W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > October to December 2010

Re: Proposed change to the OWL-2 Direct Semantics entailment regime

From: Bijan Parsia <bparsia@cs.man.ac.uk>
Date: Mon, 6 Dec 2010 17:55:13 +0000
Cc: public-rdf-dawg@w3.org
Message-Id: <46682DF7-3130-4EB6-AAF7-2102C6428E62@cs.man.ac.uk>
To: Guido Vetere <gvetere@it.ibm.com>
Hi Guido,

Thanks for writing!

On 6 Dec 2010, at 17:00, Guido Vetere wrote:

> Bijan Parsia <bparsia@cs.man.ac.uk> 
> > [...] 
> > I'm still interested in talking with such people regardless of the group outcome. 
> Hi everybody, 
> Enrico told me that you might want to hear from implementers about the distinguished\non-distinguished variables debate. 
> I've not been able to go through the entire thread and to thoughtfully consider the entire question. But I can tell you that we are implementing an OWL2-QL+UNA engine here at my center, mainly for data integration purposes (please refer to Cangialosi et al. Accessing Data Through Ontologies with ONDA, Proc. CASCON 2010 for a preliminary exposition). 


> Actually, we do support conjunctive queries with non-distinguished variables. Also, we did some practical experience with users, including Banca d'Italia (the Italian Central Bank) by integrating legacy databases, and I can tell you that, in general, customers find conjunctive queries a pretty nice way for accessing data through ontologies. 

So the key point at issue is whether having non-distinguished variables made a difference. We all believe, I'm sure, that conjunctive query rocks (hence our participation). The point at issue really is this particular feature.

> We don't use SPARQL as a query language (we adopt a datalog-style syntax instead) but we might support (some) SPARQL as a front-end in the future, as long as it does not misses relevant features. I cannot tell if making all variable distinguished would definitely prevent form covering some relevant use case.

That's unfortunate. It would be really helpful to find some cases where users would *notice* the difference. Best is when they would rely on non-distinguished variables.

I would be interested in exploring this, perhaps off line, as a research question, by the by.

> However, I don't see a compelling reason to inhibit non-distinguished variables,

I hope I don't give offense by asking the following clarificatory question: Do you really mean variables which range over unnamed individuals, or do you just mean variables which are projected away (in the Datalog world, these coincide as there are no unnamed individuals; hence my question)?

> provided that supporting this feature would be up to implementers. After all, not all SPARQL features are going be supported by all implementers, I guess. 

The problem is that nondistinguished variables get prohibitively harder as you raise the expressiviety of the logic. Ideally, we would like to make it easy for a user to port a query from an RDF engine to a OWL QL engine to an OWL DL engine and get compatible results (if the engines all support all of SPARQL). Nondistinguished varibles makes that impossible.

On the flip side, would you find it extremely burdensome to add nondistinguished variables as an extension? I see you already depart from the OWL spec by imposing the UNA, would this departure discourage you from implementing SPARQL at all?

Thanks very much for your time.

Received on Monday, 6 December 2010 17:55:33 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:01:02 UTC