Re: OWL semantics

[...]

> Moreover, I would be interested in your suggestions as to how one
> might go about proving that such an axiomatisation is correct. As you
> refer to engineering, I presume that this means testing. Testing wont
> PROVE anything (there are infinitely many possible inputs), it can
> only show that you didn't find any errors yet. In fact, given that the
> axiomatisation is in FOL, which is undecidable, even inputs that would
> in principal lead to errors may not reveal anything due to
> incompleteness in the FOL reasoner.

of course not only testing, but lots of theoretical work too
over the past 2300 years many smart people have already done
lots of sound axiomatisations; no doubt that that will go on

-- ,
Jos De Roo, AGFA http://www.agfa.com/w3c/jdroo/

Received on Tuesday, 6 August 2002 06:25:47 UTC