<trackbot> Date: 06 September 2012

<Luc> scribe: simon miles

admin

Luc: outlines agenda

<Luc> proposed: to approve Minutes of the Aug 9, 2012 Telecon

<Luc> http://www.w3.org/2011/prov/meeting/2012-08-09

<TomDN> +1

<Curt> +1

<SamCoppens> +1

<satya> +1

<khalidBelhajjame> +1

<hook> +1

+1

<Paolo> 0 (not attended)

<Luc> accepted: Minutes of the Aug 9, 2012 Telecon

<stainPhone> 0

Luc: open actions for Paulo and Paul

Paul: Not yet done overview slide for Wiki

F2F4

Luc: Idea to co-locate F2F4 with ISWC

<Luc> November 9 - 10, 2012

Luc: available before ISWC workshops/tutorials, 9-10
... any blocking reasons against?

<pgroth> +q

Luc: Vote or just accept those dates?

Paul: Don't think we need a vote, but need sign-up page

<pgroth> yes

<Luc> ACCEPTED: F2F4 will take place in Boston, on November 9 - 10, 2012

Luc: details available later, but hosted at MIT

Ivan: at Computer Science

Luc: questions about F2F4?

<pgroth> ACTION: paul to set-up web page f2f4 [recorded in http://www.w3.org/2012/09/06-prov-minutes.html#action01]

<trackbot> Created ACTION-104 - Set-up web page f2f4 [on Paul Groth - due 2012-09-13].

prov-constraints

Luc: At last telecon, we looked at remaining technical issues on the document

<GK> (I assume MIT is easily accessible from the ISWC and/or downtown location?)

<Luc> http://lists.w3.org/Archives/Public/public-prov-wg/2012Aug/0228.html

<stainPhone> @gk yes, metro from downtown

Luc: work was done over August

<Luc> http://dvcs.w3.org/hg/prov/raw-file/default/model/releases/WD-prov-constraints-20120911/Overview.html

Luc: distributed link above for feedback (seems good so far)

<Paolo> sorry -- new VOIP client

Luc: before formal vote, is there any comment on the document?

Ivan: one thread of discussion on terminology alignment of "top-level bundles". What is current status?

Luc: Top-level bundle introduced in PROV-N, not part of PROV-DM
... it is the structure of the complete document you would write in PROV-N
... "top-level bundle" was not adequate term, because it is not a bundle
... James suggested talking about a "PROV document" instead, and this is adopted in PROV Constraints
... PROV documents can contain PROV statements or bundles
... In context of PROV-N, will propose that will align terminology to also use PROV document

stian: Thank you to the editor for addressing the issues I raised

Luc: Also thank you and other reviewers for quality of reviews

<Luc> PROPOSED: publish PROV-Constraints as Last Call Working Draft

+1

<MacTed> +1

<stainPhone> +1

<ivan> +1

<khalidBelhajjame> +1 (University of Manchester)

<TomDN> +1 (IBBT)

<Curt> +1 (NASA)

<Paolo> +1 Newcastle University

<zednik> +1 (RPI)

<pgroth> +1 (VU University Amsterdam)

<SamCoppens> +1 (IBBT)

<satya> +1, Invited expert (CWRU)

(invited expert, King's College London)

<GK> +1 Oxford U

<hook> +1 (IE)

<MacTed> (OpenLink Software)

<dgarijo> +1 (UPM)

<Luc> +1 (University of Southampton)

<stainPhone> (univ of Manchester as well)

<jun> +1

<Luc> Accepted: to publish PROV-Constraints as Last Call Working Draft

<pgroth> +q

Luc: When are we going to release the document? And what is the review period?
... Editors do not intend to make any changes, can prepare for release next Tuesday pending webmaster approval
... Release date 11 September
... Propose review period closes 10 October

<ivan> +10000 to Paul

Paul: Have to also consider some blog post to go with document, it is heavy duty and needs context to interpret
... Who will write the post?

Luc: Paolo, would you have bandwidth to draft blog entry?

<pgroth> i can then polish

Paolo: Can try to first draft something tomorrow

Luc: I could then work on it on Monday

Paul: That's fine and can also run by me

<pgroth> nice one everyone

Luc: Any other comments on constraints?

Ivan: Reminder that we should finalise the timetable tomorrow
... should become synchronised with other documents

Implementing constraints with SW technologies

Paul: We have noticed that there is discussion around implementing constraints using SW tech
... great, but not part of the WG's responsibility
... should be done by individuals if interested, but implementation should not be done in WG time, cannot respond to all questions
... encouraged but outside WG

<GK> I agree with Paul here, but I also note that we'll need interoperable implementations for REC track progress?

Ivan: Agreed that deliverables need to be delivered, but if work is done WG might decide to publish in W3C Notes and could be valuable

<khalidBelhajjame> @Paul, agreed. With Jun and Stian, we decided to specify the constraints (that are speciable within OWL), outside the context of the working group

Luc: Yes, are very keen for implementation of constraints using SW tech, but concerned about using WG bandwidth and mission creep

<zednik> https://docs.google.com/spreadsheet/viewform?formkey=dGM4cXZYMk0xaFBDT2VyRV92YkY5WkE6MQ

zednik: Added constraints implementation section to implementation questionanire
... not intended to be saying we will be creating these implementation, but part of implementation report

<pgroth> +q

Paolo: Agree not done on WG time, but cannot claim have reference implementation without compliance validator

<pgroth> mute paolo

Paul: Implement both specs, and if don't have implementations then spec can't be approved as recommendation

<stainPhone> @paolo, yes, our (khalid/jun) intent is to do what can practically be done in owl

<Paolo> agree -- and that's why Khalid's qualification worried me

<stainPhone> As owl is not a constraint language

satya: When talking about implementations/validators, are we talking about conforming to PROV-O or to constraints?

<GK> Hmmm... I thought we needed (interoperable) implementations of each feature rather than a "reference implementation"

<khalidBelhajjame> Just to clarify, within OWL, we will not aim to implement all constraints but only the one that are easy to express within OWL

<stainPhone> @gk +1

Luc: We are talking about conforming to constraints, validity is not in scope of WG

<jun> @paolo, my understanding that exploring OWL for the implementation is our first step, to see what can be implemented in OWL and what can't

Ivan: What we in general need is not a reference implementation of everything, but interoperable implementation for each feature

<Paolo> @ khalidBelhajjame, jun I am doing the same in the deductive programming space, but there are unknown -- hence the difficulty to committing to a deliverable

Ivan: but more general point, what do we really mean by the implementation of the particular document or the whole PROV enivronment?
... no fixed rules, but need to show world this is not just a paper exercise and can be put into practice
... not clear what this means for constraints document in relation to other documents
... will need to be clear when we present to the management

Luc: Not discussed yet, and may be an item of discussion of implementation report skeleton
... WG will have to identify features to implement in constraints document (e.g. type checking)
... Question is then how we demonstrate whether this is implemented properly
... In my own implementation, am using test cases

Ivan: For me, test cases is very much what I would like to see

Luc: Discussion of test cases are in scope of WG
... We want to come up with an agreement that we will not spend WG time defining constraints in SW technologies

<pgroth> @stefan can I get edit access to the implementation questionnaire?

<Luc> PROPOSED: The Working Group will not formalize constraints with Semantic Web technologies, but implementations of constraints by group members are welcomed

<GK> (Test cases are a good way to illustrate the consequences of specified features.)

<GK> +1

<jun> +1

<satya> seems reasonable

<dgarijo> +1

+1

<satya> +1

<TomDN> +1

<ivan> +1

<SamCoppens> +1

<khalidBelhajjame> +1

<hook> +1

<Curt> +1

<MacTed> +1

<Paul> +1

<Luc> ACCEPTED: The Working Group will not formalize constraints with Semantic Web technologies, but implementations of constraints by group members are welcomed

Public Comments

<Luc> http://lists.w3.org/Archives/Public/public-prov-comments/

Paul: We had a number of public comments, one needs cutting up, some we've started discussing
... All can be handled, but maybe we should readvertise to get more comments

<Luc> should we take the opportunity to ask for feedback as we release prov-constraints?

Luc: comments?

<pgroth> in particular on advertising?

<pgroth> +q

Luc: Now we've released constraints document, we can take the opportunity to go back to groups and individuals to get feedback on that
... then editors need to go through comments one by one and address
... Ivan, you said RDF group would give us feedback but not received yet

Ivan: Will prompt again, have discussion paper but not yet dealt with

<pgroth> talking to guus is difficult

Luc: Once constraints document out, easy to get back to them

<pgroth> sure

Implementation Report Skeleton

Luc: Now time to think about structure of this report
... identifying kinds of feedback expected from implementors

zednik: Have gone through and updated skeleton, Paul will update actual questionairres

Paul: Currently based on SKOS, If we decide we need to test cases that might change this

Luc: Have you received feedback on questionnaire?

zednik: Not yet, but good opportunity to now re-ask

<Dong> @Paul: I can help with the implementation report as well

Luc: Will make clear in minutes summary, please send an email again with link to questionnaire

<Luc> action on Paul to draft implementation report skeleton based on Stephan's questionnaires

<trackbot> Sorry, couldn't find user - on

<pgroth> +q

<Luc> action on pgroth to draft implementation report skeleton based on Stephan's questionnaires

<trackbot> Sorry, couldn't find user - on

<pgroth> ACTION: pgroth draft implementation report skeleton based on Stephan's questionnaires [recorded in http://www.w3.org/2012/09/06-prov-minutes.html#action02]

<trackbot> Created ACTION-105 - Draft implementation report skeleton based on Stephan's questionnaires [on Paul Groth - due 2012-09-13].

Paul: Ivan, have Google questionnaire, does it need to be a W3C poll?

Ivan: Questionnaire is just for activity of group, not deliverable?

Paul: Yes. Easy to collect by web poll

<pgroth> ok

Paul: Will copy part of result of poll into implementation report

Ivan: Yes, that's OK. How you collected data is your business

<pgroth> ok

Ivan: RDFa had many tests, running on site outside of W3C, but results collected into static page on W3C as implementation report

<pgroth> yes

PROV-XML

Luc: Some emails sent, issues raised, what is the progess?

zednik: Split up DM terms and encoded in XML schema, published and raised issues
... Reza has been working on implementation at Oracle
... also interested in constraints, and had a side discussion on this, decided not to pursue as part of WG but instead side-activity at Oracle or NASA for implementation report
... looked into Schematron and XSD inadequate for expressing constraints

<hook> Schematrons have been used by ISO working groups for constraints checks

zednik: For schema, implemented almost all terms and should not be too many issues raised

<Luc> @curt +1

Curt: Most of the issues are minor inconsistencies, and inclined to take from DM and make XML match, even copying non-normative language

<hook> We also tried to sync the XML examples with the PROV-DM examples.

Luc: Schema was still a bit behind DM, and issues raised are good

hook: We also tried to make XML examples match one-to-one those in DM document

<pgroth> +q to ask about prove xml html

hook: this forces us to make sure DM is in sync with XSD (and PROV-O)

Paul: Is editors draft of document on XML up to date, can be looked at? Or just XSD?

<pgroth> ok

<Curt> agreed -- not ready

zednik: XSD up to date, not the document

<Zakim> pgroth, you wanted to ask about prove xml html

Luc: When can say have version 1 of XML schema?

<Curt> I'll address the issues I've raised by this weekend

zednik: Need to ensure no open issues, but pretty close
... aim for Tuesday/Wednesday next week

<Luc> action stephan to produce version 1 of xml schema

<trackbot> Sorry, ambiguous username (more than one match) - stephan

<trackbot> Try using a different identifier, such as family name or username (eg. estephan, szednik)

<Luc> action zednik to produce version 1 of xml schema

<trackbot> Created ACTION-106 - Produce version 1 of xml schema [on Stephan Zednik - due 2012-09-13].

Luc: Can come back to at next week's telecon
... PROV-O spent long time looking at tools to convert their schema into HTML document

<pgroth> +q

Paul: On the editors draft, there is an XSD/HTML document, who created?

Luc: I created

Paul: We said that for the XML, we wouldnt worry about the verbiage around the schema, right? Just need schema

Luc: Want to minimise work, but need something readable. Can extract lots of text from DM document

<pgroth> sure

Luc: Paul and I will identify editor(s) for this document

Curt: Considered adding links to anchors in XSD HTML document to link to DM, rather than copy

Luc: XML schema editors, are there specific issues you want to discuss?

<zednik_> sorry, disconnected

zednik: Issue on representing prov:type
... Type of prov:type is defined as a value (effectively a literal)
... Consequence is that 42 is a prov:type, etc.
... Is that what we want?

<Curt> stephan: could discuss prov:Agent vs. prov:Person too? (or is that resolved?)

zednik: In XSD would be xsd:anySimpleType, not constraints

<pgroth> +q

<zednik_> http://www.w3.org/2011/prov/track/issues/493

Paul: QName would improve interop, but are there examples where you don't want than (Java package names?)

Luc: DM contains examples of types that are just strings

<zednik_> @curt raise as an issue

Luc: Will not resolve issue now, but is important as has potential impact on DM, so suggest continue over email and will revisit

<TomDN> bye

<ivan> bye everyone

<jun> bye

<Dong> Thanks, bye

<SamCoppens> bye

<pgroth> congrats - another last call

<pgroth> bye

<zednik_> bye

<khalidBelhajjame> bye

<Luc> trackbot, end telcon

<trackbot> Sorry, Luc, I don't understand 'trackbot, end telcon '. Please refer to http://www.w3.org/2005/06/tracker/irc for help

<MacTed> trackbot, end meeting

This is scribe.perl Revision: 1.136  of Date: 2011/05/12 12:01:43  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/Paolo/Paul/
Succeeded: s/Paolo/Paul/
No ScribeNick specified.  Guessing ScribeNick: smiles
Found Scribe: simon miles
Default Present: smiles, Luc, Curt_Tilmes, pgroth, +329331aabb, TomDN, khalidBelhajjame, Satya_Sahoo, +1.818.731.aacc, samcoppens, Ivan, GK, MacTed, stainPhone, Paolo, dgarijo, jun
Present: smiles Luc Curt_Tilmes pgroth +329331aabb TomDN khalidBelhajjame Satya_Sahoo +1.818.731.aacc samcoppens Ivan GK MacTed stainPhone Paolo dgarijo jun
Regrets: James Cheney
Agenda: http://www.w3.org/2011/prov/wiki/Meetings:Telecon2012.09.06
Found Date: 06 Sep 2012
Guessing minutes URL: http://www.w3.org/2012/09/06-prov-minutes.html
People with action items: paul pgroth