General comments on OWL2 design

The next few messages contain our comments on the current OWL2 design.
We have decided to split our comments across multiple messages to
facilitate discussion in multiple threads.

They are mostly general comments on the design decisions,
not specific comments on the documents. (although we make some editorial
suggestions at the end of this message).

We support the inclusion of all of the new language features F1-F15
in section 2 of the "New Features" doc. We are not entirely convinced
how broad the the demand is from use-cases for all of these features (of
course it's easy to find some use-cases for each feature), and there a
is the price of the additional syntactic complexity of the language due to
these extensions, but in general we're fine with these extensions.

We note however that half of the use-cases come from a single domain
that has all kinds of peculiarities, and for which it is debatable
whether it is representative of major Semantic Web take-up. Looking at
the business space of recent years, the buzz is not in bio-science, but
in instead in many other applications that require much lower
expressivity.

However, even given our doubts on the urgent need for all this
additional expressivity, we support the new language features numbered
F1-F15 from the "New Features" document.

Our next messages will each list an objection or request for change.
Some of our requests in the other messages may seem substantial, but we
think they are all easy to implement, even in this late stage of the WG's
work:
- revisit all uses of the name "OWL2" for usage consistent with
   established practice
- simply declare the non-RDF-XML syntax non-normative
- introduce a simple mechanism for indicating the (minimal) language
   fragment of an ontology.
- remove the new treatment of anonymous individuals.

Frank van Harmelen,
and many members of the Semantic Web Group at the Vrije Universiteit Amsterdam

Received on Friday, 23 January 2009 23:58:16 UTC