Re: OWL DB, OWL UML, etc.

• From: Bob MacGregor <macgregor@ISI.EDU>
• Date: Thu, 13 May 2004 09:12:53 -0700
• To: Jim Hendler <hendler@cs.umd.edu>, www-rdf-logic@w3.org
• Message-Id: <6.0.3.0.2.20040513090810.01f2dec0@tnt.isi.edu>

Its not at all clear to me that OWL subsets are adequate to capture
the primitives needed for something like database integration.
In our own work, we find it essential to be able to define compound
keys.   We also find it essential to be able to define compositions
of (range-restricted) properties.  I'm not sure if the DB integration
folks need the latter or not, but I would be very surprised if they
don't need the former.  Both of these constructs lie outside of
OWL Full, as I interpret it.

Cheers, Bob

Received on Thursday, 13 May 2004 12:19:48 UTC