W3C home > Mailing lists > Public > public-rif-wg@w3.org > January 2006

Re: [SWC] Re: RIF and QL

From: Peter F. Patel-Schneider <pfps@inf.unibz.it>
Date: Fri, 27 Jan 2006 08:36:34 -0500 (EST)
Message-Id: <20060127.083634.03291396.pfps@research.bell-labs.com>
To: bry@ifi.lmu.de
Cc: public-rif-wg@w3.org

From: Francois Bry <bry@ifi.lmu.de>
Subject: [SWC] Re: RIF and QL
Date: Wed, 25 Jan 2006 15:55:39 +0100

> [summary: 
> Resent with tag. 
> SPARQL queries and queries in other query languages in RIF rules.]
> 
> Dear RIFers,
> 
> During yesterday's (24 Jan.) telecon, the issue has been raised
> whether SPARQL queries, or queries expressed other query languages,
> could appear in bodies (= condition parts) or heads (= consequence
> parts) of RIF rules.
> 
> It has been noticed that this approach would have the advantage of
> elegantly ensuring the compatibility of RIF with SPARQL and the other
> query languages considered.

Hmm.  I don't go along with this at all, depending on the query language.  What
happens, for example, if the underlying representation language has
disjunction, or other non-vivid representational constructs?  [Here there are
problems with producing a complete account of reasoning.]  What happens if the
query language has aggregation?  [Here there are problems with providing a
comprehensive account of meaning, particularly with respect to queries in the
head, but also just for queries in the body.]  What happens if the query is to
the same source of information that the rules are affecting?  [Here there are
problems having to do with termination.]

[...]

> Francois

Peter F. Patel-Schneider
Received on Friday, 27 January 2006 13:36:50 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:33:26 GMT