W3C home > Mailing lists > Public > public-rdf-dawg-comments@w3.org > June 2007

RE: SPARQL Semantics

From: Seaborne, Andy <andy.seaborne@hp.com>
Date: Fri, 22 Jun 2007 17:05:32 +0100
Message-ID: <86FE9B2B91ADD04095335314BE6906E8013C6B83@sdcexc04.emea.cpqcorp.net>
To: <public-rdf-dawg-comments@w3.org>, "Michael Schmidt" <m.schmidt00@web.de>



-------- Original Message --------
> From: Michael Schmidt <>
> Date: 11 June 2007 11:48
> 
> Hi,
> 
> 
> apologizes if this is not the right place to post my question.
> 
> In "Semantics and Complexity of SPARQL" by Arenas, Gutierrez and
Perez,
> a compositional and an operational semantics for the evaluation of
> SPARQL have been proposed.  
> A couple of months ago, I posted a question about the semantics of
> nested OPTIONALS to the ARQ mailing list
> (http://tech.groups.yahoo.com/group/jena-dev/message/25717), a topic
> that has been widely discusses in the W3C mailing list. The answer to
> my question was that Jena - as well as the W3C working draft at that
> time (http://www.w3.org/TR/2006/WD-rdf-sparql-query-20061004/),
> followed the operational semantics, which answered my question.      
> 
> I just studied the new W3C SPARQL working draft semantics definition
> and it seems that, in meantime, the semantics changed and now follows
> the compositional approach. In spite of an extensive search in the W3C
> mailing list archive, I could not find any threads stating about this
> decision, but instead only threads containing general discussions and
> differences between both semantics. Have there been any public
> discussions on that topic, in particular why the semantics has been
> changed? And why did you use/propose the operational semantics before?
> I guess the main reason must have been not-commutativity of operator
> AND.         
> 
> Any help would be appreciated,
> Michael
> _______________________________________________________________
> SMS schreiben mit WEB.DE FreeMail - einfach, schnell und
> kostenguenstig. Jetzt gleich testen! http://f.web.de/?mc=021192 

Michael,

Apologies for the delay in replying.

The working group has been discussing this over a period of time, both
via email and  also in the teleconferences (which are publicly minuted).

There have been several independent proposals for a compositional
algebra for SPARQL, both inside and outside the working group.  As
editor, I prepared material that I thought reflected the views of the
members of the working group and wider community in a form that would
bring consensus.  The formal point where the working group accepted the
approach would have been when it decided to publish the Last Call
publication of 26 March 2007.  That is, the working group decision was
based on specific text.

Your message jena-dev/message/25717 was Oct 2006.  Sicne then, ARQ 2.0
has been released which follows the current SPARQL CR and evaluates
using scoped normal form evaluation (it also notices when it can get the
correct results with a more space-efficient strategy).

	Andy

--------------------------------------------
  Hewlett-Packard Limited
  Registered Office: Cain Road, Bracknell, Berks RG12 1HN
  Registered No: 690597 England
Received on Friday, 22 June 2007 16:05:46 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:14:51 GMT