- From: Dan Connolly <connolly@w3.org>
- Date: Fri, 09 Sep 2005 18:17:43 -0500
- To: RDF Data Access Working Group <public-rdf-dawg@w3.org>
On Thu, 2005-09-08 at 08:18 -0500, Dan Connolly wrote: > Given recent discussion and comments, I'm inclined to add > three issues to our issues list... > http://www.w3.org/2001/sw/DataAccess/issues > > rdfSemantics -- should queries of equivalent graphs > give the same answers? Any practical advice about queries > over infinite graph such as all the RDF axiomatic triples? > owlDisjunction -- the worker example evidently doesn't work > well with SPARQL as of the 21 July 2005 LCWD. Are there > mature designs that work better? At a minimum, we should > be explicit that we don't handle this. Those two are now open. http://www.w3.org/2001/sw/DataAccess/issues#rdfSemantics http://www.w3.org/2001/sw/DataAccess/issues#owlDisjunction > owlIntegration -- some explicit explanation of how OWL integrates > with SPARQL seems worthwhile. I don't know if that's a separate issue. We'll see how the more concrete ones go. > Is that too many or too few issues to cover recent discussions? I see a request to re-consider SOURCE/GRAPH and UNSAID/OPTIONAL/BOUND in Enrico's comments... "The construct GRAPH, combined with FROM clauses, allows to bind variables/restrict graphs the triples came from. The main problem with this feature is that the RDF Dataset deļ¬nition ... does not have a corresponding semantics in the RDF model theory. ... We reckon that the possibility of referring to graph names enables some interesting features of the query language; however, we believe that this should be excluded from cSPARQL." -- http://www.inf.unibz.it/krdb/w3c/sparql-notes-fub.html But I think the WG already considered that argument in Hensinki... "... DanC argued against taking us out of the scope of positive conjunctive queries against RDF graphs. ..." -- http://www.w3.org/2001/sw/DataAccess/ftf4.html It's possible that our discussion of rdfSemantics will result in adopting a requirement for soundness/completeness semantics, in which case it will be necessary to reconsider the SOURCE and UNSAID issues. But for now, I'm leaving them closed. -- Dan Connolly, W3C http://www.w3.org/People/Connolly/ D3C2 887B 0F92 6005 C541 0875 0F91 96DE 6E52 C29E
Received on Friday, 9 September 2005 23:17:52 UTC