W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > July to September 2012

Re: Comment BB-1 draft response (was RE: SPARQL TC 2012-09-04 Agenda)

From: Andy Seaborne <andy.seaborne@epimorphics.com>
Date: Tue, 04 Sep 2012 10:45:45 +0100
Message-ID: <5045CDC9.2040808@epimorphics.com>
To: public-rdf-dawg@w3.org


On 04/09/12 08:38, Polleres, Axel wrote:
>
> The comment BB-1 (Querying only the default graph from the data store)
> Has now a draft response at:
>   http://www.w3.org/2009/sparql/wiki/CommentResponse:BB-1
>
> Please support/comment,
> Thanks,
>
> Axel
>

A couple of points:

1/ The response discusses the comment - and asks some implicit questions 
e.g. "it might be the case that you suggest here" seems to asking for 
further discussion.  Is that the intent?

2/ Service description is important here in describing what data is in 
an endpoint.  It would point to the VOiD defn of the default graph.

If we're happy with that, add text to refer to the service description work.

3/ The portable way to query only the default graph is to not use GRAPH 
in the query.  I think the confusion here is that someone coming from a 
world where the default graph is the union of named graphs.

I suggest stressing this by changing you example to

"""
SELECT *
WHERE { ..... }         <-- No use of GRAPH.
"""


	Andy

Just discussion on:

SELECT *
FROM DEFAULT

I don't know that it means! Maybe there is an assumption there is a real 
storage default graph that can be masked for query by the union of named 
graphs.  This seems to be getting into defining what is intentional a 
reflection of variability.

The real, concrete default graph can also be put in the dataset as a 
named graph, so it is accessible via FROM and FROM NAMED. (c.f. Anzo?)
Received on Tuesday, 4 September 2012 09:46:22 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:15:49 GMT