Barry, (01)
For problem-oriented special cases, there are good reasons
for using approximations that are known to be physically
false. But in a neutral core that is designed to support
*all* possible applications, it would be a serious mistake
to include any axiom that is known to be false. (02)
> The problem is that as soon as anyone puts forward
> anything to rise to the top, or to be part of the
> supremum, or whatever we call it, you will shoot it
> down by pointing that it is not consistent with, say,
> Quantum Mechnics. (03)
A good way to handle such issues is to adopt the proposal
for "hubs", which are fairly large packages of theories
that are tailored for various application domains. (04)
> I also submit that, given the need to acquire users of
> whatever results from our work, most of whom will not be
> specialists in ontology, the need to be consistent with
> a four-dimensionalist ontology... (05)
Indeed. Each hub should be designed to classify the subject
matter according to the manner in which specialists in that
field are trained to think and talk. (06)
For example, we could have a hub for automotive engineering,
which includes Newtonian mechanics and many other special
cases, which are known to be false under extreme conditions,
but which are also known to be accurate within the limits
of experimental error for ordinary cars and their occupants. (07)
Yet even for automobiles, there are components that require
theories outside the hub: electromagnetic theory for the
radio; relativistic effects on the GPS navigational system;
and quantum mechanics for designing new kinds of paint or
for studying the molecular composition of the fuel and its
combustion inside an engine. (08)
Similar considerations hold for a medical hub: a surgeon
working with organs visible to the naked eye or with low-power
magnification doesn't need relativity or quantum mechanics.
But those considerations become crucial for molecular biology,
pharmaceuticals, and MRI scanners. (09)
As these examples show, even mid-world things may have components
that are crucially dependent on the most advanced physics. If
we divide the ontology into hubs for various purposes, we must
recognize that many components in one hub will depend on theories
that are handled in detail by other hubs. In one hub, those
components will behave like black boxes with warning labels that
say "No user serviceable parts inside"; in the other hub, however,
those black boxes will be transparent "white boxes", whose inner
workings are central to the subject matter of that hub. (010)
Preserving the mid-world phenomena and speech patterns while
satisfying those constraints should not be difficult: (011)
> ... the need to be consistent with a four-dimensionalist ontology
> according to which there are no dogs but only doggy processes,
> is not a sensible constraint. (012)
Matthew's ontology handles 4 dimensions quite nicely while saving
the phenomena and terminology of ordinary speech. Similarly, the
Cyc ontology and Whitehead's ontology accommodate the view of dogs
as processes without requiring anyone to modify speech patterns
or feeding habits when playing with their pets. (013)
Summary: In order to accommodate *all* hubs, the core must not
contain any axioms known to be physically false. But there can be
as many hubs as necessary, each designed from the perspective of
people who work in a particular specialty. Since every industry,
ranging from medicine, to automobiles, to aerospace requires
multiple specialists, each hub may need components that are
treated as "black boxes" in that hub, but as "white boxes" in
some other hub. (014)
John (015)
_________________________________________________________________
Message Archives: http://colab.cim3.net/forum/ontac-forum/
To Post: mailto:ontac-forum@xxxxxxxxxxxxxx
Subscribe/Unsubscribe/Config:
http://colab.cim3.net/mailman/listinfo/ontac-forum/
Shared Files: http://colab.cim3.net/file/work/SICoP/ontac/
Community Wiki:
http://colab.cim3.net/cgi-bin/wiki.pl?SICoP/OntologyTaxonomyCoordinatingWG (016)
|