14:57:51 RRSAgent has joined #ldp 14:57:51 logging to http://www.w3.org/2013/02/04-ldp-irc 14:57:53 RRSAgent, make logs public 14:57:53 Zakim has joined #ldp 14:57:55 Zakim, this will be LDP 14:57:55 ok, trackbot; I see SW_LDP()10:00AM scheduled to start in 3 minutes 14:57:56 Meeting: Linked Data Platform (LDP) Working Group Teleconference 14:57:56 Date: 04 February 2013 14:58:09 chair: Arnaud 14:59:07 cygri has joined #ldp 14:59:29 SW_LDP()10:00AM has now started 14:59:34 Ashok has joined #ldp 14:59:35 +cygri 14:59:49 +??P14 14:59:55 zakim, who is on the phone? 14:59:55 On the phone I see cygri, ??P14 15:00:01 zakim, ??P14 is me 15:00:01 +AndyS; got it 15:00:02 rgarcia has joined #ldp 15:00:02 +SteveBattle 15:00:06 +[IPcaller] 15:00:10 BartvanLeeuwen has joined #ldp 15:00:16 +??P18 15:00:21 zakim, code? 15:00:21 the conference code is 53794 (tel:+1.617.761.6200 sip:zakim@voip.w3.org), Ashok 15:00:54 +Arnaud 15:00:56 -[IPcaller] 15:00:56 +??P22 15:01:09 nmihindu has joined #ldp 15:01:09 +??P24 15:01:18 hi 15:01:24 zakim, ??P22 is me 15:01:24 +rgarcia; got it 15:01:27 Kalpa has joined #ldp 15:01:30 +[IPcaller] 15:01:34 zakim, kalpa is with me 15:01:34 +kalpa; got it 15:01:35 +??P27 15:01:42 +[IBM] 15:01:43 Zakim, ??p27 is me 15:01:43 +BartvanLeeuwen; got it 15:01:46 +OpenLink_Software 15:01:47 Zakim: [IPcaller] is me 15:01:47 +[IPcaller.a] 15:01:48 dret has joined #LDP 15:01:55 Zakim, [IBM] is me 15:01:55 +SteveS; got it 15:02:02 Zakim, OpenLink_Software is temporarily me 15:02:04 +TallTed; got it 15:02:05 Zakim, mute me 15:02:05 TallTed should now be muted 15:02:26 +??P28 15:02:28 Zakim: [IPcaller] is me 15:02:32 +bblfish 15:02:45 hi 15:02:54 zakim, who's here? 15:02:54 On the phone I see cygri, AndyS, SteveBattle, ??P18, Arnaud, rgarcia, ??P24, [IPcaller], BartvanLeeuwen, SteveS, TallTed (muted), [IPcaller.a], ??P28, bblfish 15:02:57 roger has joined #ldp 15:02:57 Dave_Snelling_Fujtsu has joined #ldp 15:02:58 cygri has cygri, kalpa 15:02:58 On IRC I see dret, Kalpa, nmihindu, BartvanLeeuwen, rgarcia, Ashok, cygri, Zakim, RRSAgent, sergio, TallTed, AndyS, betehess, SteveS, Arnaud, stevebattle, jmvanel, bblfish, 15:02:58 ... trackbot, Yves, sandro, ericP 15:03:05 zakim, +??P28 is me 15:03:05 sorry, dret, I do not recognize a party named '+??P28' 15:03:08 +Sandro 15:03:11 zakim, ??P28 is me 15:03:12 +dret; got it 15:03:16 Zakim: [IPcaller] is me 15:03:21 Do you all remember when Arnaud used to to be worried there was not enough conversation going on the mailing list? 15:03:24 zakim, ??18 is me 15:03:24 sorry, Ashok, I do not recognize a party named '??18' 15:03:31 +Yves 15:03:33 We'll use a shorter nick name in future. :-) 15:03:34 Zakim, who's here? 15:03:35 On the phone I see cygri, AndyS, SteveBattle, ??P18, Arnaud, rgarcia, ??P24, [IPcaller], BartvanLeeuwen, SteveS, TallTed (muted), [IPcaller.a], dret, bblfish, Sandro, Yves 15:03:37 cygri has cygri, kalpa 15:03:37 On IRC I see Dave_Snelling_Fujtsu, roger, dret, Kalpa, nmihindu, BartvanLeeuwen, rgarcia, Ashok, cygri, Zakim, RRSAgent, sergio, TallTed, AndyS, betehess, SteveS, Arnaud, 15:03:37 ... stevebattle, jmvanel, bblfish, trackbot, Yves, sandro, ericP 15:03:54 Dave_Snelling_Fujtsu - you can change nick now, with `/nick MyNewNick` 15:03:56 zakim, ??P18 is me 15:03:57 +Ashok; got it 15:04:05 +roger 15:04:09 mute bblfish 15:04:15 Zakim: [IPcaller] is me 15:04:19 zakim, mute bblfish 15:04:20 bblfish should now be muted 15:04:26 JohnArwe has joined #ldp 15:04:51 +JohnArwe 15:05:19 +??P8 15:05:25 zakim, unmute bblfish 15:05:25 bblfish should no longer be muted 15:05:32 -[IPcaller] 15:05:44 i can do it for 30min before yves takes over... 15:05:59 +[IPcaller] 15:06:11 Zakim: [IPcaller] is me 15:06:13 q? 15:06:27 Ok, I can do it 15:06:46 srcibe: dret 15:06:55 s/srcibe/scribe/ 15:07:08 chair: Arnaud 15:07:13 scribenick: dret 15:07:18 agenda: http://www.w3.org/2012/ldp/wiki/Meetings:Telecon2013.02.04 15:07:38 Zakim: [IPcaller] is sergio 15:07:49 +1 15:07:50 +1 15:07:55 topic: Adminstrative Stuff 15:07:56 zakim, ??P8 is me 15:07:56 +nmihindu; got it 15:07:56 zakim, [IPcaller] is sergio 15:07:57 +sergio; got it 15:08:04 thx cygri 15:08:16 minutes of january 28 are approved (no objections) 15:08:38 what is the URL for the registration? 15:08:39 next meeting next week, please register for f2f2 15:08:46 topic: Actions 15:08:55 dret, http://www.w3.org/2012/ldp/wiki/F2F2 15:08:56 there are no actions pending review 15:09:05 thanks 15:09:22 thanks, cygri 15:09:44 http://www.w3.org/TR/2013/WD-ldp-ucr-20130131/ 15:10:34 LDP UCR FPWD was published (but there wasn't an open action) 15:10:42 topic: Issues 15:11:24 ISSUE-46 raised by roger should result in a wiki page 15:11:39 ISSUE-46? 15:11:39 ISSUE-46 -- services and LDP -- raised 15:11:39 http://www.w3.org/2012/ldp/track/issues/46 15:11:41 q+ 15:11:46 q? 15:12:09 svillata has joined #ldp 15:12:35 maybe ISSUE-46 should be closed and a wiki page created as a "wish list"? roger agrees, stevebattle assists with making sure things get into UCR 15:12:42 if issue-46 is handled via wiki page or any other on-line resource, let's link 46 to that target site 15:12:48 close ISSUE-46 15:12:48 Closed ISSUE-46 services and LDP. 15:12:52 q+ 15:12:55 q- 15:13:11 +??P0 15:13:20 ISSUE-47 15:13:20 ISSUE-47 -- publish ontology -- raised 15:13:20 http://www.w3.org/2012/ldp/track/issues/47 15:13:46 Zakim, ??P0 is me 15:13:46 +svillata; got it 15:13:55 proposed: open issue-47 15:13:56 +1 15:13:57 +1 15:13:59 +1 15:14:00 +1 15:14:00 +1 15:14:01 +1 15:14:02 +1 15:14:12 +1 15:14:12 +1 15:14:14 +1 15:14:14 +1 15:14:21 +1 15:14:31 open ISSUE-47 ; bblfish will become the "editor" of the ontology 15:14:46 resolved: open issue-47 15:14:57 The minimal ontology on the issue page looks good though 15:14:57 ISSUE-48 15:14:57 ISSUE-48 -- Profile mechanism is Needed -- raised 15:14:57 http://www.w3.org/2012/ldp/track/issues/48 15:15:27 q+ 15:15:40 Proposal: close issue-48 as out of scope for this WG 15:16:29 agree - does not need spec'ing - server can enforce anyway. 15:16:30 +1 15:16:33 +1 15:16:36 +1 15:16:40 +1 15:16:40 +1 15:16:41 +1 15:16:41 +1 15:16:45 need use cases from owner 15:16:55 close ISSUE-48 15:16:55 Closed ISSUE-48 Profile mechanism is Needed. 15:16:58 could come later 15:16:59 +1 15:17:00 +1 15:17:05 resolved: close issue-48 15:17:38 Thanks for getting that published Yves. 15:17:54 UCR have been published, thanks everybody for making this happen 15:18:00 http://www.w3.org/TR/2013/WD-ldp-ucr-20130131/ 15:18:02 http://www.w3.org/TR/ldp-ucr/ 15:18:12 topic: UCR 15:18:18 topic: LDP spec 15:18:23 The link to the published UC&R is broken on the wiki 15:18:34 It returns 404 15:18:34 topic: Open Issues 15:18:51 stevebattle, it's a wiki ;-) 15:18:58 :) 15:19:17 ISSUE-9 15:19:17 ISSUE-9 -- Should properties used in BPR representations be BPRs? -- open 15:19:17 http://www.w3.org/2012/ldp/track/issues/9 15:19:56 fixed (ldp-ucr link) 15:19:59 TallTed has joined #ldp 15:20:22 http://lists.w3.org/Archives/Public/public-ldp-wg/2012Dec/0100.html 15:20:32 should we accept rgarcia's proposal? 15:20:40 proposed: close issue-9, accepting Raul's proposal 15:20:43 Ah yes, so this was my test: "Predicate URIs used in BPR representations should be HTTP URLs which when dereferenced provide the definition of said term." 15:21:02 arnaud proposes to close the issue, accepting this proposal 15:21:17 q+ 15:21:20 q- 15:21:59 cygri asks about plans for a best practices document, maybe separate from the specification 15:22:13 Arnaud points to the deployment guide 15:22:15 http://www.w3.org/2012/ldp/wiki/Deployment_Guide 15:22:27 -??P24 15:22:49 cygri points out that we shouldn't require things that are not necessary, and we shouldn't overconstrain the protocol 15:23:07 Yes - I agree it's best practice. 15:23:10 -Ashok 15:23:39 q+ 15:23:43 ack me 15:23:55 ack bblfish 15:23:58 maybe we could always keep SHOULD's in the spec? 15:24:08 But agree they certainly shouldn't be LDP Resources. 15:24:12 q+ 15:24:25 ack cygri 15:24:27 bblfish agrees with cygri 15:25:15 q+ 15:25:29 f2f discussion around use of datatypes: everybody said maybe forbidding datatypes is not such a great idea, we shouldn't do it if this doesn't break anything 15:25:44 Proposed: close issue-9, this is really best practice and should be moved out of the spec into the deployment guide 15:25:53 as a result of the f2f, the datatypes were moved to the deployment guide 15:25:59 ack sergio 15:26:51 q+ 15:26:54 sorry sergio, i have a hard time understanding you. could you summarize on IRC for the minutes? thanks. 15:26:58 -[IPcaller.a] 15:27:00 ack ragarcia 15:27:09 ack rgarcia 15:27:10 This proposal should state that the text of 4.1.8 should change. 15:27:12 +[GVoice] 15:27:26 the point is that there are many open points at the spec, so maybe we should combine it with some best practices 15:27:30 +1 15:27:32 +1 15:27:34 +1 15:27:37 +1 15:27:40 +1 15:27:42 +1 15:27:47 +1 This is general LD good practice, not LDP specific. 15:27:47 +1 15:27:54 arnaud proposed to close issue-9, straw poll about this 15:27:57 +1 15:27:58 +1+0 15:27:58 +1 15:28:03 +1 15:28:04 oops +0 15:28:12 +??P2 15:28:23 close ISSUE-9 15:28:23 Closed ISSUE-9 Should properties used in BPR representations be BPRs?. 15:28:34 zakim, ??P2 is me 15:28:34 +Ashok; got it 15:28:42 ISSUE-10 15:28:42 ISSUE-10 -- Include clarifications and guidance around ETags -- open 15:28:42 http://www.w3.org/2012/ldp/track/issues/10 15:28:45 resolved: close issue-9, this is best practice, move text to deployment guide 15:29:26 some discussions on the mailing list, rgarcia presents an evolved proposal 15:30:02 There was something weak etags which sounds good to me 15:30:09 Proposal I believe is within here http://lists.w3.org/Archives/Public/public-ldp-wg/2013Feb/0035.html 15:30:14 what are the different ways to go? 15:30:59 q+ 15:31:00 rgarcia says there's agreement to weak ETags, but HTTP/1.1 requires strong ETags for certain functions 15:31:10 ack bblfish 15:31:28 -Ashok 15:31:34 From Prierre-Antoine: "The server MAY provide a strong etag (ref), but only if it can guarantee that the same graph will always be serialized the exact same way (byte-wise). This is not always the case, as the order of triples or blank node labels are not significant in RDF and may vary across serializations. If the server can not ensure that, the etags it provides MUST be weak etags (ref)." 15:31:35 q+ 15:31:35 q+ 15:31:55 bblfish suggests that this info should be put into the spec 15:31:57 q+ 15:31:59 q+ 15:32:09 ack dret 15:32:09 ack dret 15:32:29 My question was "Do we need to specify how the server creates etagas" 15:32:45 Ashok, I don't think so 15:32:53 +1 15:32:53 dret++ 15:32:56 ack steves 15:32:57 +[IPcaller] 15:33:09 q+ 15:33:15 I recommend that people read https://tools.ietf.org/html/draft-ietf-httpbis-p4-conditional-21#section-2.1 15:33:24 my suggestion is to link to the specs for informative purposes 15:33:25 no I don't htink so 15:33:32 ack ashok 15:33:32 q+ 15:33:49 ashok asks whether we need to specify how ETags are created 15:33:53 ack stevebattle 15:34:12 stevebattle says we should say something, but maybe best practices 15:34:17 ack dret 15:34:34 -[IPcaller] 15:34:42 Best practice is going to be bigger than the spec :) 15:34:48 xD 15:34:55 q+ to ask if the spec needs to say whether weak ETags are sufficient to identify identical semantic content 15:35:43 ack ericP 15:35:43 ericP, you wanted to ask if the spec needs to say whether weak ETags are sufficient to identify identical semantic content 15:35:43 qq+ 15:35:47 yves points out that HTTPbis already talks about this, so maybe just link to that 15:35:48 q+ 15:37:13 arnaud points out the specific text of the issue: should we leave the spec like it is, or does it need a change? 15:37:32 ack stevebattle 15:37:36 arnaud says people seem to converge towards not changing the spec 15:38:11 stevebattle says that HTTPbis maybe not have specific text about ETags for RDF media types 15:38:15 i think that a decision like "identical weak ETags indicate identical graphs" still needs to go on the LDP spec 15:38:38 q+ 15:38:43 that seems like a normative behavior that folks will need to be able to count on and test 15:39:01 more in "if weak etags are used, equality of weak etags MUST mean identical graphs" 15:39:23 ack rgarcia 15:39:28 we cannot change HTTP at that level (you're kind of inventing "strong ETags specifically for RDF content") 15:39:31 then leave the use and creation od etags (weak or not) that to implementation detail 15:40:00 Ruben has joined #ldp 15:41:08 that sounds interesting. 15:41:36 not sure how well that would work for clients that just talk HTTP 15:41:59 ISSUE-10 remains open for now 15:42:09 http://www.w3.org/2012/ldp/wiki/Deployment_Guide 15:42:15 ok 15:42:17 thanks 15:42:32 topic: ISSUE-37 (The Model) 15:43:15 +1 to specific textual proposals 15:43:17 ! 15:43:59 http://www.w3.org/2012/ldp/wiki/ISSUE-37 has been growing 15:44:23 no discussions around http://www.w3.org/2012/ldp/wiki/ISSUE-37#Proposed_Spec_Section_.28for_the_LDP_Spec.29 on the mailing list so far 15:44:23 +[IPcaller] 15:44:32 makes sense 15:44:41 Is the "Home Resource" and "Containers" section part of which proposal? 15:45:56 +1 SteveS how do I read the HR/Containers section, as Erik's, Henry's, neither? 15:46:04 +q 15:46:08 everything from the linked section to the end of the page is intended to be the text 15:46:43 ack roger 15:46:50 the "Erik Wilde", "Henry Story" chunks - are those alternatives? mut excl, additive? 15:47:00 bblfish wants to se a better definition of the resource that are made available 15:47:04 that's exclusive 15:47:42 q+ 15:48:02 oops, sorry, i didn't see the changed structure so far; i think bblfish's section is just for the intro section... 15:48:06 ack steves 15:48:10 ( mind you we should discuss that ) 15:48:36 steves points out that the text currently is a bit confusing (mixed authors) 15:49:00 +1 to that 15:49:01 A link is a URI? 15:49:12 please mail proposals on the mailing list 15:49:19 you're asking me, AndyS 15:49:31 ok. 15:49:37 i'll try to keep up with change proposals as they come in... 15:49:37 ok 15:49:44 issue-34? 15:49:44 ISSUE-34 -- Adding and removing arcs in weak aggregation -- open 15:49:44 http://www.w3.org/2012/ldp/track/issues/34 15:49:50 topic: ISSUE-34 15:50:12 generally - I am unclear everyone is talking about the same thing. May be over-solving some issues. 15:50:28 http://www.w3.org/2012/ldp/wiki/ISSUE-34 15:50:34 http://lists.w3.org/Archives/Public/public-ldp-wg/2013Feb/0014.html 15:50:51 AndyS, if yes: a link is a URI with a link relation (a type), i.e. it is something you follow because you want to do something in the protocol flow. 15:50:56 q+ 15:51:22 ack john 15:51:27 Are you sure that was option 3? 15:51:54 I read option 3 as every direct member of a container is aggregated. 15:52:10 JohnArwe says this is mixing the interaction model and the way how to find out about the interaction model while you're using it 15:52:12 That's a "no" then :-) URIs aren't typed. Two concepts of "link" RDF/LD link = URI. Part of a triple. Atom link - entry in a container = several triples. 15:52:30 Sorry - I confused myself 15:52:37 s/aggregated/composed/ 15:52:38 q+ 15:52:42 doh! 15:52:46 ack ashok 15:53:26 ashok is that a data or behavioral question in your view? 15:53:32 q+ 15:53:48 q- 15:54:00 "class" I could hear as "onotology", when I think the distinction is behavioral (and any ontology is a consequence of that) 15:54:02 http://www.w3.org/2012/ldp/wiki/ISSUE-34 15:54:19 John, not sure I understand your question 15:54:53 +q 15:55:08 @ashok, when you said the first question is "one class or two", I'm trying to understand what You mean by 'class' there... rdfs:Class vs behvaioral 15:55:42 q+ 15:55:56 ack roger 15:55:58 Ah, I meant rdfs:Class 15:56:14 AndyS: a link in atom often could be seen as just one triple, but not just a URI. but it can have more complex information as well, such as media type information associated with the link as well "go to URI x if you want to edit this resource which is of media type y" 15:56:54 ack john 15:56:57 q+ 15:56:59 s/behva/behav/ 15:57:04 ack bbllfish 15:57:34 ack bblfish 15:57:39 q+ to ask for motivating use cases for containership (presuming that still means automagic deletion) 15:57:47 q+ 15:58:00 bblfish just wanted to highlight that there are several ways to do aggregation 15:58:18 ack ericP 15:58:18 ericP, you wanted to ask for motivating use cases for containership (presuming that still means automagic deletion) 15:58:50 -TallTed 15:59:34 ack steves 15:59:46 dret, I prefer the submission style-design. additional complexity make LDP harder to use ; different to link in other LD (and HTML for that matter) [but I'd drop the recursive delete as not cool URIs but that's fine tuning] 15:59:50 +q 16:00:40 SteveS points out that there are different interaction models underlying the proposals of bblfish and JohnArwe 16:00:59 ack roger 16:01:01 Arnaud proposes to start from JohnArwe's proposal 16:01:58 sounds good 16:01:58 -cygri 16:01:59 I'm not sure about that either. 16:02:17 0 16:02:19 0 16:02:21 0 16:02:21 Link to John's proposal for issue 34? 16:02:25 +1 16:02:28 -1 16:02:30 0 16:02:30 0 16:02:38 http://www.w3.org/2012/ldp/wiki/Issue-34:_Back_to_Basics 16:02:41 proposal: let's pick JohnArwe's proposal for ISSUE-34 16:02:43 +1 16:02:44 +1 16:02:46 Kalpa has left #ldp 16:02:51 Ta - capitalization:-( !!!! 16:02:55 I think it's worth looking at it in more detail for next week 16:03:01 straw poll abandonded, will be help next week 16:03:10 -bblfish 16:03:15 By 16:03:15 Ruben has left #ldp 16:03:16 -roger 16:03:17 -BartvanLeeuwen 16:03:17 -SteveS 16:03:18 -sergio 16:03:19 bye 16:03:19 -[IPcaller] 16:03:21 -svillata 16:03:21 -Arnaud 16:03:22 -JohnArwe 16:03:26 -Yves 16:03:30 yes 16:03:35 SteveS_ has joined #ldp 16:03:36 -AndyS 16:03:46 -Sandro 16:03:55 -rgarcia 16:04:23 AndyS has left #ldp 16:06:20 SteveS__ has joined #ldp 16:12:37 You should be using if-match and if-none-match with etags 16:18:00 Also, if we only support weak etags for RDF we have to make sure that PATCH operates on the semantics rather than on the syntax. 16:18:42 bhyland has joined #ldp 16:19:28 I'm not sure layering on http changes it though 16:23:15 how do we close this to start generating the minutes? 16:24:24 gonna cost you 16:24:39 :-) https://cgi.w3.org/member-bin/wiki_scribe 16:25:19 CGI! oh sweet nostalgia! 16:25:40 it'll still be CGI, just wont be on a machine called CGI. :-) 16:26:34 -SteveBattle 16:27:25 Zakim, who is here? 16:27:25 On the phone I see dret, nmihindu, ericP 16:27:26 On IRC I see bhyland, SteveS, TallTed, roger, dret, nmihindu, cygri, Zakim, RRSAgent, sergio, betehess, Arnaud, stevebattle, jmvanel, bblfish, trackbot, Yves, sandro, ericP 16:30:31 stevebattle2 has joined #ldp 16:31:59 https://dvcs.w3.org/hg/ldpwg/file/39114b935c70/tests/basic 16:35:41 ericP: currently I was thinking about testing my draft implementation in Apache Marmotta 16:36:56 I'll try to use those test suites 16:38:13 ericP: It would nice if we actually parameterize those URLs 16:38:26 ericP: but it looks great 16:42:56 hope for the next telco I'll have some tests running 16:43:12 according my interpretation, of course 16:43:15 :-S 16:47:22 https://dvcs.w3.org/hg/ldpwg/raw-file/default/tests/Overview.html#createdResource 16:47:45 SteveS has joined #ldp 16:47:54 https://dvcs.w3.org/hg/ldpwg/file/39114b935c70/tests/basic/NetWorth_1_4.trig 16:49:44 stevebattle has joined #ldp 16:50:46 ericP: for instance right now I don't have a clear strategy to create LDPR or LDPC 16:50:52 nmihindu: do you? 16:57:33 i made a guess and went for it 17:00:54 -dret 17:02:12 -nmihindu 17:06:05 ericP: summarizing? 17:06:20 I mean, to check if we are thinking about the same 17:07:13 disconnecting the lone participant, ericP, in SW_LDP()10:00AM 17:07:14 SW_LDP()10:00AM has ended 17:07:14 Attendees were cygri, AndyS, SteveBattle, Arnaud, rgarcia, kalpa, BartvanLeeuwen, SteveS, TallTed, bblfish, Sandro, dret, Yves, Ashok, roger, JohnArwe, nmihindu, sergio, svillata, 17:07:14 ... ericP, [IPcaller] 17:28:05 nmihindu has joined #ldp 17:35:40 sergio, umm, i sumarizing the conversation with dret about tests? 17:37:17 ok 17:37:25 let's tak tomorrow 17:37:32 s/tak/talk 17:41:11 TallTed has joined #ldp 18:05:36 AndyS has joined #ldp 19:05:07 SteveS has joined #ldp 19:05:27 SteveS has joined #ldp 19:15:59 bblfish has joined #ldp 20:12:58 SteveS_ has joined #ldp 20:23:03 Zakim has left #ldp 21:31:58 bhyland has joined #ldp 21:54:42 bblfish has joined #ldp 21:55:21 bblfish_ has joined #ldp 22:08:16 bhyland has joined #ldp 22:39:42 SteveS has joined #ldp