Provenance Working Group Teleconference

29 Nov 2012


See also: IRC log


Curt_Tilmes, pgroth, Luc, MacTed, smiles, [IPcaller], +1.315.330.aabb, SamCoppens, GK, dgarijo, TomDN, jcheney, Satya_Sahoo, jun, khalidBelhajjame, stain
Paolo_Missier, Hua_Hook
Paul Groth
Curt Tilmes


<trackbot> Date: 29 November 2012

<pgroth> Scribe: Curt Tilmes


<pgroth> http://www.w3.org/2011/prov/meeting/2012-11-22

<pgroth> Minutes of Nov. 22, 2012

0 (not present)

<smiles> +1

<pgroth> Accepted: Minutes of Nov. 22, 2012 telcon

pgroth: open action items


pgroth: action 153 should be complete -- will double check
... request for transition to CR sent out


<Luc> it's for december

tlebo: 116 -- haven't done yet, will soon

<pgroth> close Action-128

<trackbot> ACTION-128 Add a statement on informative and normative in prov-o closed

<pgroth> close action-129

<trackbot> ACTION-129 Editor check prov-o closed

pgroth: action 122 still open, complete by next week

<pgroth> close action-133

<trackbot> ACTION-133 Draft a first one page overview closed

SamCoppens: action 134, timetable for prov-dictionary has been proposed

<Luc> http://dvcs.w3.org/hg/prov/raw-file/default/model/prov-dictionary.html

<Luc> we said we would do an "internal" release

<pgroth> ace Luc

pgroth: after dec 14, there is a moratorium until after xmas -- then we can release prov-dict FPWD

<pgroth> close action-134

<trackbot> ACTION-134 draft a timetable for prov-dictionary for the next teleconference closed

<pgroth> close action-135

<trackbot> ACTION-135 Create a mention of document closed

<Luc> @Sam, did you see the current draft for dictionary

<TomDN> yes, we did, thanks Luc

pgroth: 3 actions on stephan on questionnaire will discuss later

<Luc> @Tom, great!

<TomDN> Good place to start :)


<pgroth> http://lists.w3.org/Archives/Public/public-prov-wg/2012Nov/0295.html

<Luc> @Tom, you may want to move this document into a separate directory in hg

<TomDN> @Luc: ok, will do

<Luc> PROV WG is in good company, HTML5 went CR at same time as us


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

pgroth: special thanks to editors of standards and Jun for org ont feedback

<GK> Jun say's she's received offline ack of org ontology feedback

pgroth: small PROV-Overview for intro to PROV
... feedback from khalid, paolo, one other, minor changes, concerns about picture, all approved going to FPWD

<smiles> I read it through and seemed good

<dgarijo> @curt: the other was me

<pgroth> proposed: release of PROV-Overview as a FPWD

<tlebo> +1

<dgarijo> +1

<smiles> +1


<zednik> +1

<Dong> +1

<TomDN> +1

<jun> +1

<jcheney> +1

<SamCoppens> +1

<GK> +1

<satya> +1

<pgroth> accepted: release of PROV-Overview as a FPWD

pgroth: will add recommended changes next week
... timetable to release discussed last week

<Luc> Luc summarized the timetable to CR publication. CR documents must be ready by noon GMT on November 26th, for final check by chairs and Ivan. Announcement on Nov 27th. CR transition teleconference on Dec 6, and planned publication on Dec 11th. Notes to be ready for staging on Dec 5th. Vote for release of notes on Nov 29th. Questionnaires must be ready by Dec 11th.

Luc: many documents being released, we would like to go by the 4th noon

<dgarijo> very clear

<GK> Plan fopr PROV-AQ?

<smiles> Yes

<GK> OK.

pgroth: postponing PROV-AQ for now for review


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

Luc: extracted text from mention from PROV-DM, N, CONSTRAINTS, O, XML and added the intro, editted into a note
... paolo reviewed and recommended release

<smiles> I skimmed it, and it looked fine to me

Luc: Curt pointed out a typo, recommended releasse

<Zakim> GK, you wanted to say I've reviewed links docs, have some issues with a couple of points but not fatal

GK: have read the document, I have significant comments I will write up, but ok to release as a draft

<Zakim> jcheney, you wanted to ask a pedantic question

GK: will write up and send after this call

jcheney: as a note, it should only have informative references

Luc: right, will fix

<pgroth> proposed: Release of PROV-Links as a FPWD

<smiles> +1

<dgarijo> +1


<tlebo> +1

<stain> +1

<zednik> +1

<GK> +1

<Dong> +1

<satya> +1

<TomDN> +1

<jcheney> +1

<pgroth> accepted: release of PROV-Links as a FPWD


<stain> quick Q: would the XML bit work as an extension to PROV-XML when it is using the same namespace?

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

dgarijo: reviews from craig, luc, simon
... haven't discussed with kai (?) yet
... will contact him and discuss and ready by next tuesday

<SamCoppens1> +1 fro release of PROV-links (just got disconnected when voting)

<GK> @stian hopefully the PROV-XML spec will have extension points that can be "colonized" by the links/mention elements. Haven't actually looked, though.

pgroth: are the changes blockers to FPWD?

<dgarijo> YES, I'll do that today

<dgarijo> thanks, Simon

dgarijo: no, some references need to fix, actual content no blockers

<GK> @smiles … you can include Luc's references quite easily - I've done that for PROV-AQ; it's quite easy

<dgarijo> +q

Luc: the example for figure 2 was valid, not invalid as document indicated -- must be fixed to clarify constraints

dgarijo: correct, I need to fix that, will work on that before week-end

pgroth: add a box noting some sections are still under discussion prior to draft
... but should fix the constraint now

<pgroth> proposed: Release of PROV-DC as a FPWD

<smiles> +1

<dgarijo> +1

<TomDN> +1

<tlebo> +1

<satya> +1

<jcheney> +1

<SamCoppens1> +1

<zednik> +1

<stain> +1


<pgroth> accepted: Release of PROV-DC as a FPWD

pgroth: all notes should be ready before next tuesday

Implementation Questionnaire

<zednik> https://www.w3.org/2002/09/wbs/99999/prov-implementation-survey

zednik: looked at WBS to host survey
... have implemented implementation questionnaire -- need feedback

Luc: can 1 individual submit multiple implementation reports

zednik: you can do it by logging in with multiple email addresses, otherwise it retrieves your old report

(that's a kludge)

<stain> perhaps you can do the + trick for making additional email addresses.. for instance: stian@s11.no, stian+anotherone@s11.no, stian+blah@s11.no

pgroth: some questions are awkward to ask with WBS
... should we pursue with WBS given the issues?

<Luc> the good think is that we see all questions that are being asked

<Luc> the good thing is that we see all questions that are being asked

Luc: I understand the issues, they're awkward, but acceptable -- what do you recommend?

<Luc> +1

pgroth: I think WBS is a bit better than google docs, can return and continue editting
... most respondents will already have a W3C account, tallying is nice, on W3C site is good

<tlebo> returning to the survey is +100. It's reassuring for a tentative implementer.

pgroth: a little annoying with multiple implementations, (e.g. vocabulary + application), but those
... are already separate questionnaires

<stain> Feature Coverage" should have some more details about using the +++ ratings for different feature level. Add "Indicate covered features by selecting one of the following in the dropdown lists"

pgroth: could easily work around the multiple email issue

zednik: very nice to be able to come back to it, email in responses works too, a lot of advantages, some awkwardness in questions, but we can cope with it
... reporting is nice, haven't seen in detail yet

<stain> oh hang on, I can fix it myself. Done!

<tlebo> @stian ;-)

<zednik> text version of the questionnaire: https://www.w3.org/2002/09/wbs/99999/prov-implementation-survey/text

pgroth: it may be easier in putting together the implementation report -- can refer to these results

<Luc> Stephan and Paul, thanks for doing this!

Constraints Test cases

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

<Dong> http://dvcs.w3.org/hg/prov/raw-file/default/testcases/process.html

Dong: lots of test cases, examples from documents, a table in the doc describes coverage by test cases
... can do gap analysis to figure out which we still need

<Luc> @zednik, is it possible to change the URL and have them point to the date urls for the CR documents? also, can we have links for everything?

Dong: I've done categorization, but could use help with that

pgroth: several comments online about doc organization

<Luc> @Dong, same comment as for Stephan, we need to use urls to CR documents

Dong: Haven't revised doc yet, plan is to restructure doc for implementors
... will do that very soon

Luc: table for coverage, is it complete?

Dong: we cover all constraints, but some only pass, not fail yet

<zednik> @Luc, is your question concerning the links in the Feature Coverage question? If so, yes. I will need to determine the best link anchor for the Person, Organization, etc.

stain: it says the implementation can use the representation to test the constraint -- would be good to know which representation was used for each case

Dong: good idea -- would be good to have representations equivalent, but in some cases that may not be perfect. That is doable, we'll add that to the cases and extract into the repot

pgroth: have we decided on the test case naming convention?

Dong: arbitrary name + pass/fail + constraint number

pgroth: other two tables from PROV-O examples / test cases, those don't have constraint numbers

Dong: yes, they are complex and cover multiple constraints, they aren't unit tests, but they should still pass the validator

Luc: the links to the documents should point to the dated URLs to the other docs (questionnaire document too)

Dong: will do for release

Luc: we want them prior to the telecon

Dong: will do

Other Documents

smiles: working on other docs, will update primer this week-end

pgroth: XML status? voted to release as FPWD already, all staged and ready to go?

<zednik> discconnected from audio, calling back in

<zednik> back

Curt: I think the changes were all made and everything is staged

zednik: it is staged now, the CR links don't resolve

Luc: Curt raised a question about how to handle XML namespace


<Dong> Sorry, I have to go.

<pgroth> \

pgroth: we merge the various namespaces together for PROV-O, can we do something similar for PROV-XML?

<Dong> Bye all.

Luc: If you look at the ontology, we have mentionOf defined in a separate document, but it still uses the prov: namespace

<pgroth> yes

<stain> http://dvcs.w3.org/hg/prov/raw-file/default/xml/releases/WD-prov-xml-20121211/Overview.html uses <xs:any namespace="##other"/> -- so here you can't extend this using the same namespace

<stain> you would have to use <xs:any /> instead

Luc: a separate ontology defines mention, but the subject/object are prov entities, is that ok? I don't think we can do that with XML.

pgroth: with PROV-O, you can load both files into your reasoner

Luc: will this work with XML?

stain: I don't think that will work with XML

Curt: could manually maintain separate complete XSDs, one with just core, one with core+extensions

Luc: this can affect overall extensibility approach

<stain> @Curt two schema versions could be tricky - if I import a schema which is using the 'core' schema, then I can't (easily!) use the 'extended' schema.

Luc: I don't think we can use the same approach to extend XML we use for OWL

pgroth: XML is a note, so we could just put all the notes extensions together
... it isn't a specification

Luc: that may give more weight to dictionaries, mention, etc. than we intend to have

<GK> "Can't do it with XML schema" sounds wrong to me. I'm not expert on XML, but I don't see what breaks if one has multiple schema documents that define different aspects of an XML structure, creating some elements in a common namespace. The main thing that I see is that the core schema MUST have suitable extension points.

<Luc> yes you can import

pgroth: Can we check to see if we can technically handle it by combining them?

<pgroth> no

zednik: dictionary/mention will have notes, could we use separate namespaces for them?

pgroth: wouldn't be consistent with other docs, want to use just one namespace

<satya> * sorry have to leave now

<satya> bye

I'd like to experiment a bit before making a decision

<stain> I can try to experiment as well

Luc: add a note in the document that the namespace may change

<tlebo> I still need to play with it, too.

<TomDN> gotta go, bye

<stain> but there is a reason why XML community has moved away from XSD and over to lightweight, less strict formats like RelaxNG

Luc: there may not be a technical solution, we want the schema to validate

<pgroth> all of them

there is no overlap

but if you use mentionof and validate against core without links, it should fail

<Luc> xml does not work with uri, per se

<stain> http://dvcs.w3.org/hg/prov/raw-file/default/xml/releases/WD-prov-xml-20121211/Overview.html shows lots of overlap and non-overlap, for instance <prov:wasGeneratedBy> (overlaps with prov:wasGeneratedBy) and <prov:time> (not overlapping with prov:atTime)

Address in next revision past FPWD

<Luc> I also would like to check that this is implementable

<GK> I would assume we can have multiple schemas defining terms in a common namespace, and do some background checking with XML experts.

It's just a note, and its just FPWD, people know it might change

<tlebo> +1 curt

<stain> I'm going now, but I'll check. Have we got an ISSUE for this (common email thread) ?

<dgarijo> bbye

<tlebo> bye bye!

<SamCoppens1> bye

I'll make an issue

<pgroth> curt, i'll sort out the minutes

<pgroth> trackbot, end telcon

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.137 (CVS log)
$Date: 2012/11/29 17:09:46 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.137  of Date: 2012/09/20 20:19:01  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

No ScribeNick specified.  Guessing ScribeNick: Curt
Found Scribe: Curt Tilmes
Default Present: Curt_Tilmes, pgroth, Luc, MacTed, smiles, [IPcaller], +1.315.330.aabb, SamCoppens, GK, dgarijo, TomDN, jcheney, Satya_Sahoo, jun, khalidBelhajjame, stain
Present: Curt_Tilmes pgroth Luc MacTed smiles [IPcaller] +1.315.330.aabb SamCoppens GK dgarijo TomDN jcheney Satya_Sahoo jun khalidBelhajjame stain
Regrets: Paolo_Missier Hua_Hook
Agenda: http://www.w3.org/2011/prov/wiki/Meetings:Telecon2012.11.29
Found Date: 29 Nov 2012
Guessing minutes URL: http://www.w3.org/2012/11/29-prov-minutes.html
People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.

[End of scribe.perl diagnostic output]