W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > May 2002

Re: Layering OWL on RDF: the case for unasserted triples

From: Jim Hendler <hendler@cs.umd.edu>
Date: Thu, 30 May 2002 17:56:23 -0400
Message-Id: <p0511172bb91c4ac233ae@[10.0.1.4]>
To: Brian McBride <bwm@hplb.hpl.hp.com>, "Jonathan Borden" <jonathan@openhealth.org>, <w3c-rdfcore-wg@w3.org>
Cc: "Guus Schreiber" <schreiber@swi.psy.uva.nl>

At 9:33 PM +0100 5/30/02, Brian McBride wrote:
>At 15:57 30/05/2002 -0400, Jonathan Borden wrote:
>[...]
>
>>The OWL model theory itself remains an open issue and different members of
>>the WG have different ideas about how this might be accomplished.
>
>Thanks Jonathan.  That's a comprehensive reply.
>
>I think I heard:
>
>   o Whether or not dark triples will be used by webont is still an 
>open question

Brian - I think it is clear that at this point WOWG will create a 
mechanism to achieve this effect.  Several options are open.  If RDF 
Core creates rdf:unasserted, then we would almost definitely use it 
(it's the only thing where we are close to consensus).  If not, then 
we are not in consensus about which of two ways we might build would 
be less problematic -- but I think it safe to say that either one 
will cause some angst because we will have to either be less 
semantically compatible with RDF Core, have a weaker language on our 
part, or have to solve some problems that right now look like 
research.

I think this is what John means when he says "not reach consensus" or 
do so by attrition -- we are grappling with hard issues that have 
impact on the future of the sem web, so it's not surprising we have 
to work carefully.

>
>   o Never the less, webont are asking RDFCore to consider how dark 
>triples might be provided

yes, we have gotten two kinds of feedback from members of RDF Core - 
one says "Give us the one answer you want worked out in detail, and 
we'll accept or reject"  the other is "Don't tell us how to do our 
job, it's RDF Core that should decide how to provide DTs"  -- clearly 
we cannot do both, but I think  John has come as close as humanly 
possible -- he has outlined an approach we believe would work, but 
has done it without trying to be over-constraining on RDF Core

>Did I hear right?

yes.

Again, we do have OWL only solutions on the table, but they do not 
appear to most of us to be as good as what we could do if RDF 
provides some mechanism.

Let me be clear on one thing - whatever decision RDF Core makes, 
webont will go on to design a successful ontology language that is 
maximally compatible with what you build - our job will be easier if 
one of the approaches works well for you, but it will not be a 
show-stopper either way.

One other thing I'd like to state for the record -- and I invite 
everyone to quote me publicly on this:
  The unasserted triples issue is a hard one, but it is not a 
show-stopper.  The 99% of stuff our groups are in agreement on, and 
building together, is what makes the Semantic Web the unstoppable, 
dynamic project that it is -- what we're ALL doing is very important 
and moving amazingly well.  DTs is a technical issue, with 
ramifications at the margins - but the effort your group has made in 
providing a strong base for the semantic web is incredible, and to be 
much lauded.  We're hoping OWL will be a strong motivator for people 
to use what you are building, and we also appreciate all the help 
your members are providing in helping us to create our language 
correctly.  It is important that the outside world understand that 
while there are some hard architectural issues we are attacking, none 
of them is going to stop the semantic web from happening -- and 
happening sooner than they expect!!
  <Hendler gets down off his soap box>

  -Jim H.
  co-chair WebOnt

-- 
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 Thursday, 30 May 2002 18:20:46 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:48:16 EDT