[topicmapmail] basics II - Inheritance in Topic Maps

[Robert Barta]

On Tue, Jun 03, 2003 at 12:07:15PM +0100, Murray Altheim wrote:
> Robert Barta wrote:
> > Unfortunately, we do not have an official TM ontology language yet.

> I have been enjoying your exchange on this subject and don't mean to
> diminish the importance of it, but I'm *glad* we don't have an official
> TM ontology language.....
> one religious school among many. Now, if there was some way of creating
> a TM-based meta-ontology language that was agnostic as to the model theory,
> I'd be in favour and would probably be working on it right now. But I'm
> doubtful. This doesn't mean I'm not interested in an ontology language,
> but I don't think of "official" ontologies as anything but coercive in
> the end.

Murray,

As strange as it may sound, I can share you sentiments regarding an
"official" language. In fact, for relational models there are a few,
albeit quite similar schema languages, for XML you have a zoo - and
that will remain, it seems to me. There is no indication that for
ontologies this will be different.

The reason I see in favor of standardized languages is that the TM
paradigm without an ontology definition language (and without a query
language) seems incomplete to me: For all the applications I have
written I had this "wouldn't it be nice" feeling.

> I'd be keen on hearing your thoughts on what might constitute some
> sort of meta-language that *might* be relatively agnostic.

I assume that with 'meta-ontology language' you mean a language which
allows to define ontologies. Yes, this is what AsTMa! (and others)
tries to do.

I am not sure whether this can be defined _without_ any underlying
model. What we try here is to use a model which is rather close to
Steve N's reference model, although streamlined a bit. So I plan to
define the language semantics fully in terms of a Topic Map model and
not some other model theory. This is as agnostic as my imagination
goes.

\rho