RDF Data Access WG meeting -- 28 Nov 2006

<jeen> pfft. no qualified names....

I'm qualified.

<scribe> Scribe: Andy Seaborne

<scribe> ScribeNick: AndyS

<kendallclark> after 3 years, i forget the code! :)

<kendallclark> 2, 3, wahtever it's been

It's RDFF.

<kendallclark> heh

<kendallclark> i didn't know that!

KC: little email traffic for 2 weeks leads to a short agenda each week

Also: http://www.w3.org/2001/sw/DataAccess/rq23/rq24-algebra.html

EricP: Q about duplictae answers and bNodes in BGPs

<kendallclark> http://lists.w3.org/Archives/Public/public-rdf-dawg/2006OctDec/att-0167/21-dawg-minutes.html

EricP: Elias will be along soon.

Minutes accepted.

RESOLVED to accept the minutes of Nov 21.

AndyS available.

Pat (for some of the time)

Eric notes it's during XML2006.

<scribe> Scribe: EricP

action items (this is prolly a no-op)

<AndyS> Scribe next time: EricP

<kendallclark> ACTION: ericP to bring up test on bags for next meeting. [recorded in http://www.w3.org/2006/11/28-dawg-minutes.html#action01]

<kendallclark> ACTION: ericP to seek clarification on http://www.w3.org/mod/20061110085518567.00000002912@bmacgregor1 [DONE] [recorded in http://www.w3.org/2006/11/28-dawg-minutes.html#action02]

<AndyS> DONE action 1

<AndyS> ACTION: ericP to seek clarification on http://www.w3.org/mod/20061110085518567.00000002912@bmacgregor1 [CONTINUES] [recorded in http://www.w3.org/2006/11/28-dawg-minutes.html#action03]

<AndyS> (scribe will clear the mess up later :-)

<kendallclark> ACTION: KendallC to close formsOfDistinct issue [recorded in http://www.w3.org/2006/11/28-dawg-minutes.html#action04]

<kendallclark> ACTION: PatH to review the proposed tests in http://lists.w3.org/Archives/Public/public-rdf-dawg/2006JulSep/0169 and say yay or nay [recorded in http://www.w3.org/2006/11/28-dawg-minutes.html#action05]

<kendallclark> PatH to draft replacement vanilla entailment section for WG consideration

<AndyS> ACTION: PatH to draft replacement vanilla entailment section for WG consideration [CONTINUES] [recorded in http://www.w3.org/2006/11/28-dawg-minutes.html#action06]

<kendallclark> ACTION: KendallC to remember that the wee, lost filter tests should be put to the question (re: http://lists.w3.org/Archives/Public/public-rdf-dawg/2006OctDec/0099.html ) [recorded in http://www.w3.org/2006/11/28-dawg-minutes.html#action07]

convene, boot up

test suite

<AndyS> KC: An output of the WG

<AndyS> ... issue is now where we are in procss terms on this. Figure out where we go next.

<patH> Jeen said 'eep'

<kendallclark> which i tried to turn into "yes"

<AndyS> Lee: does Eric / Andy have a timeline for the QL spec?

<kendallclark> EliasT: where you going to work on test suite?

<jeen> (dunno what's wrong) anyway, I'd be happy to spend time on test suite, particularly on isolating a set of tests specifically for use in testing a SPARQL parser implementation and a SPARQL engine implementation. Mostly that means sorting out old tests, etc.

<AndyS> AndyS: timeline - algebra mostly there -need integration into doc.

<AndyS> ... discussion on the algebra can be now.

<AndyS> ... integration wil be in the new year.

<AndyS> ... test suite process can be done now - most tests not affected.

<AndyS> KC: Need a process owner.

we need a mom

<AndyS> ... coordinate work as the chair changes (swivels?)

well, the tests do

<AndyS> Lee: will talk to Elias.

<jeen> ok, I give up.

<patH> Jeen might try tapping out morse.

<AndyS> I've been using a test suite for the algebra and have some new corner cases (just 3) Rest of tests went through fine.

<AndyS> KC: Thinks Jeen is agreeing to be the owner

<kendallclark> wonderful

<AndyS> ... Jeen agrees to take the lead.

<EliasT> I was working on the "protocol" test suite and be of help to jeen as I last recalled.

<LeeF> AndyS, when you say "rest of tests", what exactly do you mean? (considering that some tests seem to be completely broken?)

<AndyS> All the DAWG tests I think are valid (approved and unapproved) + my own.

<LeeF> I see.

<AndyS> ~ 500.

<jeen> ok but I don't know a deadline

<AndyS> ??? ACTION: Jeen propose test suite process.

<jeen> ah ok

<AndyS> ... does not imply "do it all"

<jeen> phew :)

<AndyS> ACTION: Jeen propose test suite process (not do it all). [recorded in http://www.w3.org/2006/11/28-dawg-minutes.html#action08]

duplicate answers

http://www.w3.org/2001/sw/DataAccess/tests/#rdfsemantics-bnode-type-var

<AndyS> EricP: bad laptop

<kendallclark> http://lists.w3.org/Archives/Public/public-rdf-dawg/2006OctDec/0164.html

<patH> Im having trouble hearing Eric.

<kendallclark> start of the thread

<AndyS> (after Eric finishes)

<patH> I can't folow the audio at this point.

<kendallclark> it's hard to hear Eric today

<Zakim> AndyS, you wanted to ask Pat about existentials and counting

<AndyS> Data: { :a :p 1. :a :p 2 .}

<AndyS> Query: :a :p []

<AndyS> Query: ?a :p []

<AndyS> Q is whether you get one answer or two?

<AndyS> Query: ?a :p ?v

:a or :a, :a

(compare to (:a 1) vs, (:a 1),(:a 2))

<LeeF> What do current implementations do? (Mine treats bnodes as unprojected variables.)

mine does as well

<jeen> That's what we do as well

<AndyS> ARQ can do either - but woudl like a decision :-)

<AndyS> Pat: instinct - get two

<LeeF> ARQ is so wishy-washy ;-)

<LeeF> I wonder who (specific orgs or general communities) will object to treating blank nodes as unprojected variables -- Will the DL community object to that?

<kendallclark> (I don't think it will.)

<AndyS> I think we are defining for simple entailment only. DL replaces the whole of BGP matching with what it wants. There is no overall constraints on dups.

<kendallclark> (well, it won't if that can be made compatible w/ sparql-dl, which doesn't exist yet except in people's heads)

<LeeF> Oh right, Andy, thanks.

<AndyS> http://lists.w3.org/Archives/Public/public-rdf-dawg/2006OctDec/0140

+1 on duplicates for many reasons

<AndyS> Pat: duplicates make no difference to semantics.

<AndyS> ... reports Bijan woudl be happy with that (absence of DISTINCT)

<AndyS> KC: hers agreement

<AndyS> KC: hears agreement

perhaps approve the test first

<kendallclark> first or second, no matter, i think

<AndyS> This editor is informed by the consensus hear

<AndyS> AndyS: wil try to incorporate Jorge's style of approach: wil wait for Pat's text.

PROPOSED: approve http://www.w3.org/2001/sw/DataAccess/tests/#rdfsemantics-bnode-type-var

<LeeF> Seconded.

<jeen> Seconded

steve and andy have both tried it in their impls

or at least, that's the feedback i got a while ago

<AndyS> Goods OK to me (scanning the HTML)

<AndyS> RESOLVED

<kendallclark> 0 abstensions or objections

convene, boot up

algebra updates

<kendallclark> eric: i hate zakim! :)

andy: am hppy with structure. looking for comments on the design

<AndyS> http://www.w3.org/2001/sw/DataAccess/rq23/rq24-algebra.html

i think i can fix the last query publication CSS

i"m afraid to commit

<AndyS> CSS changes don't change the semantics :-)

<LeeF> ACTION: LeeF to review http://www.w3.org/2001/sw/DataAccess/rq23/rq24-algebra.html [recorded in http://www.w3.org/2006/11/28-dawg-minutes.html#action09]

<jeen> uhm.... (closes eyes, takes deep breath) yes, action me.

<AndyS> ACTION: Jeen review rq24-algebra [recorded in http://www.w3.org/2006/11/28-dawg-minutes.html#action10]

+1 to adjourn

<AndyS> ADJOURN

<AndyS> Meet next time - one week - Eric to scribe

<jeen> ok, seeya all.

<kendallclark> ciao

<patH> byebyee

- DRAFT -

RDF Data Access WG meeting

28 Nov 2006

Attendees

Contents

action items (this is prolly a no-op)

convene, boot up

test suite

duplicate answers

convene, boot up

algebra updates

Summary of Action Items

Scribe.perl diagnostic output