Provenance Working Group Teleconference -- 07 Feb 2013

<trackbot> Date: 07 February 2013

<pgroth> trackbot, start telcon

<trackbot> Meeting: Provenance Working Group Teleconference

<trackbot> Date: 07 February 2013

<pgroth> Scribe: Tom De Nies

<pgroth> tom are you to scribe?

<pgroth> you ready to scribe?

yes, hold on, trying to dial in

<pgroth> ok

ok: )

Admin

<pgroth> http://www.w3.org/2011/prov/meeting/2013-01-31

pgroth: approval of last week's minutes

<dgarijo> +1

<khalidBelhajjame> +1

<Curt> +1

<hook> +1

<ivan> 1

<TallTed> +0

<CraigTrim> +1

<pgroth> accepted: Minutes of Jan. 31, 2013

pgroth: open actions: Luc needs to FAQ and I responded to Clark & Parsia
... This closes all outstanding comments on the CR.
... Next week Luc and I can't make it, so we will cancel next week's telecon

PROV-XML

pgroth: We want to converge on some outstanding issues
... First issue. Which identifiers do we use in PROV-XML?
... editors are for the use of qnames

zednik: Qnames seem to have the least issues.
... so best remaining option

<hook> http://www.w3.org/2011/prov/wiki/Prov-XML_Identifiers

pgroth: comments?

hook: Before we vote, could we list the disadvantages of qnames?

<tlebo> If you're excluding SADI!

<tlebo> ?!

hook: for example, full URIs are not valid values of xs:QName

tlebo: This design is currently excluding SADI?

hook: you must use a namespace

tlebo: So is prefix an XML element?
... ignoring the existing XML ways to do this and defining our own

zednik: We didn't go down that path. We use the standard XML constructs

<pgroth> Proposed: Adopt qnames for identifiers in PROV-XML

<ivan> +1

<SamCoppens> +1

<TallTed> +1

<zednik> +1

<khalidBelhajjame> +1

<Curt> +1

<hook> +1

<jcheney> +1

+0 (havent read all options)

<dgarijo> +0 (I haven't been following the discussions)

<satya> 0, I have not read it

<pgroth> Accepted: Adopt qnames for identifiers in PROV-XML

<Curt> I summarized the attribute ordering issue here: http://www.w3.org/2011/prov/wiki/Prov-XML_ElementOrdering

pgroth: Next issue is ordering of attributes to support "nice" JAXB

zednik: I think there is consensus now.
... Initially, the group didn't want to change the schema and work via JAXB mapping/etc. But those options seemed too complex, so Curt and I worked on adding ordering yesterday
... So it's reasonable to proceed with ordered attributes

Curt: I think it's cleaner and more elegant

<pgroth> Proposed: use order attributes in prov-xml

Curt: If we can change our schema to work with such a widely used tool, it's worth doing

<ivan> +1

<SamCoppens> +1

<zednik> +1

<Curt> +1

<hook> +1

<jcheney> +1

<TallTed> +1

<dgarijo> +1

<pgroth> resolved: use order attributes in prov-xml

pgroth: Last issue was still under discussion
... How are the xsd files organized and is it appropriate for the use of legacy xml?

zednik: Stian put together a nice page with all options
... We've reorganized the schemas to include the core schema
... Everything seems to be working. There's still some discussion between Luc and Stian on the mailinglist.

Luc: I don't think there is consensus. Still have question.

<zednik> http://www.w3.org/2011/prov/wiki/ProvXMLNamespaces#Substitution_groups_and_abstract_elements

Luc: There are schema's out there describing provenance, and we want to be able to include these in PROV-XML
... The proposed solution with the substitution group seems to require changes to legacy schema's.
... and this is undesirable.

zednik: The substitution groups are meant to include elements that are in the provenance namespace.
... for everything else we still have xs:any
... so you can use elements that don't use the prov: namespace

Luc: Ok, thank you for the clarification.
... So, as we were talking about JAXB, does this also work with JAXB?

zednik: yes, the current version works with JAXB.

<pgroth> cool

Luc: So dictionary and mention work as well?

zednik: yes.

Luc: Why do we need a sequence of choice and abstract elements in the document element?

zednik: I think the latest schema might fix that
... will doublecheck.

pgroth: So are you ok with the current solution, Luc?

Luc: Yes, it seems it's ok now.

pgroth: Seems the issue is resolved now then.
... To wrap-up, is there anything outstanding now?

zednik: just fixing up the text.

pgroth: Once the text is updated, we can go to internal review.

Luc: We need to make sure to get the PROV-XML reviews back in time for the PR pulication then

Implementation Report

pgroth: Dong and I have been working on the report

<pgroth> https://dvcs.w3.org/hg/prov/raw-file/default/reports/prov-implementations.html

pgroth: Dong did an impressive amount of work.
... The report is almost done. Some small stuff remaining to do.
... We'd like to send out the report for review on Monday, and set up the timetable for publication and going to the next stage.

Ivan: Maybe we can get the votes in by email by the end of next week
... since there is no telecon next week

Luc: It may be good to try and speed up the process. But we need to allow them to respond/revise.

<dgarijo> there is a vocabulary named Jun Zhao :O

Luc: after the review.

<Luc> + section 7

Ivan: It looks good. We still need something that says: we met the exit criteria and this is why.
... We have a lot of implementations, 54, which is very impressive.

<pgroth> everybody keep your enthusiasm up :-)

pgroth: Once we say we want to go to PR, do we need the big long email again?

ivan: yes, just like we did with CR.
... I can help.

<Luc> is it time to think of a publication date, to finalize our documents?

ivan: Am I correct that there is not that much editing left to be done?

pgroth: Just PROV-Constraints

jcheney: will do that tomorrow

ivan: excellent.

Luc: If we are going to vote by email, and the outcome it positive, we want to be ready for PR as soon as possible. So we need the URLs with the publication date.
... So I suggest that Ivan starts the process to get the date agreed.

ivan: either March 7th or 12th

pgroth: I think we should aim for the 12th

ivan: yes, that seems without risk.

Luc: so the url's?

ivan: I think it's safe that the editors start preparing the documents for 2013.03.12 as date

<Luc> can we check all notes will be ready then?

pgroth: For health and safety reasons of half the chairs, we should ask the webmaster for the date :)

ivan: will do
... Do note: what we send out for PR is the final document. So be sure to spellcheck etc. and be 100% sure

+q to ask if this also goes for the final notes

<tlebo> so, when are the PR drafts finalized?

<Zakim> TomDN, you wanted to ask if this also goes for the final notes

TomDN: Can we still touch the notes after we send out the PRs?

<dgarijo> ok!

<Luc> do we have time to review those that still need review?

pgroth: yes, but we'd like the final version of the notes on the 12th

<pgroth> ACTION: paul to remind note editors about March 12 deadline [recorded in http://www.w3.org/2013/02/07-prov-minutes.html#action01]

<trackbot> Created ACTION-164 - Remind note editors about March 12 deadline [on Paul Groth - due 2013-02-14].

pgroth: We will send out a note via email about this

Zakim. mute me

ivan: So april 23rd will probably be the final final final final date for the notes as well.

<dgarijo> +q

Luc: I want to make sure that we have the time to get all notes ready by March 12th
... Some still need review
... for example PROV-XML, maybe PROV-AQ, PROV-DC, ...

ivan: There's no last call for notes, so the 12th is a WD publication, so April 23rd is the final date

Luc: but some reviewers blocked the release of PROV-AQ

<dgarijo> -q

Luc: and we need a vote for the synchronized release of the notes
... So we need an appropriate amount of time.
... Will we have enough time?

<satya> sorry, I have to leave

pgroth: We have 4 weeks to complete reviews
... PROV-XML looks almost done, PROV-DC we will hear about next, and PROV-AQ will probably be done in time.

PROV-DC

<dgarijo> http://www.w3.org/2011/prov/wiki/AnswersToProvDCReviewers

dgarijo: update on PROV-DC. Been working on response to reviews and edits in the document
... Raised new issues for the remaining things to do.
... Main issues are summarized on wikipage
... about dct:isVersionOf, we have enough material to discuss with dct people.

pgroth: Would like review now or after issues are resolved?

dgarijo: Would like time to finish the issues. But if someone wants to have a look, be my guest. People can also answer in comment to the wikipage

<pgroth> http://dvcs.w3.org/hg/prov/raw-file/default/dc-note/dc-note.html

<tlebo> when is the dead deadline for the implementation report?

pgroth: OK thanks. We still have time for extra implementations. So submit!

+q to ask about implementations on Notes

TomDN: Where can we put implementations of notes?

<Zakim> TomDN, you wanted to ask about implementations on Notes

pgroth: On the implementation wiki page

<dgarijo> bbye

<Dong> bye

bye

<pgroth> trackbot, end telcon

- DRAFT -

Provenance Working Group Teleconference

07 Feb 2013

Attendees

Contents

Admin

PROV-XML

Implementation Report

PROV-DC

Summary of Action Items

Scribe.perl diagnostic output