Provenance Working Group Teleconference -- 06 Dec 2012

<trackbot> Date: 06 December 2012

<Luc> trackbot, start telcon

<trackbot> Meeting: Provenance Working Group Teleconference

<trackbot> Date: 06 December 2012

<pgroth> you can add Stephan Zednik

<pgroth> :-(

<Luc> i need a scribe please

<pgroth> i can if no one else

I can scribe

<tlebo> sorry, I might get pulled away during the call.

<scribe> Scribe: smiles

admin

<Luc> proposed: to approve the minutes of last week's teleconference

<tlebo> +1

<ivan> +1

<jun> +1

<Dong> +1

<stain> 0: not read (the link?)

<Luc> resolved: the minutes of last week's teleconference

<stain> http://www.w3.org/2011/prov/meeting/2012-11-29

<pgroth> no

Luc: Actions open, Tim's for later, Luc's not done, Stefan's will be covered later

congratulations

<Luc> http://lists.w3.org/Archives/Public/public-prov-wg/2012Dec/0017.html

Luc: Transition request approved for four new notes: overview, links, dc mapping, and xml
... congratulations to editors

ivan: Checked with Dennis, who will check the release documents at the weekend, no problem with publication

Luc: May not be available at all times on Monday, Tuesday, to deal with problems

ivan: Publication will usually happen Tuesday morning Boston time

<pgroth> do we need a blog post?

ivan: need blurb for W3C homepage
... once publication is out, blog post would be good

pgroth: I can prepare a blog post, ready for Tuesday

<TomDN> (sorry I'm late, other meeting ran late)

ivan: Whatever goes to the W3C homepage, I repeat it on the activity blog
... so better if one of chairs write this text

<pgroth> yes

ivan: could be separate blog post with more details, which can also go to activity blog

pgroth: If something comes up regarding publication, we can directly edit the HTML?

ivan: yes

Getting Things Ready for Implementation

Luc: There are outstanding questions regarding questionnaire

pgroth: Questionnaire has not changed much since last week, couple of things Stefan needs to do

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

pgroth: he has copied three questionnaires into one and added question about PROV constraints
... he still needs to create the other questionnaires
... for different kinds of products: implementation, vocabulary

Luc: questionnaire ready for management telecon tomorrow?
... does not have all links to concepts in CR documents

pgroth: I can update this questionnaire with links

Luc: currently links point to sections, not definitions
... Any questions on questionnaires?

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

Dong: (regarding test cases) Above are the current documents
... Issues raised by internal reviewers addressed, almost ready to go
... except that test cases missing for some constraints

Luc: The test cases do not need to be frozen by 11th, need to be good but as others implement we may come up with other examples

Dong: Good. In current version of test case document, we ask people to email results back to us, maybe be better to use questionnaire

<pgroth> that's a good idea…

Dong: I will see if the questionnaire allows us to have text fields they can copy these into instead

pgroth: Could be possible, but would have to go through test case numbers in WBS

<Luc> if you look at the current questionnaire, you have free text boxes

Dong: Do not to intend to list all test cases as options, but just provide big text fields
... need volunteer reviewers for test case document

Luc: There are a couple of lines with no test cases listed, but for which we do have test cases

Dong: Yes, some to add to test suite

<pgroth> +10

Luc: Thanks to Dong for all the work: 240 different tests, for each of 3 serialisations
... Now need reviewers for test cases

pgroth: Can I do it while I implement?

Luc: That's fine, though ideally you should not come back on 31 Jan with problem
... will also ask James C to review

<Luc> action jcheney to review the test cases

<trackbot> Created ACTION-154 - Review the test cases [on James Cheney - due 2012-12-13].

<jun> I am in a similar position as Paul

<jun> I will review them while implementing them

<pgroth> yes

<Luc> action paolo to review the test cases

<trackbot> Created ACTION-155 - Review the test cases [on Paolo Missier - due 2012-12-13].

<tlebo> interested, but not willing to commit :-)

Luc: Anything else regarding implementation?

ivan: We are ready for tomorrow
... They will review the call sent to the chairs and links on that page

prov-xml

<pgroth> stain

<stain> http://www.w3.org/2011/prov/wiki/ProvXMLNamespaces

stain: Have investigated namespace issue
... have looked at how to deal with extensions, include or redefine namespaces
... propose use of substitution groups to allow for extensions
... namespace overlap issues become clean, as can reuse same namespace and use xsd:includes
... need to say something about how others do extensions, manage adding complex type elements with attributes, etc.

Luc: do not have editors on call, so cannot make decisions, but can pass Stian's wiki page to initiate debate

<Luc> action zednik to review http://www.w3.org/2011/prov/wiki/ProvXMLNamespaces and come back with a response

<trackbot> Created ACTION-156 - Review http://www.w3.org/2011/prov/wiki/ProvXMLNamespaces and come back with a response [on Stephan Zednik - due 2012-12-13].

Luc: Can discuss at next telecon

pgroth: Would be good to get this settled so we can get on with improving the namespace page

stain: Last week Tim raised some concerns about same namespaces in OWL and XML

tlebo: Haven't yet considered, will do for next week

Luc: Thanks Stian for work, we will now digest

<stain> .. specially if we need to leave hooks in the core.xsd

Luc: Request that current set of issues on XML should be addressed by next release: early February

<stain> and are there any overlaps with elements and predicates?

ivan: Just checked: OWL 2 has an XML serialisation and uses same namespace as the RDF ontology

<tlebo> Thanks, @ivan

prov-aq

GK: <alien sounds>

<GK> Ity seedms my outgoping audiop is poor

<GK> I'm on an ADSL line, so maybe bandwidth starved

<Luc> ... alien text too ;-)

<pgroth> I can do it

<pgroth> and correct me if I'm wrong

<GK> I have some notes I can paste in:\

<GK> PROV-AQ - draft at: http://dvcs.w3.org/hg/prov/raw-file/default/paq/prov-aq.html

<GK> Has been updated:

<GK> Change 'provenance information' to 'provenance description' (ISSUE 601); extensive editorial changes should be reviewed

<GK> Add note about service description and LDP consideration thereof

<GK> Add non-commital paragraph about accessing provenance bundles

<GK> Update forward provenance (pingback) in response to comments; fix text around VoID example

<GK> Added specification for pingback link header (ISSUE 600)

<GK> Expanded discussion of provenance service discovery to include prov:hasProvenanceService

<GK> Remove speculative non-specification text to be covered in FAQ (ISSUES 426, 598)

<GK> Added section with table of URIs and what they dereference to (ISSUE 424)

<GK> Make treatment of direct retrieval and service for provenance access more equally visible (ISSUE 422)

<GK> Point out that provenance services can accept paraneters other than just 'target' (ISSUE 420)

<GK> Added definitions for accessung and locating (ISSUE 417)

<GK> Added section on Link: headers and content negotiation (ISSUE 416)

<GK> Added icon to distibguish external links (ISSUE 400)

<GK> Updsate and cross-link table of prov: URIs defined

<GK> Changed link relations to URIs; removed IANA considerations section

<GK> Hyperlink concept definitions to themselves (per request from Tim 2012-06-05)

<GK> Update security considerations with note about use of provenance as part of audit/enforcement mechanism

<GK> Update list of PROV documents, copied from PROV-DM

<GK> Cut back on verbiage in sect 3,4 rbitrary data

<GK> Revised section 3 descriptions in terms of prodcuers and consumers

<GK> Updated security considerations to mention audit

<GK> Add TODO for producer/consumer roles, add note about multiple links, add reference to RFC3986 for %-escaping

<GK> Update security considerations, note about non-RDF service desription, PROV-O link, acknowledgements

<GK> The aboce is from the HG log

<stain> intergalactic issue

<GK> So where we are:

<GK> - Most issues have been addressed, pending review

pgroth: Gone through all the issues, and GK has been addressing them (see changes above)

<GK> - http://www.w3.org/2011/prov/track/products/5

<GK> - Have review comments from Jun to fold in

<GK> - update of OWL document [fragment] is still to do.

<GK> - still need to review for use of RFC2119 language (even though non-REC, is still a spec)

pgroth: Still more to do, but should be done by next week (GK?)

<GK> Most;y, buy some outstandinmg issues

<GK> Recent issues:

<GK> - SPARQL endpoint discovery

<GK> - raised issue with LDP about service description format (http://www.w3.org/2011/prov/track/issues/425). I p[ropose to stiock for now with RDF, as that avoids defining yet another data format, and maybe update later to follow LDP recommendations

<GK> -- issue of RDF having multiple MIME types (cf. email from Eric Wilde aka @dret)

pgroth: After that, need to read over

Luc: Group will not see doc before Christmas?

<GK> I think there are issues that can be revioewed; e.g. forward provenance

pgroth: Yes, there seems no point in doing so, as will not be able to give proper attention

<GK> There';s some discussion about how to handle SPARQL endpoint discovery

<pgroth> yeah, first - second week of january

Luc: Early Jan, should have internal review of document

<pgroth> will this work...

<Zakim> GK, you wanted to say there are a couple of specific issues to discuss

<pgroth> no

<GK> I'

<stain> it was good as you said "hang on"

<stain> perhaps GK really IS in a different galaxy..!

<GK> 1. SP{ARQL endpoint discovery - I propose a new link relation

<GK> 2. service descrtiption format

<GK> 3. pingback; aka "forward provenance"

<GK> These are issuyes where there are substantial changes/new material

<pgroth> sounds good

Luc: Use mailing list to point out issues to discuss (as GK lists above)

<pgroth> perfect

<GK> Acxk. Siounds good.

<GK> Ack.

prov-sem

<pgroth> thanks graham

<pgroth> !

<Luc> http://lists.w3.org/Archives/Public/public-prov-wg/2012Dec/0022.html

<Luc> http://www.w3.org/2011/prov/wiki/FormalSemanticsED

Luc: Had apologies from James, but sent brief update: has been converting the rules into LaTeX/HTML
... He asks for (1) people to go through document and check for typos in converting rules, (2) people to help convert into W3C Note style

ivan: Could contact staff contact in MathML group, as they may have experience in publishing formal notation/maths
... Can forward James' mail

Luc: Anyone else willing to go through the document? Tom?

<TomDN> yes

<Luc> tom?

<GK> I'll look if I have time

pgroth: Is it at a state where we would want to show it to others?

Luc: No I don't think so

<TomDN> I can take a look at it as reviewer, but cannot actively help with writing

prov-dictionary

Luc: Not formal review yet, see James' mail for what to check

<pgroth> is that too late?

TomDN: Schedule will be as said last week, next version out by next Thursday for internal review

Luc: Timing fine for FPWD with proposed recommendations
... AOB?

<tlebo> thanks, bye!

thanks

<TomDN> bye

<khalidBelhajjame> bye

<pgroth> sure

<Dong> bye

<GK> Yay! +10 mins

<jun> bye

<GK> Bye

<Luc> trackbot, end telcon

- DRAFT -

Provenance Working Group Teleconference

06 Dec 2012

Attendees

Contents