Key questions on Common Upper Ontology
4/20/2004 6:54
AM
Note from Irene,
I think the vagueness of definitions is in issue here.
Although, not necessarily the vagueness about what an ontology is, but rather
about what a "standard upper ontology" is. I'd also like to
understand better lattice of theories and ideas about multiple modular upper
ontologies mentioned by Leo.
Replacing "ontology" with "program
spec" doesn't do much for me. For several reasons, one of them is that a
"standard upper program spec" is not a term that is widely used or
has a commonly accepted and well understood definition. So we have replaced one
term that needs definition with another term that needs definition.
John suggested that a common upper program spec is "a
specification of those data elements that are common to some fairly large set
of programs. There are many such things
in existence, such as the IEEE floating-point specs and various standards for
times, dates, currency, etc."
Does this definition mean that any ontology designed for
re-use is an upper ontology?
I thought that an "upper" ontology is the one
that you use by subclassing the lowest classes. For example, an upper ontology may say, "there are living
things and there are inanimate things". I would use it by then
specializing and saying that "people are living things", "fish are
living things", "hats are inanimate things". This assumes that I
find the classification of all the things into living and inanimate both reasonable
and useful and every thing I need to model would fit under (and benefit from)
this classification.
Ontologies of dates, measures and so on are rather
different from the example above. They are focused, specialized and get re-used
"directly", without subclassing, mostly by referencing to them. For
example, "A meeting has a start time and end time". I am not going to
define "time", I am just going to take it from the DATE ontology.
Most (if not all) of highly useful and re-usable ontologies I came across were
of this kind.
I believe that it would be confusing to call both of these
very different types of models upper ontologies.
Regards,
Irene