Re: "what is an ontology?" stuff in requirements abstract/intro

Pat Hayes wrote:

> >Dan,
> >
> >Most of these comments appear to be improvements, so we'll incoroprate
> >them into the document. I'll work on a rewording of "the problems with
> >lack of semantics in XML" to describe specific problems for specific
> >uses.
> >
> >Jeff
> >
> >
> >Dan Connolly wrote:
> >>
> >>  Our requirements document is going to
> >>  be the first exposure that some folks
> >>  get to our work; I can imagine
> >>  it showing up in Robin Cover's XML
> >>  new stuff; he'll probably grab the abstract:
> >>
> >>  "This document specifies goals, requirements, and usage scenarios for
> >>  the OWL web ontology language."
> >>
> >>  I can imagine xml-dev/www-talk folks saying, "er..
> >>  gee, thanks; now what's an ontology
> >>  language?"
> >>
> >>  That's elaborated later in the document; let's
> >>  see if there's some text to grab... yes:
> >>
> >>    Put simply, an ontology is just a set of
> >>    standard vocabularly terms along with some
> >>    formal definitions of the terms.
> >>
> >>  Lightly edited:
> >>
> >>    An ontology is vocabularly of terms along
> >>    with some formal definitions of the terms.
> >>
>
> Err.. I'd prefer it if you could avoid using the word 'definition'
> here. The point being that ontologies only describe, in general,
> rather than define.  If we say that a description is a definition
> then we can get into all kinds of trouble, eg consider Peter's
> pseudo-paradox problem which arises from thinking of an RDF
> description as a *definition* of OWL syntax.
> <snip>

I agree,   i use the term description unless I am talking about necessary and
sufficient conditions for class membership.

>
> >.....>
> >>    The term ontology may be unfamiliar to many readers of
> >>    this document.
> >>
> >>  That seems superfluous. I suggest striking it.
> >>
> >>    This notion of ontologies comes from Artificial Intelligence,
> >>    where ontologies are used to allow heterogeneous systems to
> >>    exchange and reason with information.
> >>
> >>  I'd suggest either citing specific work in this area
> >>  or striking the reference to Artificial Intelligence.
>
> I agree. In any case, you could equally well cite data modelling
> languages, say; and the basic ideas go back way before AI if you want
> to get historical, at least to the 1940s and maybe the 1880s.
>

I agree that particular work should be cited.  in the paper i cited, i
referred to a number of particular pieces of work in ai and relied on an
article guarino wrote in the 1998 FOIS book that had MANY citations in it on
ontologies.
I also cited that webster dates ontology to 1721
and at least husserl used the term in the late 1800's with a citable
reference i had from 1900.

>
> >  >
> >>    One of the problems with using ordinary XML is that the
> >>    elements and attributes defined by DTDs or XML Schemas do
> >>    not have any semantics associated with them;
> >>
> >>  I think a lot of folks in our audience see the lack
> >>  of semantics in XML as a feature, not a problem.
> >>  Even myself: I don't see a lack of semantics in XML
> >>  as a problem with XML, any more than the lack
> >>  of semantics in s-expressions or binary trees
> >>  is a problem.
>
> It's an opportunity for developers, but a problem for content
> providers. Who do we think we are talking to here?
>
> Pat
> --
> ---------------------------------------------------------------------
> IHMC                                    (850)434 8903   home
> 40 South Alcaniz St.                    (850)202 4416   office
> Pensacola,  FL 32501                    (850)202 4440   fax
> phayes@ai.uwf.edu
> http://www.coginst.uwf.edu/~phayes

--
 Deborah L. McGuinness
 Knowledge Systems Laboratory
 Gates Computer Science Building, 2A Room 241
 Stanford University, Stanford, CA 94305-9020
 email: dlm@ksl.stanford.edu
 URL: http://ksl.stanford.edu/people/dlm
 (voice) 650 723 9770    (stanford fax) 650 725 5850   (computer fax)  801
705 0941

Received on Monday, 18 February 2002 15:51:35 UTC