Re: SPARQL Aggregation Consensus?

Adrian Walker wrote:
> Hi All --
> 
> Is there a consensus emerging please on the syntax and semantics* of 
> SPARQL aggregation over RDF?
> 
> If I understand correctly, there are some implementations doing their 
> own things, but as yet no standard.

This is correct. We're hoping that this list will be a reasonable forum 
to encourage implementors to be aware of each other's progress on areas 
including aggregation that are not addressed in the current SPARQL 
specification.

> If that's indeed the case, is there a timeline for a standard?

Not at this time. The working group is in the finishing stages of 
publishing a recommendation for the "first version" of SPARQL. It's my 
hope that the work on SPARQL 'extensions' will continue in the various 
SPARQL implementations, and that a new round of standardization will 
begin once there is renewed energy and mature designs to draw from.

Lee

> Surely, SPARQL should avoid the multi-vendor situation that  SQL got in 
> to on this matter?
> 
> Thanks for educating me about this,    -- Adrian
> 
> * Semantics as in, ideally, a model theory that says what the result of 
> any aggregation must be.                 
> 
> Internet Business Logic
> A Wiki and SOA Endpoint for Executable Open Vocabulary English
> Online at www.reengineeringllc.com <http://www.reengineeringllc.com>    
> Shared use is free