W3C home > Mailing lists > Public > public-owl-wg@w3.org > May 2008

functional style syntax

From: Alan Ruttenberg <alanruttenberg@gmail.com>
Date: Wed, 28 May 2008 12:24:15 -0400
Message-Id: <8C3327CA-BBB1-45FE-A430-8F38AE3211F9@gmail.com>
To: OWL Working Group WG <public-owl-wg@w3.org>

At the risk of ticking off the less RDF inclined in the group (for  
which I sincerely apologize in advance) but in the interest of  
perhaps making OWL more accessible to the wider SW audience familiar  
with RDF, I wonder if we should consider some of the choices made in  
the functional syntax.

The sort of thing I have in mind are, for example:

Currently: PropertyAssertion( a:parentOf a:Peter a:Chris ) => a:Peter  
is a parent of a:Chris.

First, the order of argument would be more intuitive if it followed  
the S P O pattern that the triple takes, and it might even be worth  
calling this

Statement(a:Peter a:parentOf a:Chris) to maximize the parallel

Similarly, we have now

ClassAssertion( a:Person a:Peter )

This differs both from OWL 1, and from the common ordering. This  
could become

InstanceOf(a:Peter a:Person )

or

TypeAssertion(a:Peter a:Person)

There are other possibilities, but I won't include a complete review  
here. Rather I wanted to open up the subject to discussion.

I realize that one gets attached to names and that these changes  
might be viewed as unpleasant for some in the working group. So  
first, let me reassure that it is *not* my intention to offend.  
Rather I am hoping that despite an initial potential distaste WG  
members might consider this from the point of view of what might help  
get the broadest adoption of OWL.

Regards,
Alan
Received on Wednesday, 28 May 2008 16:31:13 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 28 May 2008 16:31:15 GMT