Re: Layering document

At 12:38 AM -0500 9/27/02, pat hayes wrote:
>[snip]
>Pat
>
>PS. the document now begins by defining OWL/RDF to be what was 
>called "OWL encoded in fast OWL", ie sticks exactly to the mapping 
>from the abstract syntax, and develops the whole semantics and makes 
>the main layering point purely in that context. Then either you can 
>stop reading, or go on to where it (will) discuss how to 'recognize' 
>OWL/RDF in a wider RDF context, and what happens when you use the 
>semantic conditions on fragments of OWL/RDF that are not 
>abstract-syntax well-formed (hence my earlier query about isolated 
>restriction property triples, BTW) . That corresponds roughly to 
>"fast-OWL". Then it explores further, and looks at what happens if 
>you relax the OWL/RDF restrictions to allow the OWL vocabulary to be 
>used on more, er, relaxed RDF graphs. That goes into what were "weak 
>OWL" and "large OWL" now re-christened and lumped together as 
>varieties of 'wild' OWL/RDF. This way, all the really wild stuff 
>about Russell contradictions and such matters is placed right at the 
>end.
>
>I hope this will be more acceptable to everyone.


Pat, I'm sorry, but it is not clear to me this reflects the consensus 
I heard emerging on the call. I  thought I heard consensus emerging 
that "large owl" was too extreme for the group at this time.  I also 
believe I heard consensus arising around the idea of having OWL/RDF 
be what you were calling weak owl and OWl/RDF restricted to the 
abstract syntax being haled as a way to get the extra benefits of 
consistency and speed.   I'm troubled that continuing to work large 
owl back in is not within the group's consensus, nor is the limiting 
of the "normative" OWL/RDF to Fast Owl (unless it can be shown that 
Fast Owl can meet all of our requirements - a point on which I've not 
yet heard anything close to consensus).

I appreciate the hard work you're putting into this, and the 
difficulty of dancing to so many masters,
  Jim H.

-- 
Professor James Hendler				  hendler@cs.umd.edu
Director, Semantic Web and Agent Technologies	  301-405-2696
Maryland Information and Network Dynamics Lab.	  301-405-6707 (Fax)
Univ of Maryland, College Park, MD 20742	  240-731-3822 (Cell)
http://www.cs.umd.edu/users/hendler

Received on Friday, 27 September 2002 07:11:42 UTC