[Person-ontology] Ontology Requirements for Distributed Data Sharing

[Paul Trevithick]

Pat Hayes wrote:

<snip>
> While I agree that "cross-context data sharing" is important, I don't
> believe that the Higgins project's approach to it is either necessary
> or desirable, and I would strongly urge that we do not agree to
> follow their approach. IMO, the Higgins project has placed far too
> much emphasis on this aspect, and the result has warped and distorted
> their use of OWL and XSD, which is highly unconventional. 
<snip>

It is true that HOWL is a "warped and distorted use of OWL and XSD." We got
to this point by (a) having a set of requirements (b) selecting OWL 1.0
because it seemed to be in roughly the right space, looked like it would
become popular, and had some good tool support and then (c)
warping/distorting OWL to make it meet our requirements. Unfortunately, as
has been pointed out, it really only "works" within the narrow Higgins
community of practice and doesn't interoperate with any efforts of other
communities. 

There are two paths forward for Higgins. One is for someone to step forward
and educate us on a non-warped, undistorted way that OWL can be used to meet
the requirements. The other is to abandon OWL and use something more
expressive (e.g. CLIF, IKF, SUMO-KIF etc.) as has been recently suggested on
this list. The former would be preferable, but may not be possible. 

-Paul