RDF Data Access Weekly -- 20 Mar 2007

<LeeF> Scribe: LeeF

<patH> be there in a second

<SteveH> hi

minutes from last week -> http://www.w3.org/2007/03/13-dawg-minutes

approved minutes http://www.w3.org/2007/03/13-dawg-minutes

next meeting 27th of march, scribe: ericP

action items

<scribe> ACTION: ericP to draft text about a LOOSE keyword and run it by w3 folks [recorded in http://www.w3.org/2007/03/20-dawg-minutes.html#action01]

to see if we're abusing the "at risk" mechanism [DONE]

<scribe> ACTION: LeeF to seek guidance about at-risk features from the CG [DONE] [recorded in http://www.w3.org/2007/03/20-dawg-minutes.html#action02]

<scribe> ACTION: ericP to mark sections 2 and 3 informative, Appendices A, B and D [recorded in http://www.w3.org/2007/03/20-dawg-minutes.html#action03]

normative in the text and table of contents and 1.1 document outline [DONE]

<scribe> ACTION: LeeF to close bnodeRef issue [DONE] [recorded in http://www.w3.org/2007/03/20-dawg-minutes.html#action04]

<scribe> ACTION: LeeF to close nested optionals issue [DONE] [recorded in http://www.w3.org/2007/03/20-dawg-minutes.html#action05]

<scribe> ACTION: LeeF to close openWorldValueTesting issue [DONE] [recorded in http://www.w3.org/2007/03/20-dawg-minutes.html#action06]

<scribe> ACTION: PatH to investigate closing the entailment issue [DONE] [recorded in http://www.w3.org/2007/03/20-dawg-minutes.html#action07]

<scribe> ACTION: EricP to run the yacker tool over and annotate the existing tests [CONTINUES] [recorded in http://www.w3.org/2007/03/20-dawg-minutes.html#action08]

<scribe> ACTION: LeeF or EliasT to reply to Bjoern regarding (not) POSTing application/sparql-query documents [CONTINUES] [recorded in http://www.w3.org/2007/03/20-dawg-minutes.html#action09]

<scribe> ACTION: LeeF to remember that the wee, lost filter tests should be put [CONTINUES] [recorded in http://www.w3.org/2007/03/20-dawg-minutes.html#action10]

Duplicate solution cardinality

<EliasT> LeeF: I spoke to Ivan and Ralph and they agreed that we can make use of the at risk feature in our publication and mark LOOSE at risk.

<scribe> Scribe: EliasT

UNKNOWN_SPEAKER: we can also make use of an informative section for LOOSE but that might not neccessarily fit this situation.

ericP: LOOSE keyword shouldn't affect implementations too much since they only need to know how to parse it. They don't have to do any extra work when encountering it.

patH: I'll keep my mute button on for this discussion.

<LeeF> PROPOSE: SPARQL SELECT queries with no keyword following SELECT must

<LeeF> return the precise cardinality of duplicate solutions specified by the

<LeeF> algebra; SPARQL contains a @@ LOOSE keyword that allows duplicate

<LeeF> solutions to be returned with cardinality of at least 1 and no greater

<LeeF> than that specified by the algebra. The @@ LOOSE keyword will be marked as

<LeeF> an at-risk feature of SPARQL.

<ericP> i nominate "LAX" as the keyword

<sdas2> DISTINCT PLUS

AndyS: We should discuss the keyword because that will dictate the semantics

<LeeF> LOOSE

<LeeF> LAX

<LeeF> INDISTINCT

<LeeF> DISTINCT PLUS

<patH> how about VOMIT

<LeeF> "SOME" ?

<Souri> Thanks Jeen

<patH> 'indistinct' sounds like you can't tell the difference betwen the answers.

<AndyS> -1 to LAX : 0 to LOOSE

<patH> LAX is an airport code, which might be a problem?

<LeeF> AT-LEAST-ONE-OF-EACH

WHATEVER?

<AndyS> REDUCED?

<patH> SILLY?

<Souri> +q

<patH> MORE (seriously)

<SteveH> what about LOOSE DISTINCT

<SimonR> It's a little bit disturbing that there's no english word for this concept....

<SimonR> BAG

<LeeF> PARTIAL ? (same problem as SOME)

<ericP> from www.wordsmyth.net:

<ericP> lax[1]: not rigorous, strict, or careful; loose; lenient.

<ericP> loose[1]: not restrained or confined; free.

<ericP> reduce[1]:to make less in amount or size.

<SteveH> I like REDUCE best I think

<patH> how about FREE?

<SimonR> DUPLICATED

<Souri> REDUCED DUPLICATES ?

LeeF: AndyS and Souri have a good comment that PARTIAL and SOME might imply that you are getting less than at least one for each unique answer.

ericP: (answering to Lee's what about REDUCED) +0

<LeeF> REDUCE or REDUCED

<SimonR> REDUCTIBLE

<AndyS> RUSTED

<Souri> +q

ericP: +1 REDUCED

<ericP> +1 for REDUCED

<LeeF> PROPOSE: SPARQL SELECT queries with no keyword following SELECT must

<LeeF> return the precise cardinality of duplicate solutions specified by the

<LeeF> algebra; SPARQL contains a REDUCED keyword that allows duplicate

<LeeF> solutions to be returned with cardinality of at least 1 and no greater

<LeeF> than that specified by the algebra. The REDUCED keyword will be marked as

<LeeF> an at-risk feature of SPARQL.

<SteveH> REDUCE works better with DISTINCT, surely?

<SteveH> otherwise it would be DISTINCTED

<SimonR> DISTINGUISHED

AndyS: Will the text highlight that this could affect future counting?

ericP: I don't think it does.

AndyS: If you do SELECT REDUCE DISTINCT COUNT is meaningless

LeeF: I'm inclined to say no because we don't know what COUNT looks like

<AndyS> SELECT REDUCE COUNT(*)

SimonR: This is specifically for the case when you don't want to count things

<ericP> second

<jeen> second

<SimonR> Abstaining.

AndyS: Abstain

RESOLVED

<ericP> RESOLVED, AndyS and SimonR abstaining

<LeeF> ACTION: Eric to incorporate text for REDUCED into rq25, including text noting that REDUCED does not play nicely with counting [recorded in http://www.w3.org/2007/03/20-dawg-minutes.html#action11]

issue entailmentFramework

<LeeF>

LeeF: patH do you think we have address the entailment framework issue in the document?

patH: yes

<LeeF> PROPOSED: that version 1.59 of rq25 addresses and closes http://www.w3.org/2001/sw/DataAccess/issues#entailmentFramework

<ericP> second

RESOLVED

<LeeF> ACTION: LeeF to close #entailmentFramework issue [recorded in http://www.w3.org/2007/03/20-dawg-minutes.html#action12]

active graph

<LeeF> http://lists.w3.org/Archives/Public/public-rdf-dawg/2007JanMar/0163.html

<ericP> PatH's mail, Subject: comments on section 12 (and a little more)

<LeeF> ACTION: AndyS to incorporate and explain the notion of active graph in section 8 and/or wherever else appropriate [recorded in http://www.w3.org/2007/03/20-dawg-minutes.html#action13]

LeeF: Anybody who does not think it's a good idea to move to Last Call?
... Is anybody uncomfortable the way we are moving towards last call?
... Nobody raised any issues or concerns with my two questions. Fantastic.

<LeeF> PROPOSED: To publish as a Last Call working draft rq25 v1.59 plus text added for relevant action items (REDUCED and active graph)

<LeeF> PROPOSED: To publish as a Last Call working draft rq25 v1.59 plus text added for relevant action items (REDUCED and active graph) and text for remainder of PatH's Section 12 review

<LeeF> PROPOSED: To publish as a Last Call working draft rq25 v1.59 plus text added for relevant action items (REDUCED and active graph) and text for remainder of PatH's Section 12 review plus query results for example query at beginning of section 11

AndyS: I think it's OK

second

ericP: second, much later after Elias

LeeF: Good job everybody. I appreciate all of your hard work.
... but you are not off the hook just yet

<LeeF> last bytes editorial call at 15:00 UTC on Friday

<Souri> What's corr EDT?

<LeeF> 11:00AM EDT

<AndyS> DAWG telecon +30mins

LeeF: Over the next few weeks we are going to be focusing on answering comments and the test suite.
... If everything goes well, we'll be working on the implementation report.

ADJOURNED

<SteveH> bye

<SimonR> 15:20Z

<scribe> ACTION: ericP schedule teleconf for Friday [recorded in http://www.w3.org/2007/03/20-dawg-minutes.html#action14]

- DRAFT -

RDF Data Access Weekly

20 Mar 2007

Attendees

Contents

action items

Duplicate solution cardinality

issue entailmentFramework

active graph

Summary of Action Items

Scribe.perl diagnostic output