W3C home > Mailing lists > Public > www-webont-wg@w3.org > August 2002

Re: "Definition" of Ontology

From: Leo Obrst <lobrst@mitre.org>
Date: Tue, 06 Aug 2002 16:35:10 -0400
Message-ID: <3D5032FE.E9C51007@mitre.org>
To: Christopher Welty <welty@us.ibm.com>
CC: www-webont-wg@w3.org

Chris,

I heartedly concur with your paper's ontology definitions (though I would put "thesaurus" to the right of "taxonomies"
in your figure and do so in my own "ontology spectrum" figure), since the latter in practice are very ill-defined and
at least standard thesauri, though focused on terms rather than concepts, define 4-5 "semantic" relations. Taxonomies
tend to be completely arbitrary and semantically inconsistent (though of course, *we* well-define them).

So in general, I think we need an initial definition which is even simpler. Perhaps OMG doesn't need this, I don't
know.  Given my experience at trying to communicate/define ontology/ies extremely simply to non-ontologists, less
technical folks, i.e., to  managers, prospective users , and  technical non-ontologist developers (together our main
audience), I've come up with the following, which I argued for adoption into our OWL requirements document:

http://lists.w3.org/Archives/Public/www-webont-wg/2002Feb/0093.html

The actual wording was slightly modified in the requirements doc: http://www.w3.org/TR/webont-req

Leo

Christopher Welty wrote:

> Webonters,
>
> I noticed in Evan's recent message about OMG that one action item was to
> accept "a definition of ontology".  People (in computer science) have been
> trying for the past decade to define what we mean by this term, but most
> definitions I'm aware of are fairly vague.  Exclusive definitions tend to
> leave out things that should rather obviously be included, and inclusive
> definitions seem to allow things that rather obviously shouldn't.
>
> I still prefer an inclusive view, and last year a philosopher in the area
> of ontology (Barry Smith) and I, in order to introduce a conference which
> attempted to bring together philosophers and computer scientists
> interested in ontology (FOIS - http://www.fois.org), wrote a paper
> (attached) which discusses this point.
>
> Every attempt to "define" ontology I'm aware of has been based on the
> definer's experience(s), and usually ends up being more of a description
> of "what I call ontology" (as one such person once put it).  The attached
> article is different in that it has been extensively researched and tries
> to explain (rather than define) the meaning of "ontology" by tracing the
> history of the term, and how it came to be used (in both fields) the way
> it is used today.  I find such etymological explanations are much more
> enlighting, because language evolves.
>
> -Chris
>
> PS: Note that the attached article is copyrighted by ACM, and I have
> permission to redistribute it as long as the copyright remains, ostensibly
> as a publicity measure to attract attention to the availability of the
> FOIS proceedings.
>
> PPS.  The reference is: Smith, Barry and Chris Welty. 2001. Ontology: Towards a new synthesis. In
> Chris Welty and Barry Smith, eds., Formal Ontology in Information Systems. Pp. iii-x. Ongunquit, Maine: ACM Press.
>
>
>
> Dr. Christopher A. Welty, Knowledge Structures Group
> IBM Watson Research Center, 19 Skyline Dr.
> Hawthorne, NY  10532     USA
> Voice: +1 914.784.7055,  IBM T/L: 863.7055
> Fax: +1 914.784.6078, Email: welty@us.ibm.com
>
>   ------------------------------------------------------------------------
>                             Name: fois-intro.pdf
>    fois-intro.pdf           Type: Acrobat (application/pdf)
>                         Encoding: base64
>                  Download Status: Not downloaded with message

--
_____________________________________________
Dr. Leo Obrst  The MITRE Corporation
mailto:lobrst@mitre.org Intelligent Information Management/Exploitation
Voice: 703-883-6770 7515 Colshire Drive, M/S W640
Fax: 703-883-1379       McLean, VA 22102-7508, USA
Received on Tuesday, 6 August 2002 16:35:30 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:57:51 GMT