Saturday, November 19, 2005
Center of Excellence Proposal
à
Paul Prueitt’s comment about
the use of the
Definition an ontology is a set of concepts
I'd just like to clarify some of the points.
I agree with the sentiments expressed by Barry Smith:
There are no concepts in
BFO. (Ontology is not psychology
or linguistics.) BFO is a
top-level ontology, with 38 types
(universals, categories, kinds, classes)...
To be precise, ontology is a branch of philosophy. Applied ontology is the application of ontological theories in some other field, such as linguistics, artificial intelligence, or library science.
When used with the indefinite article, *an ontology* is an expression of the definitions of some ontological categories or types in some language, which ideally should be some version of logic. When expressed in language X, one might adopt the terminology used for the syntactic categories of language X, such as "predicate" in predicate calculus when a predicate is used to define an ontological type.
When I use conceptual graphs as the version of logic for expressing an ontology, I use the word "type", which is standard terminology in conceptual graphs. But I often qualify that term as "concept type" or "conceptual relation type" when I refer to the syntactic categories of conceptual graphs.
I would not use the word "concept", by itself, to refer to a type or category of an ontology.
John Sowa