Provenance Working Group Teleconference

01 Nov 2012


See also: IRC log


Luc, [IPcaller], +1.315.330.aaaa, tlebo, MacTed, +1.818.731.aabb, smiles, GK1, +1.661.382.aacc, CraigTrim, Satya_Sahoo, +1.818.731.aadd
Paolo_Missier, Paul_Groth, Daniel_Garijo, Curt_Tilmes
Luc Moreau
tlebo, smiles


<trackbot> Date: 01 November 2012


<tlebo> I'll scribe

<Luc> scribe: tlebo

luc: good afternoon.
... trying to close issues leading to F2F

<Luc> Proposed: approve Minutes of the October 25, 2012 Telecon

admin issues

<Paolo> +1

<jun> +1

proposed: approve last week's minutes

<zednik> +1


<Luc> Accepted: Minutes of the October 25, 2012 Telecon

accepted: minutes approved.

outstanding actions

tim has not done the cross referencing yet.

tim: the cross referencing is editorial

Stian's 118 can be closed

121 - Tim's action on subclassing.

exit criteria

luc: not that we do NOT have a telecon next Thur., since F2F is Friday.

<Luc> http://lists.w3.org/Archives/Public/public-prov-wg/2012Oct/0305.html

luc: paul sent proposal to accept criteria.
... all responses were positive.
... can we record approval? any objections?

<Luc> accepted: CR Exit Criteria defined at http://www.w3.org/2011/prov/wiki/ProvCRExitCriteria are approved

luc: questionaire. Is it ready to submit to W3C management team?

zednik: trying to fix Paul's last issue on it. Implementation consuming external construct. Will address them today and get confirmation, will send it out.
... will try to finish it today.

luc: will try to have an email vote ahead of the F2F

Prov-o Issues

<Luc> http://www.w3.org/2011/prov/track/products/10

<Luc> http://www.w3.org/2011/prov/track/products/3

349 can be closed

479 can be clsoed b/c we are not using Trig.

491 was Stian handling it.

Luc: Tim to draft response and contact reviewer.

Tim: I''l hunt down everthing there...

<Luc> Rephrasing in prov-dm: http://dvcs.w3.org/hg/prov/rev/363ce30cec66

Tim: 552 is the only big issue, and I'll start it after this call.

luc: new definitions include "relation"
... change of text, but not of the ontology.

tlebo: my approach will be to change the text.

luc: draft response, group approves, convey change to commenter, and have acknowledgment before F2F.
... can we vote, so that the acknowledgement can be sent to the commenter on Monday.

<Luc> http://lists.w3.org/Archives/Public/public-prov-comments/2012Nov/0000.html

tim: my impression was that it is just an announcement.

luc: they are using wasInfluencedBy, but the rec suggest to use a subproperty.

tlebo: yes, we can look at their use and suggest to use a subproperty.

PROV-Constraints issues

<Luc> http://www.w3.org/2011/prov/track/products/12

jcheney: isssues were raised from the feedback last week.

<jcheney> http://www.w3.org/2011/prov/wiki/ResponsesToPublicComments#PROV-CONSTRAINTS.28Under_review.29

jcheney: haven't heard any feedback beyond minor typos.
... unless objection, would like to send the responses to the commenters.
... two more issues that are leftover from last call. internal.

<jcheney> http://www.w3.org/2011/prov/track/issues/559

jcheney: made propsoed change and considers it done. Do we need to send a formal response?

<jcheney> http://www.w3.org/2011/prov/track/issues/522

jcheney: the comment spread from issue on DM

luc: we need to approve responses.

<Luc> PROPOSED: The group endorses the responses to issue-556, issue-576, issue-582, issue-586, issue-587, issue-588, issue-584, issue-579, issue-585, issue-583, issue-580, issue-577, issue-578, issue-581

jcheney: the constraints doc is not addressing how to implement them in OWL etc.


<GK> 0 (not reviewed or followed discussion)

<smiles> +1 (though I admit to not having read them all thoroughly...)

<khalidBelhajjame> +1

<zednik> +1

<Luc> accepted: The group endorses the responses to issue-556, issue-576, issue-582, issue-586, issue-587, issue-588, issue-584, issue-579, issue-585, issue-583, issue-580, issue-577, issue-578, issue-581

<jun> +0 (haven't been able to review)

jcheney: I'll contact the commenters instead of Paul

luc: any other changes?

jcheney: all that are necessary are done.
... some quick issues to get feedback:

<jcheney> http://www.w3.org/2011/prov/track/issues/559

jcheney: 559 was a side effect, do we need to make a formal response?

luc: was an internal comment.

<jcheney> http://www.w3.org/2011/prov/track/issues/575

jcheney: links between identifier fields.

<jcheney> http://www.w3.org/TR/prov-dm/#entity.id

luc: i'll do it after the call
... in responses, can you ask them to acknowledge and whether they are satisfied?

prov-n issues

<Luc> http://www.w3.org/2011/prov/track/products/11

luc: only 2 issues
... ivan was fine, but we want to confirm...
... namespace declarations and bundles. bundles do not "inherit" prefixes.

@gk probably would have stated it better...

gk: same point as Tim, you were responding to URIs of entities

luc: do we need to revise?

<jcheney> Currently says: "The scope of a namespace declaration directly occurring in a document is the document itself, excluding the bundles it may contain."

luc: one benefit is that bundles are totally distinct.

<jcheney> This forces you to repeat prefixes in common cases, including many natural-looking examples in prov-dm

<Luc> scribe: smiles

<GK> I was saying that I thought the current design is maybe sub-optimal for human use, but not fatally flawed.

jcheney: Missed original change to excluding namespaces in top-level bundle; changing the scope to include the namespaces from top-level might not break anything?

Luc: Would need to get advice on whether this is only an editorial change
... How strongly do you feel about this?

jcheney: There are situations where it is useful to think of bundles as independent and copy-paste without worrying about top-level, but using namespace without repeating declaration throughout a document is common
... Natural expectation is that if you state something then embedded scopes would include that thing

<tlebo> FWIW, Turtle and Trig permits the prefixes to be defined once and used in any of many "bundle".

Luc: Are we willing to go ahead with change even if it is not just editoral, i.e. back to LC

<tlebo> This is not worth going back to Last Call for.

jcheney: It is something that people will anyway notice and complain about in implementation

<GK> Maybe ask Ivan if this would need to redo last call. It seems a small change, albeit technical.

<ivan> if I may, just on IRC...

<ivan> this is not a design change

<Luc> thanks ivan

<ivan> this is a change in a syntactic sugar

<Luc> ok

<ivan> it would influence implementations, but that is not the issue for LC

Luc: Given the above, it would be good to know the WG's preference

<Luc> show of hand: allow scope of prefix declarations

<tlebo> +1 because that's how it's done in Turtle/Trig, -1 b/c PROV-N is for examples in Recs :-)

satya: Please clarify question

<GK> +0.5 to top-level namespaces visible in contained bundles in PROV-N (next question: can they be redefined in inner bundles?)

<jcheney> See namespace scope rules in http://dvcs.w3.org/hg/prov/raw-file/default/model/prov-n.html#expression-NamespaceDeclaration

Luc: Currently in PROV-N, when declaring prefix at top level and want to re-use in nested bundle, have to redeclare prefix in bundle

<jcheney> Currently, this is illegal:

<jcheney> document prefix ex <http://example.org/> bundle ex:b1 entity(ex:e1, ...someattr1...) entity(ex:e1, ...someattr2...) endBundle endDocument

<jcheney> because ex is not re-declared inside the bundle

Luc: @GK, yes, prefixes could be redefined in inner bundles

<jcheney> I http://dvcs.w3.org/hg/prov/raw-file/default/model/prov-n.html#expression-NamespaceDeclaration

jcheney: Suggest just changing sentence in PROV-N from "excluding the bundles it may contain" to "including..."

<tlebo> reusing throughout and allowing redefinition in inner bundles would make it much more useful for humans to read and write.

jcheney: People have been writing examples that assume the proposed change without realising

<GK> Seems to me easier to change the text than the examples

<Luc> proposed: change scope of prefix declaration so that it includes bundles

<satya> +1 to change text


<GK> +1

<jcheney> +1

<ivan> +1

<jun> +1

<tlebo> +.5

<zednik> +1

<Luc> accepted: change scope of prefix declaration so that it includes bundles

Luc: Next issue: MIME types
... @GK, did you look at changes implemented?

GK: No

Luc: Can you prioritise issue 573?
... Everything else covered in PROV-N aside from two issues above

PROV-DM issues

<Luc> http://www.w3.org/2011/prov/track/products/2

Luc: Two issues waiting for confirmation from reviewers on changes
... and issue 475 on mention
... every other change implemented and logged

<Luc> http://lists.w3.org/Archives/Public/public-prov-wg/2012Oct/0321.html

Luc: Regarding issue 475, did not get response on issue
... GK suggested dropping this feature at risk
... Conclusion of my response is that we want feedback from implementers
... As we exit CR phase, can look at suitability
... Any feedback now?

<tlebo> +1 to keep it in, marked as at risk.

<Luc> http://www.w3.org/2011/prov/wiki/ResponsesToPublicComments#ISSUE-475_.28Mention.29

Luc: Is the group satisfied with this answer, and to vote now?
... or vote by email?

+1 to vote by email

<zednik> I like vote by email

<jcheney> +1 email

<tlebo> vote by email seems to be the resounding response. I know I could use a bit more time to work through it entirely.

Luc: Will send an email with a deadline


zednik: We have made significant changes to XML note, including table of features mapping PROV-DM to XSD, plus examples with excerpts of schema

<zednik> http://dvcs.w3.org/hg/prov/raw-file/1816251fd592/xml/prov-xml.html

zednik: Now looking for feedback on the HTML

Luc: Thanks. There is an opportunity to release first WD of this doc with CRs of other specs. Will it be ready?
... Next week we would take a vote, discuss with W3C management to publish CRs, so two weeks for XML document before publishing

zednik: Probably get ready in two weeks, but need to address WG comments, ensure there is no gaps (e.g. bundles)

Luc: Would be nice to have review taking place during week

<jcheney> sure

Luc: Who is willing to review PROV-XML document?

<tlebo> one what timeframe?

<tlebo> thanks. Then no :-)

Luc: Feedback in time for F2F
... Question for reviewers: Can we release this document as a FPWD
... Intent to include XML in primer. Feasible to include for release?

smiles: For primer, if given XML fragments, then quick job to add

zednik: Possibly, but need to look at examples in primer

<GK> @luc - the text/provenance-notation MIME registration revisions look fine to me at http://dvcs.w3.org/hg/prov/raw-file/default/model/prov-n.html#media-type

<hook> might need to check latest fragments for bundles per recent updates

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.137 (CVS log)
$Date: 2012/11/01 16:07:27 $

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)

Found Scribe: tlebo
Inferring ScribeNick: tlebo
Found Scribe: smiles
Inferring ScribeNick: smiles
Scribes: tlebo, smiles
ScribeNicks: tlebo, smiles
Default Present: Luc, [IPcaller], +1.315.330.aaaa, tlebo, MacTed, +1.818.731.aabb, smiles, GK1, +1.661.382.aacc, CraigTrim, Satya_Sahoo, +1.818.731.aadd
Present: Luc [IPcaller] +1.315.330.aaaa tlebo MacTed +1.818.731.aabb smiles GK1 +1.661.382.aacc CraigTrim Satya_Sahoo +1.818.731.aadd
Regrets: Paolo_Missier Paul_Groth Daniel_Garijo Curt_Tilmes
Agenda: http://www.w3.org/2011/prov/wiki/Meetings:Telecon2012.11.01
Found Date: 01 Nov 2012
Guessing minutes URL: http://www.w3.org/2012/11/01-prov-minutes.html
People with action items: 

[End of scribe.perl diagnostic output]