W3C home > Mailing lists > Public > public-webont-comments@w3.org > October 2002

OWL-Lite comments

From: Aseem Das <aseem.das@blackpearl.com>
Date: Wed, 2 Oct 2002 13:16:26 -0400 (EDT)
Message-ID: <49FCDC8730A25C459F818F417BEC2256049C4E@bp01.blackpearl.com>
To: public-webont-comments@w3.org

I currently work on ontology tools and solutions at BlackPearl and worked on
them previously at VerticalNet and recently reviewed the OWL feature
synopsis document.  In my current work and also previous work at
VerticalNet, I find it very useful that international organizations like w3c
are working to produce an ontology language like OWL-Lite. I find work on a
language such as OWL Lite to be the right strategy for acceptance by
application developers and tool builders. It is a simpler language to
understand and implement and from my experience sufficient for most business
commercial requirements. If OWL Lite had been available when we started the
VerticalNet Ontobuilder/OntoServer we would have implemented it first and
then incorporated the more sophisticated features of OWL as business demands
required it.  We are working with the same approach at BlackPearl currently.
Some specific comments: 
In the ontology tools developed at VerticalNet, we restricted the names of
individuals & classes to be unique within an ontology and that did not give
us any problems. The "differentIndividualFrom" feature will give rise to a
large number of declarations, depending on the number of individuals in the
ontology (N combinatorial 2 - where N is the number of individuals which are
unique). We also used full cardinalities quite a bit and restricting them to
be just 0 and 1 in OWL-Lite may not be sufficient for most business reqs. We
also used "hasValue", often to give default values to properties either upon
definition or upon attachment.
Aseem Das 
Received on Thursday, 3 October 2002 03:40:16 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:09:28 UTC