DAWG Weekly Meeting

 

 

<LeeF> dangnabbit

<LeeF> tricksy spacebar

<kendallclark> aaaa is FredZ

<kendallclark> Scribe: ericP

<scribe> scribe: Elias

<kendallclark> scribe: EliasT

<kendallclark> http://lists.w3.org/Archives/Public/public-rdf-dawg/2006AprJun/0102.html

<AndyS> +1 617 761 6200

<AndyS> then 7333#

<kendallclark> code: 7333#

<AndyS> CR exit process? Has anything moved on since the CR request?

<ericP> next meeting: same bat time, same bat channel scribe: FredZ

<scribe> Scribe: EliasT

<AndyS> The other bot is RRSAgent which tracks actions and makes the log

<scribe> Chair: kendallclark

<ericP> KendallC: [introducing the FredZ to scribing technique] record any decisions

<ericP> ... best effort to record what people say

FredZ, http://www.w3.org/2001/12/zakim-irc-bot.html

FredZ, http://www.w3.org/2002/03/RRSAgent

<ericP> last working meeting

<ericP> RESOLVED: accept last working group meeting record as a true record

convene, take role, approve last meeting's minutes

<AndyS> http://www.w3.org/2001/sw/DataAccess/rq23/#func-RDFterm-equal

<ericP> Operator Mapping

kendallclark: ericP did you propose text and tests to add boolean {boolean <> = !=} to SPARQL?

ericP: I only sent email, have not checked in text to CVS

AndyS: do tests exist?

ericP: I wasn't paying too much attention.

<ericP> aforementioned tests

<LeeF> FILTER ?a = ?b = ?c .

<LeeF> AndyS gets a parsing error

<LeeF> so does ericP

<ericP> http://www.w3.org/2005/01/yacker/uploads/SPARQL?lang=perl&text=ASK+%7B+%3Ax+%3Ay+%3Az+FILTER+%28+%3Fa+%3D+%3Fb+%3D+%3Fc+%29+%7D&action=validate+text

<kendallclark> *ah*

ericP: FredZ proposes a syntax that would be ambigous. AndyS and ericP get parser errors.

<kendallclark> Steve: sure, we just got started, really

<AndyS> http://sparql.org/validator?query=SELECT+*%0D%0AWHERE%0D%0A+++%7B+FILTER+%28+%3Fa+%3D+%28%3Fb+%3D+%3Fc%29+%29+%7D&linenumbers=true

<ericP> http://www.w3.org/2005/01/yacker/uploads/SPARQL?lang=perl&text=ASK+%7B+%3Ax+%3Ay+%3Az+FILTER+%28+%3Fa+%3D+%28+%3Fb+%3D+%3Fc+%29+%29+%7D&action=validate+text

<scribe> ACTION: LeeF to fix mimetype in json-reults [CONTINUES] [recorded in http://www.w3.org/2006/06/06-dawg-minutes.html#action01]

<scribe> ACTION: EricP to propose text and tests to add {boolean < > [recorded in http://www.w3.org/2006/06/06-dawg-minutes.html#action02]

= !=

boolean} to SPARQL [CONTINUES]

<LeeF> looking now

<LeeF> There's an incorrect reference to text/json in the document. that's my action.

<LeeF> the mime type that we use is orthogonal to my action.

<scribe> ACTION: EliasT to submit mime-type registration for json-results. [recorded in http://www.w3.org/2006/06/06-dawg-minutes.html#action03]

<scribe> ACTION: AndyS: to draft of open-world = tests (unknown datatypes) [CONTINUES] [recorded in http://www.w3.org/2006/06/06-dawg-minutes.html#action04]

<LeeF> ACTION -4

<ericP> DONE

<LeeF> ACTION: AndyS: to draft of open-world = tests (unknown datatypes) [DONE] [recorded in http://www.w3.org/2006/06/06-dawg-minutes.html#action05]

<AndyS> A != B on RDF-terms is currently not(RDF-term-equals) which is not open world

<ericP> 1 != "1"^^foo:int

AndyS: in an open-world is unknown.

ericP: would like it to result in a error.

<ericP> error is the tightest way to describe the result

<AndyS> Currently :: not( 1 = "1"^^foo:int) => true

<kendallclark> +1

patH: if we give a specific error in the case, bots could use that to perform private world calculations.

<AndyS> FILTER(?x=?y)

<SteveH> FILTER(termEqual(?zx, ?Y))

<ericP> FILTER(?x=?y || ?x SAMETERM ?y)

<ericP> FILTER(?x=?y || ?x !SAMETERM ?y)

<AndyS> Issue: Things like FILTER (?x = <x>) currently work.

<AndyS> which is intuitive.

<SteveH> +1

track action items

<scribe> ACTION: AndyS to resend a proposal to solve equality testing over email, ericP and patH to give eyes. [recorded in http://www.w3.org/2006/06/06-dawg-minutes.html#action06]

<kendallclark> http://lists.w3.org/Archives/Public/public-rdf-dawg-comments/2006Jun/0008.html

<scribe> ACTION: AndyS to contribute a couple tests for computed-properties/virtual-prop/builtin. one using lists, and one not using lists [DONE] [recorded in http://www.w3.org/2006/06/06-dawg-minutes.html#action07]

AndyS, was that right?

<SteveH> UNION is not compositional, having variables that appear only in UNION blocks is legal

<AndyS> More concrete example would be good - they showed it was compositionaly under the restrictions they assume.

<kendallclark> message seems full of concrete examples of their main claims...

<kendallclark> cool :)

<SteveH> +1 to most conservative

<AndyS> SELECT * { pat OPT pat1 OPT pat2 }

<AndyS> SELECT * { { pat OPT pat1 } OPT pat2 }

<ericP> SELECT * { { pat OPT pat1 } . {} OPT pat2 }

<kendallclark> we'll cover for you Elias :)

<kendallclark> you slacker

<kendallclark> -wink-

<ericP> SELECT * { { pat OPT pat1 } { OPT pat2 } }

<ericP> SELECT * { { pat OPT pat1 } { { } OPT pat2 } }

<AndyS> (OPT (OPT pat pat1) pat2))

<AndyS> http://www.hpl.hp.com/techreports/2005/HPL-2005-170.html

<ericP> 2006-06-06T12:55:56Z <ericP> "Paradoxical SPARQL query?", "Mismatch with multiple OPTIONAL and grammar?"

<AndyS> which is May 0009

<AndyS> "FILTER evaluation problem" is Apr 0025

<AndyS> OPT is binary - gives an interpretation of the query

<AndyS> This goes back to an ealier dialog when they had a different design for optional

<kendallclark> ACTION: EricP to send mail about 3-part interpretation of optionals [recorded in http://www.w3.org/2006/06/06-dawg-minutes.html#action08]

<kendallclark> ACTION: PatH to kibbitz with EricP about 3-part interpretation of optionals [recorded in http://www.w3.org/2006/06/06-dawg-minutes.html#action09]

<kendallclark> ACTION: KendallC to tell Martin Soto that we're considering his comments [recorded in http://www.w3.org/2006/06/06-dawg-minutes.html#action10]

<AndyS> hence I believe its already been answered

<kendallclark> ADJOURNED