- From: james anderson <james@dydra.com>
- Date: Wed, 1 Aug 2012 10:00:56 +0200
- To: SPARQL Comments <public-rdf-dawg-comments@w3.org>
good morning; On 2012-08-01, at 08:25 , Polleres, Axel wrote: > Dear David, > >> Thank you. I am satisfied with this resolution, providing >> that "virtual graphs" is added to the wish list for >> consideration in the next version of SPARQL: >> http://www.w3.org/2009/sparql/wiki/Future_Work_Items > > In order to not add to many new items to this list, please > note that this page already has an explicit link to all the discussed > features in the beginning of this WG and that were not adopted for > work by the SPARQL 1.1 Working Group: > http://www.w3.org/2009/sparql/wiki/Category:Features > > It seems to me that this your proposal of "vitrual graphs" is > covered in the sense that > it is a variant of what we had alredy noted under the feature name > "Composite Datasets": http://www.w3.org/2009/sparql/wiki/ > Feature:CompositeDatasets if the "Composite Datasets" feature is, or should eventually become, the subject of active discussion, it would be worthwhile for the proponents to explain, 1. how a mechanism like "COMPOSE GRAPH" permits more than enumerating the respective arguments in a "FROM" or "FROM NAMED" clause 2. how the "cohort" use case, which the document claims to be the primary motivation, is not supported by graph variables in subqueries. 3. why the "from all named graphs" use case motivates either an addition syntactic element (FROM *) or the equivalent "COMPOSE GRAPH" long form, when a standard designator - an iri specified as part of the SPARQL standard, would be sufficient. best regards, from berlin --- james anderson | james@datagraph.org | james@dydra.com | http:// dydra.com
Received on Friday, 3 August 2012 15:31:15 UTC