Hi Pat, (01)
>
> Yes so far, but ...
>
> > Metaclasses - predicates that take predicate class predicates=20
>
> I have never understood what the intended meaning of "metaclass" was.
> After learning mathematical set theory, it is hard to imagine what
> could be more "meta" than a set. (02)
Agreed. I believe that the term 'metaclass' comes from programming languages
and refers to a second (maybe higher)-order classes. Unlike classical set
theory which has classes all the way down, these start with non-class
ur-elements and then classes of the ur-elements. In this structure, classes
of first-order classes are meta-classes. (03)
> > However, if you mean the distinction =
> > between
> > particularity and generality, then I am not so sure. Surely it is =
> > natural
> > (though maybe na=EFve) to think what distinguishes a particular,
> > such as =
> > Pat
> > Hayes, from a class of things, such as the class of Welshmen, is
> > that =
> > Pat
> > Hayes cannot have members, whereas the class of Welshmen can and
> does.
>
> It is natural at first blush, but it gets quite hard to keep it up for
> an extended length of time. For example, one way to reconcile the
> temporal fights is to introduce the notion of a series of time-
> snapshots of a 4D entity, this being the nearest thing in the 4D world
> to a 3D continuant. If you do that to (the 4-D) Pat Hayes, then I
> become the set of my instantaneous snapshots. (04)
I have the impression that it is only enemies of 4D that propose the set
option (I am not suggesting you are an enemy), this seems to me to avoid 4D
altogether as all 4D objects become sets of 3D objects. Friends of 4D seem
to go for the mereological fusion option - then you get the same 4D Pat
Hayes. It seems to me that the choice gives you different structures. A part
of a snapshot of Pat Hayes is also a part of the mereological fusion but not
of the set, etc. (05)
> Now, I am quite happy to
> be thought of this way, and it is sometimes very useful. But if we
> have a logically rigid distinction between things with members and
> things without, then this violates a fundamental partitioning of the
> universe. LIke the continuant/occurrent distinction, this dichotomy
> seems natural but in fact just gets in the way when one gets down to
> serious ontology engineering. One of the great merits of the CL
> absolute type freedom is that it imposes no a priori logical obstacles
> to such re-conceptualizations of entities. (06)
I suppose it depends what you are doing. If it is extremely useful not to
bother the user with questions about whether one is talking about a set or a
non-set individual, then CL seems the way to go. If however, it is more
important that the model one builds is accurate, etc. then getting the
analyst to do some thinking work may be an inevitable by-product. (07)
Regards,
Chris (08)
_________________________________________________________________
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
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (09)
|