Chris, (01)
Description logics have traditionally been used as the T-box for
defining the terminology used by the more expressive logic in the
A-box. To maximize the convenience of using any DL (including OWL)
with other logics (and procedural languages), it's important to
define the term 'class' in a way that facilitates interoperability. (02)
CP
> So a Sowa-class can change extension "at different times, places,
> and possible (or actual) worlds."
>
> But some Sowa-classes do not change extension and some do - agreed? (03)
Yes. (04)
CP
> It sounds as if you are agreeing with me. (05)
Yes, on this point. The more general point is that you can define
classes in terms of relations, but you can't define relations in
terms of classes. (06)
Definition: A *class* C has a monadic *defining relation* P.
The *extension* of C consists of all x for which P(x) is true. (07)
This definition makes the term 'class' a derivative notion whose
properties and extension are determined by P. That provides maximum
generality: you can freely use any DL ontology in any inference
engine for any subset or superset of FOL. (08)
The key word is *interoperability*. Just look at the much, much
wider suite of logics in Tim B-L's proposal of 2000 compared to
the drastically narrowed collection of tools available today. (09)
After 12 years of development, I wouldn't call that success. (010)
John (011)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (012)
|