From Provenance WG Wiki
Revision as of 11:40, 18 May 2012 by Lmoreau
Please justify/explain all edits to this page, in your "edit summary" text.
14:50:32 <RRSAgent> RRSAgent has joined #prov 14:50:32 <RRSAgent> logging to http://www.w3.org/2012/05/17-prov-irc 14:50:34 <trackbot> RRSAgent, make logs world 14:50:34 <Zakim> Zakim has joined #prov 14:50:36 <trackbot> Zakim, this will be 14:50:36 <Zakim> I don't understand 'this will be', trackbot 14:50:37 <trackbot> Meeting: Provenance Working Group Teleconference 14:50:37 <trackbot> Date: 17 May 2012 14:50:37 <Luc> Zakim, this will be PROV 14:50:37 <Zakim> ok, Luc; I see SW_(PROV)11:00AM scheduled to start in 10 minutes 14:50:46 <Luc> Agenda: http://www.w3.org/2011/prov/wiki/Meetings:Telecon2012.05.17 14:51:36 <Luc> Chair: Moreau 14:51:44 <Luc> rrsagent, make logs public 14:51:49 <Luc> zakim, who is here? 14:51:49 <Zakim> SW_(PROV)11:00AM has not yet started, Luc 14:51:50 <Zakim> On IRC I see RRSAgent, Luc, MacTed, stain, trackbot, sandro 14:52:03 <Luc> Regrets: Paul Groth, Tom DeNies 14:52:17 <Luc> Regrets: Paul Groth, Tom DeNies, Paolo Missier 14:57:34 <Paolo_> Paolo_ has joined #prov 14:57:54 <smiles> smiles has joined #prov 14:58:01 <Zakim> SW_(PROV)11:00AM has now started 14:58:08 <Zakim> +??P16 14:59:01 <tlebo> tlebo has joined #prov 14:59:10 <Zakim> + +1.315.330.aaaa 14:59:13 <Zakim> -??P16 14:59:19 <tlebo> zakim, I am aaaa 14:59:19 <Zakim> +tlebo; got it 14:59:54 <Zakim> +Luc 15:00:13 <Curt> Curt has joined #prov 15:00:20 <Zakim> +??P21 15:00:21 <Luc> Scribe: simon miles 15:00:30 <Luc> zakim, who is on the call? 15:00:30 <Zakim> On the phone I see tlebo, Luc, ??P21 15:01:04 <Luc> topic: admin <luc>Summary: Last week's minutes were approved. 15:01:31 <Zakim> +Curt_Tilmes 15:01:46 <Zakim> +[OpenLink] 15:01:56 <GK1> GK1 has joined #prov 15:01:58 <MacTed> Zakim, [OpenLink] is temporarily me 15:01:59 <MacTed> Zakim, mute me 15:02:00 <Zakim> +MacTed; got it 15:02:06 <Zakim> MacTed should now be muted 15:02:51 <Luc> proposed: to accept minute of last week's teleconference 15:03:06 <jcheney> jcheney has joined #prov 15:03:19 <Luc> http://www.w3.org/2011/prov/meeting/2012-05-10 15:03:20 <tlebo> +1 15:03:21 <Curt> 0 (not present) 15:03:23 <smiles> +1 15:03:26 <MacTed> +1 15:03:38 <Zakim> + +44.131.467.aabb 15:03:54 <Luc> resolved: minutes of last week's teleconference 15:03:54 <MacTed> Zakim, who's here? 15:03:54 <Zakim> On the phone I see tlebo, Luc, ??P21, Curt_Tilmes, MacTed (muted), +44.131.467.aabb 15:03:57 <Zakim> On IRC I see jcheney, GK1, Curt, tlebo, smiles, Zakim, RRSAgent, Luc, MacTed, stain, trackbot, sandro 15:04:09 <Luc> http://www.w3.org/2011/prov/track/actions/open 15:04:10 <smiles> Luc: Action review 15:04:20 <GK1> oops, sorry, still getting audio together 15:04:30 <GK> GK has joined #prov 15:04:40 <smiles> Luc: Action on Sandro regarding emailing emailing announcements to W3C mail list 15:04:49 <jun> jun has joined #prov 15:04:54 <tlebo> I just sent my review 15:05:02 <smiles> Luc: Actions on Tim, Graham to review constraints doc - talk about later 15:05:05 <Paolo> Paolo has joined #prov 15:05:14 <Luc> Topic: PAQ release <luc>Summary: no progress since last week. Editors need to request a publication date from the Webmaster. 15:05:41 <smiles> Graham? 15:05:41 <Zakim> +??P49 15:05:44 <Luc> graham? 15:05:59 <GK> zakim, ??P49 is me 15:06:01 <Zakim> +GK; got it 15:06:01 <MacTed> Zakim, unmute me 15:06:03 <Zakim> MacTed should no longer be muted 15:06:08 <MacTed> Zakim, mute me 15:06:11 <Zakim> MacTed should now be muted 15:06:17 <khalidbelhajjame> khalidbelhajjame has joined #prov 15:06:29 <Zakim> +??P50 15:06:31 <Zakim> +??P44 15:06:38 <jun> zakim, ?P44 is me 15:06:40 <Zakim> sorry, jun, I do not recognize a party named '?P44' 15:06:42 <Christine> Christine has joined #prov 15:06:49 <jun> zakim, ??P44 is me 15:06:51 <smiles> GK: PAQ has not been edited in past week, so not ready for release yet 15:07:05 <Zakim> +jun; got it 15:07:09 <smiles> GK: Publication release not yet requested to his knowledge 15:07:17 <smiles> Luc: Please agree release date soon 15:07:22 <Luc> topic: other documents <Luc>Summary: there was a brief overview of progress on the various other documents. For prov-o, prov-n, prov-dm, editors are addressing issues on tracker. They are on time for internal release at the end of month. 15:07:37 <Zakim> +??P51 15:07:37 <dgarijo> dgarijo has joined #prov 15:07:43 <Zakim> +??P6 15:07:48 <khalidbelhajjame> zakim, ??P51 is me 15:07:49 <Zakim> +Sandro 15:08:02 <smiles> Luc: For PROV-DM, have made a number of changes, closed many issues 15:08:26 <Zakim> -??P6 15:08:32 <smiles> ... some issues still outstanding, listed in the agenda so people who raised them can talk to them: Khalid, Yolanda, Graham, Tim 15:08:36 <Zakim> +khalidbelhajjame; got it 15:08:39 <GK> Issue 88 can be closed as far as I'm concerned 15:08:44 <Zakim> +??P3 15:08:57 <dgarijo> Zakim, ??P3 is me 15:08:57 <Zakim> +dgarijo; got it 15:08:57 <smiles> ... For PROV-N, implemented optional identifier changes, made grammar linkable for navigation, simplified presentation 15:09:04 <smiles> ... soon ready for review 15:09:18 <GK> Just closed issue 88 15:09:21 <smiles> ... For PROV-CONSTRAINTS, no progress, waiting for feedback? 15:09:25 <smiles> jcheney: confirms 15:09:35 <Luc> @graham, thanks 15:09:57 <smiles> tlebo: PROV-O, been closing issues, two requests for review before closing 15:10:25 <smiles> ... added cross-references for terms within HTML document 15:10:34 <smiles> ... latest draft linked on agenda 15:10:42 <smiles> ... feedback on the cross-references welcome 15:10:48 <Luc> q? 15:10:50 <smiles> ... on track for release June 1 15:11:13 <Luc> q? 15:11:22 <satya> satya has joined #prov 15:11:25 <smiles> smiles: PROV-Primer, not much to report from last week #15:11:28 <Luc> topic: collections organization 15:11:29 <Zakim> +Satya_Sahoo 15:11:31 <smiles> Luc: any comments on progress? <luc>Summary: this topic is a continuation from last week. Graham presented his rationale for restructuring prov-dm into two documents. Tim presented the prov-o team proposal of not separating the collection section from the prov-o document, since recent changes have significantly trimmed down the space occupied by collections. The rest of the time was spent discussing Graham's proposal: Macted supported the idea of separating "essential" concepts from "extension" concepts, though this could be done in a single document. Satya noted that some concepts were more application specific, e.g. softwareAgent. Tim observed the flat list of concepts Graham suggested lacked the kind of structure that prov-dm components currently offered: it was noted that essential concepts listed by Graham more or less corresponded to two/three existing prov-dm components (1, 2, 3). Curt defended the existing structure based on components, which allowed different views of provenance to be accommodated (e.g. process-oriented/data flow oriented/ responsibility oriented). Luc noted that the structure Graham presented was exactly the one we had two iterations ago, but it was criticised and could not be defended. Luc also indicated that we could make it clear in the current structure which concepts were more primitive than others. Section 2 of the current document (starting points, Table 2 and Figure 1) is in fact listing Graham's core concepts. We run out of time to make a decision. Chairs will come up with a proposal at the next teleconference. 15:12:21 <smiles> Luc: Several reviewers felt section of PROV-O on collections was long, and made appear more important than they are 15:12:39 <smiles> ... Paul suggested separating the collections out of the PROV-O document 15:12:54 <smiles> ... Separately, Graham suggested restructuring DM 15:13:10 <smiles> ... Last week, requested concrete proposals for restructuring 15:13:21 <Luc> http://www.w3.org/2011/prov/wiki/ProvDM_Proposal_for_restructuring 15:14:13 <smiles> GK: Proposal is in some ways quite radical, and is focused around separating central provenance patterns from those for specific processes 15:14:57 <smiles> ... The rationale is to achieve (1) separate core provenance patterns from specific applications, for comprehensibility of core idea 15:15:51 <Zakim> -??P50 15:15:54 <smiles> ... (2) Maximising interoperability with other systems doing provenance-like things 15:17:03 <smiles> ... other models including provenance seem to include core matching the core DM patterns 15:17:42 <MacTed> apropos of GK's "core" patterns... this came to my eyes today -- http://linkedevents.org/ontology/ 15:17:43 <smiles> ... (3) Minimising ontological commitment of users of model, so core embodies little semantics but captures essentials of traceability 15:18:15 <smiles> ... Core: entity, activity, agent 15:18:45 <smiles> Luc: Your proposal is to break DM document in two? 15:18:48 <smiles> GK: Yes 15:18:51 <Luc> q? 15:19:08 <Luc> paolo? 15:19:15 <Luc> just on irc? paolo? 15:20:02 <tlebo> http://www.w3.org/2011/prov/wiki/Collections 15:20:13 <smiles> tlebo: Other proposal (linked above) from Paolo 15:20:57 <smiles> ... collections and dictionaries taken out of PROV-O, classes and properties, and put into separate PROV-O-C document 15:21:03 <smiles> ... aim to simplify PROV-O 15:21:06 <Paolo> sorry guys text only, 15:21:10 <Paolo> and very unstable 15:21:37 <smiles> ... PROV-O team discussed on Monday, preferred to focus on the content of PROV-O rather than deconstructing 15:21:40 <Paolo> thanks smiles for minuting 15:21:44 <tlebo> I can fit http://www.w3.org/TR/prov-o/#description-collections-terms into 1.25 screens. http://www.w3.org/TR/prov-o/#description-qualified-terms takes up 4 screens. http://www.w3.org/TR/prov-o/#description-starting-point-terms takes up 1.5 screens 15:21:48 <Luc> @paolo, it's ok, tim filling in 15:22:31 <smiles> ... Have responded to reviews by simplifying content on collections 15:22:35 <Paolo> yes I know Tim questions the motivation for this ripping exercise 15:22:57 <smiles> ... PROV-O team prefers to keep collections in PROV-O document 15:23:16 <Paolo> we seemed to agree that it's for the prov-o team to pursue this if they want 15:23:30 <smiles> Luc: After last telecon, Paul and Luc considered logistics of taking collections out of existing documents to make new document 15:24:08 <smiles> ... short of editors and bandwidth, and goes beyond scope of original charter to give application specific extensions 15:24:36 <smiles> ... that is why Paolo suggested just extracting from PROV-O 15:25:17 <smiles> ... Tim, are you proposing not separating, as length concerns are already being addressed? 15:25:20 <smiles> tlebo: Yes 15:25:28 <tlebo> (and that was the agreement of the prov-o team) 15:25:41 <Luc> q? 15:25:56 <smiles> Luc: We have proposals to not do anything on restructuring or Graham's proposal 15:26:04 <tlebo> q+ 15:26:55 <smiles> tlebo: From explaining to other people, people latch onto those core concepts 15:27:06 <smiles> ... (as in section 1 of Graham's document) 15:28:24 <smiles> tlebo: What about components (organisational structure of current draft)? 15:28:36 <Curt> http://www.w3.org/TR/prov-dm/#data-model-components 15:28:46 <smiles> GK: Tried to stick with existing material, suggest grouped in different way 15:28:51 <Luc> ack tlebo 15:28:52 <tlebo> q- 15:29:30 <smiles> Luc: Structure proposed is more or less what we had two iterations ago, but sections in one document 15:29:40 <smiles> ... but had reviews critical of this separation 15:29:57 <smiles> ... so decided to reorganise to remove distinction of core and extension 15:30:23 <Paolo> Paolo has joined #prov 15:30:32 <smiles> ... seems to be going back, and when we get to justifying what is core, what is extension, we will have difficulties 15:31:01 <smiles> ... Second, have had feedback from people outside WG who found component structure useful 15:31:18 <smiles> ... so reluctant to abandon it if we split document 15:31:44 <smiles> ... Third, if restructure PROV-DM, then have to do the same in other documents 15:32:11 <smiles> ... lead to a multiplication of documents, as scary as large number of concepts in current model 15:32:17 <MacTed> what is background of these outside readers? philosophers, scientists, programmers, other? experience and grounding matters to whether the current structure is easy to understand... 15:33:00 <smiles> ... There are sub-types, e.g. wasRevisionOf subtype of wasDerivedFrom, and could make more explicit in structure of DM 15:33:47 <smiles> ... For example, derivations in section 6.3.1, could then have subsections for subtypes 15:34:01 <smiles> ... (4.3.1 not 6.3.1) 15:34:53 <smiles> ... or explicit marker for terms that are core, e.g. communication is not primitive as can be described in terms of generation and usage 15:35:40 <Luc> q? 15:35:43 <satya> q+ 15:36:36 <smiles> satya: As MacTed wrote above, who are the readers of the documents? which reviewers? 15:36:44 <MacTed> q+ 15:36:48 <satya> q- 15:37:00 <MacTed> Zakim, unmute me 15:37:00 <Zakim> MacTed should no longer be muted 15:37:01 <smiles> Luc: In this case, researchers he works with who've implemented data model, felt component structure helped 15:37:44 <smiles> MacTed: That kind of feedback is not very useful, need more kinds of audience 15:37:59 <smiles> ... in favour of GK's restructuring 15:38:24 <smiles> ... for PROV-O, does not seem to have discerned what is a sub-class of what, what are the overarching elements 15:38:51 <smiles> ... there really are core concepts, and refinement of those 15:39:26 <tlebo> perhaps if we organized http://www.w3.org/2011/prov/wiki/ProvDM_Proposal_for_restructuring#Provenance_core_concepts by components? 15:39:26 <Luc> q? 15:39:30 <Luc> ack mact 15:39:56 <smiles> ... root primitives need to be clearly presented 15:40:25 <smiles> Luc: Agree that root primitives need to be clearly presented 15:40:28 <Luc> q? 15:40:37 <GK> q+ to say I'll respond oif there arter no other comments 15:41:43 <smiles> GK: Responding to Luc's point, looking at document two iterations ago, while separation of core from other concepts, too much other clutter so organisation wasn't serving purpose 15:42:31 <smiles> ... Added rationale to his proposal of separation of core pattern, with principles clear 15:43:19 <tlebo> q+ to ask what if graham incorporated components into his outline? 15:43:23 <smiles> ... With regard to restructuring other documents, don't see need to do so, just restructure DM, leave others as they are 15:43:34 <smiles> ... PROV-O already does the job of pulling out core patterns 15:43:45 <jcheney> q+ to say I need to leave at 5 so can we discuss the prov-constraints review briefly before spending the rest of the meeting debating restructuring 15:44:32 <smiles> ... Regarding changing presentation, mixing text on subtypes with supertypes would be exact opposite 15:44:39 <smiles> ... of what is intended 15:44:40 <tlebo> @jcheny, I'll yield :-) 15:45:14 <smiles> ... To have to dig around in document for core ideas means much less likely specification would be deployed 15:45:22 <Luc> q? 15:45:22 <tlebo> q- 15:45:30 <GK> q- 15:45:46 <Luc> ack jcheney 15:45:46 <Zakim> jcheney, you wanted to say I need to leave at 5 so can we discuss the prov-constraints review briefly before spending the rest of the meeting debating restructuring 15:45:56 <Luc> topic: constraints document <luc>Summary: Tim and Graham reviewed/checked the latest prov-constraints document and were supportive of its new direction and structure. Editors can now continue work and address remaining issues. 15:46:08 <tlebo> go ahead, graham. 15:46:28 <jcheney> high-level impression is fine 15:46:41 <smiles> GK: Looked through constraints document, feels a lot tighter and has right approach 15:46:54 <smiles> ... definitions and inferences presented crisply 15:47:10 <smiles> ... may be able to make more comments later, but looking good 15:47:35 <smiles> tlebo: Biggest concern on last iteration was about getting into content 15:47:48 <tlebo> http://www.w3.org/2011/prov/wiki/Dm-constraints_review_2012_May_17_by_Lebo 15:47:50 <smiles> ... this version is much better organised, natural to know where to go 15:48:20 <smiles> ... minor detailed comments sent (above) 15:49:17 <smiles> jcheney: Thanks, was really looking for high level impression, thanks for going through in more detail 15:49:27 <smiles> ... good to know happy with direction 15:49:59 <GK> I think the style of revised -CONSTRAINTS will nicely complement a less formal description of -DM 15:50:06 <tlebo> @jcheney, sorry, I missed your questions in the email :-) 15:50:16 <smiles> ... after last week had more specific questions, implicitly answered in Tim's comments, but please look at questions in email 15:51:16 <tlebo> @jcheney, I'll respond to the email questions after this meeting. 15:51:27 <smiles> ... will go through issues raised to see what can be closed 15:51:30 <Luc> q? 15:52:05 <Luc> Topic: Responsibility <luc>Summary: The group is invited to suggest "nouns" for the relation actedOnBehalfOf to replace the current term 'responsibility'. 15:52:18 <Luc> http://www.w3.org/2011/prov/wiki/Renaming_the_concept_Responsibility 15:53:01 <smiles> tlebo: In definitions of assocation and attribution are responsibility, so name Responsibility is confusing and misnamed 15:53:34 <jcheney> I may have suggested "delegation" at some point 15:54:03 <smiles> ... Wiki page comments above to prompt discussion by email 15:54:16 <satya> is there an issue raised for responsibility? 15:54:37 <Luc> topic: bundles <luc>Summary: discussion on bundles to continue over email. 15:54:46 <Luc> http://dvcs.w3.org/hg/prov/raw-file/default/model/working-copy/wd6-bundle.html 15:55:03 <smiles> Luc: Circulated text addressing issues raised # SPECIAL MARKER FOR CHATSYNC. DO NOT EDIT THIS LINE OR BELOW. SRCLINESUSED=00000239