cs.umn.edu,
since that list has little traffic right now, and many people
will be interested to see CCAT-related messages. Also, since IRDS
Conceptual Graphs are to be united at the semantic level with
KIF, you should send a copy to the list srkb@cs.umbc.edu. I
think that list (dealing mostly with ontologies done in KIF) will be
interested, for now at least, in CCAT-related discussions.
The general idea I have regarding proprietary systems
is that CCAT will provide for free a repository of off-the-shelf
generic concept-systems (both a "core" of "accepted" systems
and a wider group of alternatives and extensions), whereas the
money is to be made in propriertary systems for more specific
industries and activities. The latter can be _described_ in
the CCAT manuals, with instructions on how to buy them.
(Really a form of advertising.) If the proprietary extensions
are founded upon the core-systems (or specified extended systems
or alternatives in CCAT), so much the better. Those with
intended proprietary extensions should start now in telling
CCAT what demands they will make of the core system (examples:
will you require _points_ in time -- "time-stamps" -- in addition
to time-intervals? What varieties of PART-OF relations will you
need? Will you need any kind of "A represents B" relations?).
I surmise that the Knowledge Sharing Effort of ARPA in
the USA will have a similar policy on ontologies. I think
Tom Gruber maintains a growing repository at Stanford of
freely available "tested" KIF ontologies, while working
with Enterprise Integration Technologies, Inc. on more specific,
proprietary ontologies and concept-systems. CCAT ontologies
could be fully available to KSE participants, and vice-versa.
In fact, I suggest that every KSE ontology be considered as being
in the extended CCAT repository "de jure" and "ex officio",
with possibility of admitting the IRDS Conceptual Graphs version
of it to the CCAT "core ontologies". CCAT can, in addition, have
ontologies expressed in Conceptual Graphs which cannot be expressed
in current KIF (like "actors", Sowa-contexts, connected structures,
planarity, standard numbers, transitive closures, Buchi automata,
Montague grammars, power structures, Propositional Dynamic Logics,
etc.) That is, CCAT could include more general Conceptual Graphs not
restricted to the KIF-equivalent IRDS Conceptual Graphs.
Yours truly, Fritz Lehmann
GRANDAI Software, 4282 Sandburg Way, Irvine, CA 92715, U.S.A.
Tel:(714)-733-0566 Fax:(714)-733-0506 fritz@rodin.wustl.edu
=============================================================