Digital computers don't do nuances.
That means anything that is designed to be implemented on a computer
must be precisely defined. But the methods by which people design
their implementations are open to all the same kinds of informal
intuitions and practices that scientists use to get new ideas.
I would qualify Chris P's
> I find amusing that the Gruber CS-AI sense seems to go out of its
> way not to talk about the things that the philosophy sense does.
Tom Gruber's definition is Tom's own personal view. The term
'conceptualization' comes from the book _Logical Foundations of AI_
by Genesereth & Nilsson, which is more nuanced. But there are as
many CS or AI views as there are people working in CS or AI.
I agree with Chris M:
> What, exactly, is the "expressing" relation between
> a representation and an ontology so understood?
> It seems to me that there are no scientifically
> rigorous answers to these questions.
I don't believe that it's possible to state a precise definition
of what an ontology is or should be. It's better to use
Wittgenstein's method for defining 'game':
Give a bunch of examples and say "These are ontologies.
An ontology is
anything that is more similar to these things
than to anything that is not usually called an ontology."
I wouldn't say that such a definition is "unscientific".
It allows progress, but it doesn't rule out critiques about
some practices that may be better or worse than others.
That is, in fact, how mathematicians define 'number'. They
started with positive integers. Then they added rational
numbers, irrational numbers, zero, negative numbers,
imaginary numbers, complex numbers, quaternions, numbers
modulo some integer, floating point numbers, fixed point
numbers with various criteria for what happens to overflows...
Just as the concept of number grew as a result of the way
that working mathematicians talked about their work, the
concept of an ontology is going to grow. Any attempt to
edict a definition in terms of necessary and sufficient
conditions would be counterproductive --
it would needlessly
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
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