14:58:37 RRSAgent has joined #ldp 14:58:37 logging to http://www.w3.org/2013/02/25-ldp-irc 14:58:39 RRSAgent, make logs public 14:58:39 Zakim has joined #ldp 14:58:41 Zakim, this will be LDP 14:58:41 ok, trackbot, I see SW_LDP()10:00AM already started 14:58:42 Meeting: Linked Data Platform (LDP) Working Group Teleconference 14:58:42 Date: 25 February 2013 14:59:03 +JohnArwe 14:59:26 +cygri 14:59:27 + +1.214.537.aaaa 14:59:33 +SteveBattle 15:00:00 +Arnaud 15:00:21 zakim, who's here? 15:00:22 On the phone I see [IPcaller], JohnArwe, cygri, +1.214.537.aaaa, SteveBattle, Arnaud 15:00:22 On IRC I see RRSAgent, JohnArwe, Arnaud, svillata, TallTed, jmvanel, cody, Ruben, dret, SteveS, betehess, bhyland, bblfish, stevebattle, oberger, Yves, trackbot, sandro, ericP 15:00:23 +??P24 15:00:29 zakim, IPcaller is me 15:00:30 +dret; got it 15:00:31 +[OpenLink] 15:00:35 (1 214 537.aaaa is Cody, who hasn't learned to change Zakim's prompt from phone # to name) 15:00:38 Zakim, ??P24 is me 15:00:38 +svillata; got it 15:00:43 Zakim, [OpenLink] is OpenLink_Software 15:00:43 +OpenLink_Software; got it 15:00:47 Zakim, OpenLink_Software is temporarily me 15:00:47 +TallTed; got it 15:00:49 Zakim, mute me 15:00:49 TallTed should now be muted 15:01:21 +bblfish 15:01:35 Kalpa has joined #ldp 15:01:43 hi, in train from Paris to Amsterdam 15:02:23 Kalpa has left #ldp 15:02:23 zakim, who's here? 15:02:24 On the phone I see dret, JohnArwe, cygri, +1.214.537.aaaa, SteveBattle, Arnaud, svillata, TallTed (muted), bblfish 15:02:25 On IRC I see RRSAgent, JohnArwe, Arnaud, svillata, TallTed, jmvanel, cody, Ruben, dret, SteveS, betehess, bhyland, bblfish, stevebattle, oberger, Yves, trackbot, sandro, ericP 15:02:30 zakim, aaaa is cody 15:02:30 +cody; got it 15:02:35 afternoon! 15:02:59 chair: Arnaud 15:03:07 scribe: svillata 15:03:08 scribe: svillata 15:03:15 svillata: you can use this: http://www.w3.org/2006/07/SWD/Scribing.html 15:03:16 Kalpa has joined #ldp 15:03:28 thanks bblfish 15:03:34 nmihindu has joined #ldp 15:03:37 +1 15:03:46 Topic: approving minutes Feb 18 15:03:54 approved 15:04:05 SteveS has joined #ldp 15:04:11 +[IBM] 15:04:12 zakim, who is on the phone 15:04:12 I don't understand 'who is on the phone', Kalpa 15:04:27 zakim, [IBM] is me 15:04:27 +SteveS; got it 15:04:28 zakim, who is on the phone? 15:04:28 On the phone I see dret, JohnArwe, cygri, cody, SteveBattle, Arnaud, svillata, TallTed (muted), bblfish, SteveS 15:04:32 Arnaud: F2F is coming up 15:04:37 I'll be travelling 15:05:08 ..on the monday before the F2F 15:05:09 Arnaud: indicate your participation to F2F meeting 15:05:53 Topic: ISSUE-47 15:05:56 Issue-47? 15:05:56 ISSUE-47 -- publish ontology -- pending review 15:05:56 http://www.w3.org/2012/ldp/track/issues/47 15:06:07 +??P31 15:06:33 krp has joined #ldp 15:06:44 Zakim, ??P31 is me 15:06:44 +nmihindu; got it 15:06:48 Arnaud: do we want to close ISSUE-47? 15:06:53 q+ 15:06:55 -bblfish 15:07:00 q? 15:07:13 makes sense to close it if the actions are taken. ( I can't hear much breaks up a lot in the train ) 15:07:15 +roger 15:07:31 +Sandro 15:07:59 ack stevebattle 15:08:11 Should it not have a date pattern in the URL like most W3C published schemas? How to handle new versions? 15:08:21 close ISSUE-47 15:08:21 Closed ISSUE-47 publish ontology. 15:08:23 arnaud: we now have a turtle document in the cvs ... that seems like linked data "enough" 15:08:53 ...expect editors to update ontology based on future resolutions of issues 15:09:07 cody - those date patterns are associated with the start of the WGs, not the schemas 15:09:13 Topic: LDP specification and publishing a second draft 15:09:39 thx 15:09:43 roger has joined #ldp 15:10:07 Arnaud: we have to discuss what we think we need to do for p[ublishing the second draft 15:10:21 TallTed has changed the topic to: Linked Data Platform WG -- http://www.w3.org/2012/ldp/ -- current agenda: http://www.w3.org/2012/ldp/wiki/Meetings:Telecon2013.02.25 15:10:30 q+ 15:10:31 ... what do the editors need to publish a second draft? 15:10:41 q? 15:11:07 Kalpa has joined #ldp 15:11:24 SteveS: pretty good shape wrt the resolved issues 15:11:39 -nmihindu 15:11:41 +??P29 15:11:56 zakim, ??P29 is me 15:11:56 +krp; got it 15:12:07 +??P31 15:12:17 Arnaud: effort to link from the spec the issues 15:13:19 dret has joined #LDP 15:13:46 -??P31 15:13:50 concerning draft is the relative urls resolved? 15:14:09 Arnaud: would be good to have a week to review the spec? 15:14:13 bblfish: it is an open action, minor update we can do 15:14:16 +??P31 15:14:38 I'm happy to be transparent and publish internally and externally simultaneously. 15:14:49 ... start review, and for March 11 decide whether to publish it 15:15:05 q? 15:15:12 ack john 15:15:13 q+ 15:15:17 Zakim, ??P31 is me 15:15:17 +nmihindu; got it 15:15:20 ack steve 15:17:17 Arnaud: maybe next week spec will be in a good shape, and we can decide then whether to publish it 15:18:55 which issue are we discussing? 15:19:34 ok, thanks 15:19:44 q+ 15:20:29 Topic: composition (related to ISSUE-34) 15:20:59 +Sandro.a 15:21:03 JohnArwe: the ontology itself is subjet to change 15:21:06 -Sandro 15:21:11 Think this is more narrowly issue-32 and somewhat a part of it 15:21:18 ack stevebattle 15:21:59 stevebattle: issue-34 brings to an ontology about aggregation and composition 15:22:30 -nmihindu 15:23:00 +??P28 15:23:21 ashok's email: http://lists.w3.org/Archives/Public/public-ldp-wg/2013Feb/0014.html item 2 15:23:34 Zakim, ??P28 is me 15:23:34 +nmihindu; got it 15:23:49 Arnaud: proposal now is to have two subclasses for composition and aggregation 15:24:46 ... container is a useful notion independently from aggregation/composition 15:25:03 q+ 15:25:28 ... we are discussing how many classes to define, which properties 15:25:29 ack steves 15:26:19 q+ 15:26:21 q+ 15:26:27 ack roger 15:26:49 ISSUE-34? 15:26:49 ISSUE-34 -- Adding and removing arcs in weak aggregation -- closed 15:26:49 http://www.w3.org/2012/ldp/track/issues/34 15:27:07 ack stevebattle 15:27:28 stevebattle: important to make a distinction in the ontology 15:28:33 cygri has joined #ldp 15:28:50 http://lists.w3.org/Archives/Public/public-ldp-wg/2013Feb/0115.html 15:29:09 It would be good to get feedback from Richard about issue 34 (because he originally raised the issue). 15:29:16 Arnaud: email JohnArwe sent out on Friday with a proposal 15:29:50 SteveB: as long as real behavioral difference, happy to have different classes in ontology 15:29:52 roger: I believe cygri opened on behalf of us at F2F1…but would be good to get feedback, not arguing that 15:30:56 Proposal: adopting ontology proposed by JohnArwe (http://lists.w3.org/Archives/Public/public-ldp-wg/2013Feb/0115.html) 15:30:57 +bblfish 15:31:06 +1 15:31:15 bblfish has joined #ldp 15:31:24 +1 15:32:06 No - they have different deletion behaviour. 15:32:21 cygri: no idea of what the difference is 15:32:52 @cygri: the example in the email ontology is (as resolved in 34) currently the only difference between them. 15:32:52 I'd suggest changing :Aggregation to :aggregateContainer and :Composition to :compositeContainer 15:33:17 That sounds a bit verbose to me. 15:33:21 Arnaud: when you delete the container, different behavious about the deletion of the resources it contains 15:33:27 It's going to be used a lot 15:33:41 but otherwise I'm OK with the suggested change *as a start* ... I agree with cygri that the specific differences in behavior must be explicitly noted. 15:34:27 +??P33 15:34:52 back in new train 15:34:56 cygri: having two subclasses which differ only for a sentence does not make sense, my feeling is that just using the super-class would be sufficient 15:35:19 Zakim, ??P33 is me 15:35:19 +nmihindu; got it 15:35:33 Arnaud: parent1 is aggregation and the subclass it the composition1 15:35:54 the question I would have is what happens when something is changed from an Aggreation to a Container, especially concerning the members. 15:35:59 cygri: members may continue to exist is not a constraint 15:36:15 ... it doen't commit the server 15:36:20 q? 15:36:22 q? 15:36:25 q+ 15:36:25 q+ 15:36:31 q? 15:36:34 please see my question above: 15:36:35 Zakim, unmute me 15:36:35 TallTed should no longer be muted 15:37:06 Arnaud: how do we insert this aggregation concept? 15:37:17 q? 15:37:19 please see above 15:37:23 the question I would have is what happens when something is changed from an Aggreation to a Container, especially concerning the members. 15:37:45 q+ 15:38:08 ack me 15:38:20 I don't know if we'd allow a change in container behavior dynamically... new conversation? 15:38:24 ack TallTed 15:38:26 that (in my opinion) is a very dodgy thing 15:38:40 SteveS: we can open and issue and address the question of bblfish 15:39:20 q? 15:39:29 my guess is that this will only work if you add a :contains relation 15:39:48 Arnaud: we have to make concrete proposals 15:39:50 ack stevebattle 15:39:57 Ted: if (in the end) there is no behavioral difference between Container and AggregateContainer, would you like cygri want to collapse them? 15:40:12 stevebattle: cygri's proposal appealing 15:40:23 s/Ted:/Question for Ted:/ 15:40:54 Arnaud: changing containers to something else change the spec quite a lot 15:41:05 In OOD, composition is not (typically) a subclass of aggregation. They're commonly subclasses of association. 15:41:16 q? 15:41:20 s/change /changes 15:42:00 q? 15:42:18 Isn't Container an abstract superclass that is useful for property definitions? 15:42:28 +Sandro.aa 15:42:32 -Sandro.a 15:42:46 TallTed: propose to use aggregate containers and composite containers 15:43:07 ... superclass Container 15:43:17 q+ to ask a naive question (can't we just use URLs?) 15:43:18 stevebattle: agree, we can multi-type if we even wanted to say it is a ldp:Container and a ldp:Aggregation 15:43:29 q+ 15:43:33 Yes - agreed that Aggregation and Composition are mutually exclusive classes. 15:43:35 q? 15:43:44 TallTed: proposal to change aggregation VS composition into aggregate containers/composite containers 15:43:44 ack sandro 15:43:44 sandro, you wanted to ask a naive question (can't we just use URLs?) 15:44:25 I proposed that at the last F2F and got voted down :) 15:44:28 I think it is an interesting idea 15:44:31 ack steves 15:44:51 I was going to propose that urls ending in / are LDPCs 15:45:09 mmm, I don't like "urls ending in" 15:45:16 should be opaque 15:45:18 we spoke about this at the last F2F, but since then I have changed my mind. 15:46:02 Ruben, URLs are opaque as far as emantics goes, but in fact the URI spec does give / a special significance 15:46:09 q+ 15:46:14 s/emantics/semantics/ 15:46:22 ack cygri 15:46:44 cygri: one issue was the idea of using the url structure to indicate composition 15:47:22 ... can't give any special semantics to the relations to keep the implementation really simple 15:47:56 q+ 15:47:59 I see that, but I don't find that compelling, giving the simplicity provided. 15:48:09 q? 15:48:41 ack stevebattle 15:49:26 I probably voted against stevebattle at the F2F, but now that I see how long we've spent trying to figure this out, I lean more toward simplicity. 15:49:43 I can make a proposal 15:49:44 stevebattle: is it possible to re-open the issue? 15:49:52 q+ 15:50:02 q+ 15:50:14 Arnaud: better to re-open issues when new information comes 15:50:14 stevebattle: I have an idea on how to do this in a way that is uncontrovsial 15:50:19 ack sandro 15:50:19 q- 15:50:26 ack steves 15:50:28 ro was that Sandro 15:50:58 An aggregate could generate URIs at the same level at the aggregation. 15:51:15 sandro: I think it might be new information that this is so hard to us to figure out. 15:51:21 They wouldn't be nested below the Aggregation 15:51:42 ..In the URI structure 15:51:53 I think Sandro was proposing that "if the URL is structured ..., then the client Knows the behavior is delete (or not) members." 15:52:07 I think we are arguing over minor details of class hierarchy and not fundamental behavioral difference 15:52:09 sandro, we can should get together on this. 15:52:19 yes, JohnArwe 15:52:23 s/difference/differences/ 15:52:39 proposed: use John's proposed ontology with Aggregation renamed as AggregateContainer, Composition as CompositeContainer, and better documentation 15:52:45 Arnaud: TallTed proposal from JohnArwe proposal 15:52:50 in fact -- I probably shouldn't be in the lead or critical path for this 15:53:07 +0 (not convinced about the long names) 15:53:18 Arnaud: how do we feel with TallTed proposal's? 15:53:20 +1 15:53:21 When we talk about URL structures yielding client assumptions, we'd be making it harder for any existing implementations to comply. 15:53:30 +0 (I go back to my +1 for JohnArwe's proposal) 15:53:40 +0 15:53:48 +0 15:53:51 +1 (rename things at will - I hate arguing over them, you'll win all the time ) 15:53:59 +0 15:54:01 +1 15:54:08 -0 not convinced that aggregate is needed. ted's names are an improvement 15:54:23 +0 15:54:36 vote on the original proposal? 15:54:39 Arnaud: we don't seem to have consensus 15:54:56 +/-0 15:54:59 -bblfish 15:55:17 Arnaud: JohnArwe proposal? 15:55:49 +1 (use namespaces for disambiguation) 15:56:52 I prefer the shorter local names - we don't need to append 'Container' 15:56:56 TallTed: what do you mean stevebattle as using namespaces for disambiguation? 15:57:24 yez 15:57:49 Proposal: go with the proposal by TallTed 15:57:52 s/z/s/ 15:57:59 sorry 15:58:21 Topic: LDP model section 15:58:56 resolved: go with John's proposal amended by Ted 16:00:59 Arnaud: maybe we should leave to the editors to choose among the two proposals 16:01:23 -cygri 16:01:23 Kalpa has left #ldp 16:01:30 q+ 16:01:39 ack stevebattle 16:02:03 stevebattle: the two proposals are materially the same, but I prefer Henry's proposal 16:02:22 yeah, that was just a proposal. 16:02:36 Arnaud: do we have any text to put in the second draft of the spec? 16:02:38 no complete text yet, but i can take an action for that. 16:03:48 agree that editors can take the pen, using the feedback that is there now 16:03:55 dret: we can write a complete section 16:04:12 in that case, can i have an action? 16:04:41 -SteveS 16:04:41 ACTION: dret to create complete section 16:04:41 Created ACTION-38 - Create complete section [on Erik Wilde - due 2013-03-04]. 16:04:43 -roger 16:04:45 Thanks, bye. 16:04:49 thanks everybody! 16:04:52 -cody 16:04:53 -TallTed 16:04:53 -SteveBattle 16:04:54 -Arnaud 16:04:54 -Sandro.aa 16:04:56 -svillata 16:04:56 -dret 16:04:56 One question 16:04:57 -krp 16:04:57 -JohnArwe 16:05:03 regarding the face to face coming up 16:05:17 Ruben has left #ldp 16:05:20 what's your q cody? 16:05:31 -nmihindu.a 16:05:34 The line opens at 2:00 AM - 12:00 PM Boston time. 16:05:44 Is this because of overseas participation? 16:05:55 And is that the actual meeting start/end time? 16:06:02 probably - and probably copied from F2F1 16:06:34 ...when it was in France. Usually they run 8 (or later) to 5 (or later) local time. 16:07:19 Just seems like a face to face hosted in the U.S. would require the overseas participants to join at the odd times. 16:07:26 Eric P one of the staff contacts made the arrangements - suggest email the list so he'll see your q and respond. 16:07:59 OK. Thx. 16:08:19 the assumption is most participants will be local, so local time is "it". I can attest to the effect you describe (I was in NY during the Lyon F2F) 16:09:32 ...local time also tends to dictate when rooms can be booked, when meals are available (espec in a case like F2F2 when it appears there will be no sponsors so lunch is a "go out and get it" thing) 16:10:20 I still think I am confused. 2:00 AM to start a meeting in the U.S.? 16:10:31 disconnecting the lone participant, nmihindu, in SW_LDP()10:00AM 16:10:32 SW_LDP()10:00AM has ended 16:10:32 Attendees were JohnArwe, cygri, +1.214.537.aaaa, SteveBattle, Arnaud, dret, svillata, TallTed, bblfish, cody, SteveS, nmihindu, roger, Sandro, krp 16:10:43 TallTed has joined #ldp 16:11:07 hmm, I wish I knew who was 1.214.537.aaaa 16:11:17 That is Cody 16:11:23 ah, thanks 16:11:32 I do not know yet how to tell Zakim to use my name 16:11:39 zakim is supposed to learn over time 16:12:00 zakim, aaaa is cody 16:12:00 sorry, Arnaud, I do not recognize a party named 'aaaa' 16:12:15 214 537 is appears to be Richardson, TX 16:12:17 dunno if that helps. 16:12:41 Someone already said "zakim aaaa is cody", so maybe that is why the statement no longer works 16:12:43 cody is saying it's him 16:13:02 ah. i'm slow. 16:13:39 I think it's because the call is over 16:13:51 zakim, +aaaa is cody 16:13:51 sorry, Arnaud, I do not recognize a party named '+aaaa' 16:13:55 right 16:14:21 it's ok I can fix the minutes to reflect it anyway 16:14:39 Thx. 16:15:29 arnaud your transcript should show that we attributed aaaa to cody in zakim Very Shortly after he joined. he said he did not know how to do so, so I did it. 16:15:50 ok 16:16:10 remember that zakim for attendees unions them all together. I forget if the minuting script collapsing resolved aliases or not. 16:18:06 cody, wrt to the 0200 start that is Very Likely wrong, copied from Lyon (where 0800 CET would be 0200 ET) 16:19:01 ...hence: email to list on it. EricP presumably will then check whatever he booked at MIT and make Zakim's times align, then reflect that on the page (correctly) 16:19:12 OK 16:19:28 Is there a private list email? I seem to only have the public-ldp@ 16:20:38 The charter says the group will work in public, so that's the main list. There is also member-ldp-wg for confidentail stuff like phone numbers, but that's rarely used. 16:20:39 there are two lists: public-ldp and public-ldp-wg 16:20:50 all our emails are public. there is another list (public) for non-members to append to if needed. 16:20:56 (and you are on member-ldp.wg too.) 16:21:00 Ok- got it. Thanks! 16:21:59 as a member you can post to either list 16:22:03 cody: you in vegas next week? 16:22:14 non members can subscribe to both but only post to public-ldp 16:24:44 No. I'm in Dallas/Fort Worth next week. Was unaware of Vegas. (Sorry, I am just really, really green at this). 16:25:25 What is going on in Las Vegas? IBM conf? 16:25:29 cody: (2) I also see you posed a question in IRC that may have been missed. Short answer on dates is that the month/year gets added very close to the end, because they are taken from the date it hits Rec. Until then all ns values we own should be thought of as provisional. 16:26:02 cody: (1) yeah Pulse Conf. if you were going to be there would be an opp for F2F meeting was the thought. NP. 16:27:13 Got it on the URL. Thanks. And enjoy the conference! 16:27:36 cody: (2) ...also the email contents were an excerpt; in the ttl file in mercurial the ns we're using for now is . 16:27:48 gavinc has joined #ldp 16:52:06 jmv has joined #ldp 17:37:33 bblfish has joined #ldp 17:52:41 cygri has joined #ldp 18:14:53 Zakim has left #ldp