See also: IRC log
LeeF: meeting tommorow for disposition of comments
<LeeF> minutes from Sep 18
<LeeF> minutes from Sep 25
<LeeF> approve both sets of minutes
LeeF: if you can't make next weeks call, make your feelings known about transition to PR
<LeeF> ACTION: LeeF to check and annotate open-eq-07 and -10 with mf:LangTagAwareness [DONE] [recorded in http://www.w3.org/2007/10/02-dawg-minutes.html#action02]
<LeeF> ACTION: LeeF: open and postpone an issue on unescapes XML in result set format [DONE] [recorded in http://www.w3.org/2007/10/02-dawg-minutes.html#action03]
<LeeF> ACTION: LeeF to investigate PR mechanics, put together draft transition request [CONTINUES] [recorded in http://www.w3.org/2007/10/02-dawg-minutes.html#action04]
<LeeF> ACTION: ericP to poke IETF folks about registering SPARQL media types (esp. application/sparql-query) [CONTINUES] [recorded in http://www.w3.org/2007/10/02-dawg-minutes.html#action05]
LeeF: tests were correct, details in the specification were wrong
<LeeF> link to both bugs with details
<LeeF> SPARQL Algebra Translation Bug
AndyS: there should only be 1 way to get filters into OPTIONALs
AndyS: there are tests that back this behavior up
-> Chime's reading of Andy's overview
LeeF: we can make the case that implementors were not mislead and don't need to go back to LC
<LeeF> proposed new text is:
<LeeF> [[
<LeeF> If F is not empty:
<LeeF> G := Filter(F, G)
<LeeF> ]]
<LeeF> PROPOSED: To fix the bug in attaching filters to patterns as per AndyS's suggestion in http://lists.w3.org/Archives/Public/public-rdf-dawg/2007JulSep/0177.html
<LeeF> ericP seconds
<LeeF> resolved
<LeeF> ACTION: AndyS to put replacement text for filter attachment into rq25 [recorded in http://www.w3.org/2007/10/02-dawg-minutes.html#action06]
<LeeF> ACTION: LeeF to examine 7 tests affected by change in http://lists.w3.org/Archives/Public/public-rdf-dawg/2007JulSep/0177.html in view of implementors' experiences [recorded in http://www.w3.org/2007/10/02-dawg-minutes.html#action07]
<LeeF> AndyS on simplifying {{ ... }}
AndyS: if you apply the simplification early, you end up with a different Algebra
ericP: ask implementors to run a distinguishing query
AndyS: it is not exactly clear in the process document if we can escape another LC cycle
<LeeF> ACTION: LeeF to mail out a complete test case to determine implementation behavior in the OPTIONAL+FILTER+{{...}} case [recorded in http://www.w3.org/2007/10/02-dawg-minutes.html#action08]
<LeeF> contains Chimezie's proposted new test
I wrote that while trying to figure out why: LeftJoin(Ω1, Ω2, expr) = Filter(expr, Join(Ω1, Ω2)) set-union Diff(Ω1, Ω2, expr)
isnt' equivalent to :
LeftJoin(Ω1, Ω2, expr) = Filter(expr, Join(Ω1, Ω2)) set-union Ω1
<AndyS> ARQ passes the test
ericP: does top-down / bottom-up evaluation effect this?
<LeeF> PROPOSED: Approve the test in http://lists.w3.org/Archives/Public/public-rdf-dawg/2007OctDec/0003.html and ask for implementation results for it
-> new test
<LeeF> ACTION: LeeF to solicit implementors' desired behavior for {{ ... }} case [recorded in http://www.w3.org/2007/10/02-dawg-minutes.html#action09]
AndyS: even if we don't approve the tests, we should solicit input about what implementations *want* to do (or what they do currently)
<LeeF> resolved
<LeeF> ACTION: LeeF to mark test in 0003 as approved, update test suite, notify implementors [recorded in http://www.w3.org/2007/10/02-dawg-minutes.html#action10]
<ericP> actually, i do pass it -- just didn't match the name in the manifest
LeeF: main challenge is the protocol document
LeeF: I'll be cleaning up outstanding comments
LeeF: if you are swapped in on the protocol, please contribute to thread if neccessary
AndyS and EricP: advice to LeeF -- use cvs annotate and cvs log
<LeeF> query language implementation report
ericP: a failure for a test in a facet is a failure for the whole feature set
ericP: the data we have may be sufficient for transition
<LeeF> ACTION: ericP to implement the conservative algorithm (fail a test -> fail the feature) in the implementation report [recorded in http://www.w3.org/2007/10/02-dawg-minutes.html#action11]
<LeeF> ACTION: LeeF to weed out basic facets from complex tests [recorded in http://www.w3.org/2007/10/02-dawg-minutes.html#action12]
<LeeF> email on pyrrho db results
LeeF: there is no EARL output
ericP: we should not include implementations which do a manual check instead of automated generation of EARL
AndyS: we *should* be doing automated test reporting. perhaps we don't put it in the implementation report matrix
I agree with the caution about manually generated reports (given the nature of BNodes, etc..)
<LeeF> ACTION: LeeF to solicit mechanically generated test results from pyrhho db [recorded in http://www.w3.org/2007/10/02-dawg-minutes.html#action13]
ericP: an approach for a first impl report could be to only show those for which we are 'green across the board'
<LeeF> ACTION: ericP to adapt impl report to include syntax tests [recorded in http://www.w3.org/2007/10/02-dawg-minutes.html#action14]
rssagent, make logs world-visible
<LeeF> EliasT close to having HTTP protocol testing running
<LeeF> LeeF working on creating manually generated implementation report for protocol
<LeeF> ...will work with AndyS to fit in SOAP testing in that
<LeeF> ...results from testing protocl should demonstrate 2 producers and consumers of SPARQL XML
<LeeF> example testimonials
<LeeF> + solicit testimonials from WG members
<LeeF> + ask WG members to solicit testimonials from other W3C members
<ericP> Q: testimonials from non-W3C people?
<LeeF> + ask SWEO to help provide testimonials
<LeeF> + start soon to help people get testimonials cleared with their organizations
<iv_an_ru> small spam might help :)
<LeeF> adjourned