Dear John S., (01)
>>>
However, it is integrated with the DOLCE ontology. The topics and
the analyses are sufficiently general that they could be adapted
to other ontologies. But it raises some questions:
<<< (02)
only brief responses at present I'm afraid: (03)
>>> 1. What changes would have to be made to adapt the analysis to
other ontologies, such as OpenCyc, SUMO, BFO, etc. ?
<<< (04)
reasonable axiomatisations of quality spaces in order to provide the
foundations for the rest (05)
>>> 2. Is it possible to generalize the approach so that it could
be formulated in a generic UoM to be used as a microtheory
in any or all of those upper level ontologies?
<<< (06)
if (1) is fulfilled, I expect so. Otherwise, formulating it as a
part of a generic UoM would bring with it a particular instantiation
of the idea of quality spaces in any case. (07)
>>>
3. How could it be adapted to the VIM document?
<<< (08)
same way any other approach would I guess, with a lot of work. (09)
>>> In the thread on Quantity Kinds, Ed Barkmeyer and I discussed
the issues about uncertainty and tolerance that are raised
by the VIM document. I searched Probst's dissertation and
found only one mention of uncertainty and no mention of
tolerance. How can those issues be handled?
<<< (010)
This is one of the more interesting theoretical contributions
of the thesis as a whole, quality spaces become granular
entities that bring with them notions of inherent uncertainty
and tolerance. Because it is quite central, it presumably does not then
appear as a neatly added box. I recommend a
thorough study. (011)
>>> Will any UoM ontology promote better interoperability? Or will it
create more obstacles, problems, and headaches than it can solve?
<<< (012)
well, I'll leave that to you UoMers. There are examples in the
thesis however (as well as other examples from GiS) where
not having a sensible measures ontology leads rather disastrously
to failures in your: (013)
>>> computer systems
have been sharing data and interoperating for over half a century
without detailed ontologies
>>> (014)
sometimes they do, and if one is lucky, nothing goes bang. Sometimes
however they don't and the results can be nasty. So arguments
against developing ontology on the basis that all is fine
and dandy are rather easy targets. Whether getting things better
requires all the machinery of ontologies that we all know and love
is a slightly different issue though. (015)
Best,
John B. (016)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/uom-ontology-std/
Subscribe: mailto:uom-ontology-std-join@xxxxxxxxxxxxxxxx
Config/Unsubscribe: http://ontolog.cim3.net/mailman/listinfo/uom-ontology-std/
Shared Files: http://ontolog.cim3.net/file/work/UoM/
Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?UoM_Ontology_Standard (017)
|