RE: Updated on the Features and Rationale document



> -----Original Message-----
> From: Alexandre Passant [mailto:alexandre.passant@deri.org]
> Sent: 29 September 2009 09:47
> To: Seaborne, Andy
> Cc: SPARQL Working Group
> Subject: Re: Updated on the Features and Rationale document
> 
> 
> On 29 Sep 2009, at 09:06, Seaborne, Andy wrote:
> 
> >
> >
> >> -----Original Message-----
> >> From: public-rdf-dawg-request@w3.org [mailto:public-rdf-dawg-
> >> request@w3.org] On Behalf Of Alexandre Passant
> >> Sent: 29 September 2009 08:16
> >> To: Gregory Williams
> >> Cc: public-rdf-dawg@w3.org Group
> >> Subject: Re: Updated on the Features and Rationale document
> >>
> >> Hi,
> >>
> >> On 28 Sep 2009, at 23:44, Gregory Williams wrote:
> >>
> >>> On Sep 28, 2009, at 5:52 PM, Alexandre Passant wrote:
> >>>
> >>>> Input would be appreciated on:
> >>>>
> >>>> - "Basic Federated Query": any idea of implementations besides
> >>>> DARQ ?
> >>>
> >>> I don't believe DARQ would qualify as "basic federated query" (it
> >>> goes beyond the "basic" part, imo). Wasn't this feature suggested as
> >>> a formalization of Andy's SERVICE keyword in ARQ?
> >>
> >> You're right, it seems to be from the wiki page that this is what was
> >> agreed.
> >> Can one of the champion of that feature confirm ?
> >
> > From my POV, getting anything more than basic connectivity for this
> > round of standardization looks too ambitious.  The minimum is the
> > ability of one processor to be able to call another.
> >
> > http://jena.sourceforge.net/ARQ/service.html

> > Added to F&R.
> >
> 
> Thanks, I also added links to the other implementations supporting it.
> Greg, do you want to add some more text about the RDF::Query
> capabilities ?
> 
> There are examples of queries using such federation on the wiki page
> [1].
> Which system do they come from, is that using ARQ syntax (FROM
> SERVICE) ?
> 
> > This would be an optional feature.
> 
> You mean time-permitting ?

Optional (and time-permitting).

I don't think requiring or expecting every SPARQL implementation provide SERVICE or whatever it becomes is a good idea.  It's a feature it might have and if it does, it should happen in a certain way.

Concretely: in the conformance tests, I don't think we should say "not compliant to SPARQL" just because there is no basic federated query support.

 Andy

> 
> >
> >  Andy
> >
> > When the HTML was copied to the wiki, all the section numbers were
> > copied as well.  And then mediawiki adds section numbers ...
> 
> Indeed, that's a bit of a mess on the wiki :-/
> But I'll take care of that when merging with the current F&R doc in SVN.
> 
> Best,
> 
> Alex.
> 
> [1] http://www.w3.org/2009/sparql/wiki/Feature:BasicFederatedQuery

> 
> >
> >>
> >>> If so, I also support it in RDF::Query (as listed on the original
> >>> feature page on the wiki).
> >>
> >> In that case, I'll indeed use the list of tools listed on the
> >> original
> >> feature page.
> >>
> >> Thanks,
> >>
> >> Alex.
> >>
> >>>
> >>> .greg
> >>>
> >>
> >> --
> >> Dr. Alexandre Passant
> >> Digital Enterprise Research Institute
> >> National University of Ireland, Galway
> >> :me owl:sameAs <http://apassant.net/alex> .
> >>
> >>
> >>
> >>
> >>
> >>
> >
> 
> --
> Dr. Alexandre Passant
> Digital Enterprise Research Institute
> National University of Ireland, Galway
> :me owl:sameAs <http://apassant.net/alex> .
> 
> 
> 
> 
>

Received on Tuesday, 29 September 2009 09:32:13 UTC