15:44:26 RRSAgent has joined #prov 15:44:26 logging to http://www.w3.org/2011/12/15-prov-irc 15:44:28 RRSAgent, make logs world 15:44:30 Zakim, this will be 15:44:30 I don't understand 'this will be', trackbot 15:44:31 Meeting: Provenance Working Group Teleconference 15:44:31 Date: 15 December 2011 15:44:32 Zakim, this will be PROV 15:44:32 ok, Luc; I see SW_(PROV)11:00AM scheduled to start in 16 minutes 15:44:46 Agenda: http://www.w3.org/2011/prov/wiki/Meetings:Telecon2011.12.15 15:44:56 MacTed has joined #prov 15:45:08 Chair: Luc Moreau 15:46:00 Regrets: Paul Groth, Christine Runnegar, Curt Tilmes 15:46:08 Scribe: dgarijo 15:46:15 rrsagent, make logs public 15:46:20 Topic: Admin 15:52:08 dgarijo has joined #prov 15:54:35 GK1 has joined #prov 15:54:47 hi daniel, everything set up for you 15:54:53 thanks, Luc 15:55:36 SW_(PROV)11:00AM has now started 15:55:38 Lena has joined #prov 15:55:41 +[IPcaller] 15:55:44 +Luc 15:55:59 Zakim, [IPcaller] is me 15:55:59 +dgarijo; got it 15:57:18 GK has joined #prov 15:57:44 how many grahams have we got :o? 15:58:25 Just one, but with two computers :) 15:58:36 :) 15:59:17 zednik has joined #prov 15:59:20 Agenda for today's meeting: http://www.w3.org/2011/prov/wiki/Meetings:Telecon2011.12.15 15:59:27 +??P52 15:59:32 Paolo has joined #prov 15:59:50 +[ISI] 15:59:51 +[IPcaller] 16:00:03 lebot has joined #prov 16:00:07 zakim, ??P52 is me 16:00:19 +GK; got it 16:00:24 khalidbelhajjame has joined #prov 16:00:25 +??P2 16:00:33 + +1.518.633.aaaa 16:00:41 zakim, ??P2 is me 16:00:48 + +1.315.330.aabb 16:00:50 +Paolo; got it 16:00:58 zakim, aabb is tlebo 16:01:03 Zakim, who is here? 16:01:14 +tlebo; got it 16:01:18 On the phone I see Luc, dgarijo, GK, [ISI], [IPcaller], Paolo, +1.518.633.aaaa, tlebo 16:01:33 On IRC I see khalidbelhajjame, lebot, Paolo, zednik, GK, Lena, GK1, dgarijo, MacTed, RRSAgent, Luc, trackbot, mdmdm, stain, Zakim, sandro 16:01:46 @luc - do you need scribe? 16:01:55 +??P32 16:02:01 daniel has volunteered, thank you Graham 16:02:21 zakim, aaaa is me 16:02:33 SamCoppens has joined #prov 16:02:38 Luc: agenda: http://www.w3.org/2011/prov/wiki/Meetings:Telecon2011.12.15 16:02:45 jcheney has joined #prov 16:03:05 ... aob? 16:03:06 smiles has joined #prov 16:03:14 StephenCresswell has joined #prov 16:03:29 +zednik; got it 16:03:42 +[IPcaller.a] 16:03:53 +sandro 16:03:55 ... we will propose to accept the minutes of last week next week 16:04:02 zakim, [IPcaller.a] is me 16:04:12 actions: Yolanda and Satya had some, but they were closed last week 16:04:32 Luc: scribes are still needed. Please volunteer. 16:04:38 http://www.w3.org/TR/2011/WD-prov-o-20111213/ 16:04:45 topic: releases 16:04:55 http://www.w3.org/TR/2011/WD-prov-dm-20111215/ 16:05:01 ... 2 releases: On tuesday the FPWD for the prpov-o document. Today we'll have the release of prov dm 16:05:19 ... Congratulations to everyone 16:05:20 +khalidbelhajjame; got it 16:05:28 +??P16 16:05:34 +SamCoppens 16:05:37 ... feedback from outside is already appearing, which is great. 16:06:01 zakim, ??p16 is me 16:06:07 ... it would be good to plan some disemination for the releases on Jan. Any ideas? 16:06:40 ... khalid, stian, have you planned advertising this? 16:06:41 +jcheney; got it 16:07:02 I thought the idea was to publicize all the new WDs after the holidays 16:07:07 khalid: we haven't thought about that yet. We should spend some time on that. 16:07:37 ... it will be discussed on the next prov-o telecon. 16:08:06 Luc: congrats to the team again. 16:08:34 topic: Prov-primer 16:08:49 Luc: what is the situation? 16:08:59 Simon: Yolanda made some edits. 16:09:01 YolandaGil has joined #prov 16:10:01 ... wednseday we distributed the new version of the primer, although there was 1 comment about a statement on which there is no consensus yet 16:10:14 ... feedback is welcome 16:10:29 Luc: next week we may not have many people on the call. 16:10:37 q? 16:10:37 ... could you vote today? 16:10:45 +q 16:11:12 dgarijo: if we correct the wasAssociatedWith I think we can release it. 16:11:53 Stephan: they are not synchronized, so we shouldn't release it yet. 16:12:03 ... until we fix it 16:12:16 Luc: it is not practical to synchornize releases. 16:12:18 q+ 16:12:27 ... we could release something in between. 16:12:40 The prov-o FPWD refers to the (October) DM FPWD. 16:13:05 ... if we synchronize everuthing, documents get delayed. 16:13:19 Stephan: it can be confusing. 16:13:38 Luc: is there a particular version that can be more or less aligned? 16:13:53 Simon: there is a previous one. 16:13:54 q+ 16:14:34 Luc: the picture you incorporated is from the 15th, not 18th. 16:14:42 On reflection, I think it's acceptable for FPWDs to be out of sync, on the basis we need to show the world where we're headed, but we should aim for stabilization of central concepts for subsequent releases, as far as practical. 16:14:47 Simon: we'll use the most recent in the primer 16:14:50 There are links to the HG repository drafts in the primer. 16:14:50 Luc: ok 16:15:31 Luc: I think that is worth adding a paragraph to adress Stephan's point 16:16:12 can we just have a summary of these discrepancies upfront? 16:16:16 Simon: yes, it's reasonable. 16:16:22 q? 16:16:25 zednik: ok 16:16:26 more informative than just a ref to a specific previous version 16:16:28 ack zed 16:16:30 -q 16:16:34 q- 16:16:55 q- [] 16:17:03 q- 16:17:05 ack yol 16:17:13 q? 16:17:26 Yolanda: I was going to propose what you just proposed, Luc. Last week we took things that seemed complicated from the model, so it is easier as a starting point. 16:17:58 Luc: the document will be very helpful for the community. 16:18:28 proposed: to release prov-primer as fpwd 16:18:40 +1 16:18:41 dgarijo:+1 16:18:46 +1 (UoE) 16:18:46 +1 16:18:48 +1 16:18:52 +1 16:18:55 +1 (Oxford) 16:19:06 +1 (IBBT) 16:19:22 +1 (note I'm an editor) 16:19:26 +1 16:19:44 accepted: to release prov-primer as fpwd 16:20:05 Luc: now you have green lights to start the process of releasing these documents. 16:20:37 ... no publication will take place before January, but we could start the set of requests for the publication. 16:20:53 Simon: patent information? 16:21:11 I copied the patent URI from the DM document 16:21:19 .. for PROV-AQ 16:21:24 Luc: this document will be a note, so the rest is fine. 16:22:26 GK: I added the same phrase to the prov-aq 16:22:42 Luc: there is a link wrong in the primer. 16:23:07 ... plans for the second working draft? 16:23:13 ... plans for releasing it? 16:23:30 q? 16:23:35 Simon: first of all, synchronize it with the latest dm 16:23:41 yolanda? 16:24:11 topic: prov-aq 16:24:17 Luc: share with us your plans on the next telecon, when you're available 16:24:23 http://dvcs.w3.org/hg/prov/raw-file/default/paq/WD-PROV-AQ-20111212.html 16:24:42 GK: since last week Paul has been working on it 16:24:58 ... had a response from people 16:25:12 ... will be published next year 16:25:38 ... Paul is taking care of the last issues. 16:25:59 ... We haven't started discussing the 2nd working draft plan 16:26:02 q? 16:26:13 ... still have some issues before we release the first draft 16:26:47 ... we have to look for feedback and anwer it 16:27:24 Luc: worried about discussing the release at the F2F meeting. 16:27:52 ... will you have a revised draft before? 16:27:54 q? 16:27:58 GK: will try to. 16:28:20 q? 16:28:25 topic: prov-o 16:28:27 Luc: please provide comments on these drafts. It is key for the editors to progress. 16:28:46 FWIW, PROV-AQ issues: http://www.w3.org/2011/prov/track/products/5 16:28:51 Luc: plan for the second WD? 16:29:04 q? 16:29:36 khalid: we haven talked about that yet. But We'll work on synchronize the current version with prov dm 16:29:57 http://www.w3.org/2011/prov/wiki/F2F2#Objectives 16:29:57 Luc: it would be nice if you could talk about it. 16:30:05 http://www.w3.org/2011/prov/wiki/F2F2#Inputs_for_F2F2 16:30:27 ... Reminder of the objectives and inputs we wanted to have before the second f2F meeting. 16:30:58 q? 16:31:11 topic: prov-sem 16:31:13 ... what you suggest here is in line with what we agreed on, but we have to make sure we progress on the objectives 16:31:28 http://www.w3.org/2011/prov/wiki/FormalSemanticsStrawman 16:31:28 Luc: James has been drafting the semantics. 16:32:09 James: gone through the strawman and updated the notation 16:32:45 ... the html is not on the right format, but I rather spend some time on the wiki and work on the content and what we agreed on. 16:33:01 Luc: at this stage it is better through the wiki 16:33:14 James: happy to stick with the wiki. 16:33:26 q? 16:33:32 ... discussion of wasComplement of that has to be updated 16:33:51 q? 16:34:02 ... friday is the deadline, so it is better to wait for monday 16:34:09 q+ 16:34:20 q? 16:34:25 ack GK 16:34:26 Luc: thanks for doing this work. Others will join you 16:34:54 GK: the document is similar to model theoretical semantics 16:34:59 q? 16:35:00 James: Yes. 16:35:16 topic: prov-dm 16:35:25 http://lists.w3.org/Archives/Public/public-prov-wg/2011Dec/0148.html 16:35:40 Luc: last week I sent an email with 3 proposal for events. 16:35:58 ... we had 10-15 votes during the week 16:36:20 ACCEPTED: PROPOSAL 1. Adopt the following Definition for generation. In PROV-DM, a generation record is a representation of a world event, the completed creation of a new entity by an activity. This entity did not exist before this event; this entity is available for usage after this event. 16:36:24 ... the outcome is: The first proposal to redefine generation was supported. 16:36:45 ACCEPTED: PROPOSAL 2. Adopt the following Definition for usage. In PROV-DM, a usage record is a representation of a world event: the start of an entity consumption by an activity. Before this event, this entity was not consumed or used in any form or shape by the activity, totally or partially. 16:36:50 ... the second proposal to define usage was strongly supported. It could be rephrased better though. 16:36:57 PROPOSAL 3. Replace the word event by action. 16:37:21 ... the 3rd proposal did not have enough support. 16:38:10 ... 3 quarters of the people voted no. Suggestions are milestone or Instantaneous Event. 16:38:24 ... I will not try to resolve it here now 16:38:42 ... Raise issues regarding the term 16:38:48 q? 16:38:59 q+ 16:39:03 ... happy to take any comments 16:39:39 zednik: about the last sentence of proposal 1. Can we discuss it or should I raise an issue? 16:39:59 Luc: before generation it is not an entity, because it is not generated get. 16:40:24 zednik: I understand the intent 16:40:36 Luc: If you have another phrasing, please raise an issue 16:40:50 q? 16:40:52 zednik: will do. 16:40:56 q- 16:40:56 ack ze 16:40:56 Luc: thanks 16:41:20 Luc: since last week we've made a set of changes. 16:41:21 http://lists.w3.org/Archives/Public/public-prov-wg/2011Dec/0190.html 16:41:29 ... the first one is regarding wasViewOf 16:41:37 http://dvcs.w3.org/hg/prov/raw-file/default/model/ProvenanceModel.html#record-complement-of 16:41:57 ... did anyone have the oportunity to go through the text? 16:42:14 ... spent some time discussing with paolo 16:42:24 +q 16:42:26 ... feedback would be appreciated. 16:42:30 q+ 16:42:30 q? 16:42:43 ack steph 16:43:01 q- 16:43:09 Stephen: What was discussed last week is not still reflected on prov dm 16:43:13 I was about to same the same as Stephan 16:44:08 q+ 16:44:09 ... the text mentions: Note that, as a particular case, the validity interval of e1 may be contained within the validity interval of e1. 16:44:41 ... not agreeing with the current definition of interval containment. 16:45:02 q? 16:45:05 ack pao 16:45:11 Luc: containment is mentioned as a special case of what is being defined. 16:45:16 q+ to note (again) that I think the important feature of "a viewOf b" is that all provenance about b is also true of a 16:45:41 paolo: it doesn't make sense to assert that when intervals do not overlap. 16:45:45 q? 16:46:28 paolo: I don't think that we have discussed that the validity of the intervals can be inferred from the assertions. 16:47:01 ... you can complete the model with the assertions that you make. 16:47:17 ... we should just make a decission 16:47:32 +q 16:47:52 ... the containment issue: there is no need to insist on containment or intervals 16:48:51 ... assuming or infering interval containment may not be the best thing 16:49:01 +1 to ask how a facebook account is a viewOf a twitter account. complementOf was just renamed here. 16:49:45 IMO, bob the twiotter account holder is NOT view of bob the facebook account holder 16:49:46 ack GK 16:49:46 GK, you wanted to note (again) that I think the important feature of "a viewOf b" is that all provenance about b is also true of a 16:49:50 Luc: is the example that is discussed here one that we want to capture with this relation. If it is, does it make sense? 16:49:55 also, it's not the _account_, it's the _person with_ the account. 16:50:56 @tim, did you mean to join the queue? 16:51:11 +q to ask how a facebook account is a viewOf a twitter account. complementOf was just renamed here. 16:51:12 q+ 16:51:13 GK: testability. If a is a view of b, then all the problems that b would have because of the changes would also be applicable to a? 16:51:16 q? 16:51:20 I added an alternative proposal to the wiki after last week's meeting http://www.w3.org/2011/prov/wiki/ComplementarityUseCases 16:52:06 Stephen: interval containment is useful. I have added a proposal to the wiki 16:52:07 q? 16:52:10 ack ste 16:52:50 @stephen +1 16:52:50 tlebo: looks like complement of is a rename for view of. Facebook account is a view of Tim, not from anoither Twitter cccount 16:53:32 I just looked a Stephen's entry on wiki page, think it looks good. 16:53:34 tlebo: we renamed the relation and now it can be confusing. The example is right, but not well written. 16:54:14 .. http://www.w3.org/2011/prov/wiki/ComplementarityUseCases#An_alternative_proposal_.28stephen.29 16:54:18 q? 16:54:22 ack leb 16:54:22 lebot, you wanted to ask how a facebook account is a viewOf a twitter account. complementOf was just renamed here. 16:54:33 paolo: disagree with all 3 comments. 16:54:52 ... complement of was based on a level of how much knowledge you have of an entity 16:54:55 q? 16:54:58 ack pao 16:55:04 ... it has completely disappeared here. 16:55:32 ... now we have time intervals and create minimum assertions about them. 16:56:55 ... we felt that 2 characterizations of the same entity should not have a hierarchy. 16:57:56 ... I don't think that having the 2 accounts view of each others does not violate anything. 16:57:57 q? 16:58:12 We want something with actual transitivity, not just pseudo-transitivity. I don't see what use pseudo-transitivity is. 16:58:14 I was previously looking at the wrong document ... What Paolo says is reasonable, except that I think the name viewOf should 16:58:16 q+ 16:58:18 Luc: not made much progress. 16:58:24 Still wondering how "me via facebook" is a view of "me via twitter" and "me via prov-wg call" 16:58:36 It is always a good idea to leave complimentarity discussions til the end of the agenda :-) 16:58:54 GK: problem may be with the names 16:59:07 pseudo-transitivity is simply transitivity with the additional condition on consistency of intervals 16:59:11 +1 GK! 16:59:17 ... paolo's view Of should be complement of 16:59:43 ... 2 competing set of intuitions. 17:00:23 GK: I din't see the necessity for interval containment, although it could be there in practice. 17:00:43 q? 17:01:02 -SamCoppens 17:01:08 @Paolo Pseudo-transitivity seems to allow us to derive things which are meaningless - or depend on a separate test which we might not always be able to make. 17:01:15 Luc: just to clarify this issue. Tim, are you supportive to what Graham said? 17:01:25 tlebo: yes 17:02:01 +SamCoppens 17:02:06 ... view is complement. 17:02:14 @tlebo thanks - just what I was trying to say! 17:02:17 s/view/view of 17:02:20 if you go for "abstract" vs "concrete" then you could argue that the "abstract" version has a longer lifetime than the "concrete", hence containment, but I still feel that's arbitrary 17:03:09 Luc: would not be against the renaming proposed by Tim and Graham. 17:03:11 @paolo ... I don't see it as abstract vs concrete 17:03:16 if we had both the "vertical" and "horizontal", would that help? with suitable renaming of the relation names 17:03:45 q? 17:03:49 q+ 17:03:54 Luc: I will work in a proposal. We will try to adress Stephen's concerns 17:03:54 @GK Stephen Z use that terminology 17:03:59 q- 17:04:08 q+ to clarify the changes that paolo made 17:04:15 ack gk 17:04:25 -[ISI] 17:04:46 Tom: question about the attributes in the new proposal. 17:04:57 It's also changed to become symmetric. 17:05:03 q? 17:05:05 paolo: There is no attributes anywhere at this point. 17:05:23 s/is/are. 17:06:25 Tim: will read and put a comment on the list 17:06:53 Luc: you seem to be in agreement. Can you agree on a text and send it to the rest of the group? 17:06:59 q- 17:07:08 Luc: GK, Tim, paolo 17:07:47 +q 17:08:22 GK: what os proposed in the wiki sounds reasonable. 17:08:30 s/os/is 17:08:43 paolo: maybe we are running in circles. 17:08:59 +q 17:09:09 Luc: how do we move this forward 17:09:40 stephen: transitivity **I missed something here** 17:10:26 q? 17:10:28 +q 17:10:29 paolo: we don't know normally what this intervals are. I don't see this intervals as that relevant. 17:10:36 ack steph 17:10:41 @paulo, even if we don't know what they are, the properties may be inferred existentially 17:11:03 stephen: we should be allowed to assert to assert something about the interval even if we don't know the time 17:11:40 paolo: the model is missing a facility to make clear when entities exist. 17:12:09 khalid: instead talking about intervals, talk about the validity of an entity. 17:12:14 Got to go, sorry 17:12:18 -jcheney 17:12:28 q? 17:12:33 ack khal 17:12:38 paolo: will have to ground validity. 17:12:54 khalid: we don't have means to ground it. My proposal is to let it be 17:13:56 q+ 17:14:00 Luc: we're editing the WD with this. Is it better to revert back to the previous version of that section and then review both proposals separately? 17:14:18 I'd rather keep the rewrite in as it is (so I can read the new stuff) 17:14:37 GK: the document is not the main issue. The consensus is the main issue. 17:14:48 -1 for reverting back 17:14:51 will the rewrite stand somewhere else? 17:15:03 Luc: should revert and THEN reach consensus, knowing that some people will look at the doc. 17:15:10 GK: No. 17:15:25 q? 17:15:32 Luc: ok, lets keep it 17:15:43 http://lists.w3.org/Archives/Public/public-prov-wg/2011Dec/0161.html 17:16:04 @luc: I need to properly read Paolo's current proposal, and comment. Would like to see if can find a consensus view (sic). 17:16:08 Luc: 2 other sections that we edited: the section about hadPlan (to change recipelink). 17:16:24 Luc: please add feedback 17:16:26 http://lists.w3.org/Archives/Public/public-prov-wg/2011Dec/0163.html 17:16:33 s/add/provide some/. 17:16:54 Luc: interpretation sections with the constraints for intervals 17:17:04 q? 17:17:12 ack pao 17:17:45 paolo: propose to rewrite the proposal with today's comments. 17:17:56 -SamCoppens 17:18:13 Luc: ok 17:18:27 ... send an email to get feedback 17:18:39 Luc: too late to start discussion on accounts. 17:18:52 q? 17:19:00 ... postpone discussion to next week 17:19:04 -tlebo 17:19:06 -dgarijo 17:19:07 -khalidbelhajjame 17:19:07 -[IPcaller] 17:19:08 -Paolo 17:19:10 -sandro 17:19:16 -zednik 17:19:17 @dgarijo, will shut the system down, thank you! 17:19:21 -Luc 17:19:29 ok, thanks Luc. goodbye 17:19:38 bye bye 17:22:01 -??P32 17:35:00 disconnecting the lone participant, GK, in SW_(PROV)11:00AM 17:35:04 SW_(PROV)11:00AM has ended 17:35:11 Attendees were Luc, dgarijo, [ISI], [IPcaller], GK, +1.518.633.aaaa, +1.315.330.aabb, Paolo, tlebo, zednik, sandro, khalidbelhajjame, SamCoppens, jcheney 18:48:27 GK has left #prov