14:30:15 RRSAgent has joined #dawg 14:30:21 +Kendall_Clark 14:30:24 -??P0 14:30:25 +??P0 14:30:30 +DanC 14:30:38 +??P3 14:30:40 zakim, ??P3 is AndyS 14:30:40 +AndyS; got it 14:30:44 +Farrukh 14:30:56 +??P5 14:31:05 Zakim, ??P0 is SimonR 14:31:05 +SimonR; got it 14:31:09 hmm, 9:30 here in the DC part of the east coast :> 14:31:46 someone calling in from a kitchen :> 14:32:13 +??P6 14:32:40 JosD has joined #dawg 14:32:51 regrets from .. Hiroyuki Sato 14:33:02 regrets from Yoshio Fukushige 14:33:02 regrets Yoshio Fukushige 14:33:08 and Alberto Reggiori 14:34:43 minutes of prev meeting http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0210.html 14:34:53 FarrukhNajmi has joined #dawg 14:34:54 don't seem to be in agenda 14:34:57 + +3249456aaaa 14:34:57 Zakim, who's on the phone? 14:34:58 On the phone I see SimonR, Kendall_Clark, DanC, AndyS, Farrukh, JanneS, DaveB, +3249456aaaa 14:35:15 hi jos 14:35:38 we had regrets from SteveH earlier IIRC 14:36:14 0210 accepted as minutes 14:36:26 ericp is chair of next telcon 14:36:39 donm is 2004-11-16 14:36:58 Zakim, pick a scribe 14:36:58 Not knowing who is chairing or who scribed recently, I propose Kendall_Clark 14:36:59 I'm at risk because of XML 2004 14:37:08 Zakim, pick a scribe 14:37:08 Not knowing who is chairing or who scribed recently, I propose SimonR 14:37:15 scribe is SimonR 14:37:24 regrets: DanC for 16 Nov, 23 Nov, 30 Nov 14:37:38 (regrets for me 23, 30th also) 14:37:44 agenda + next SPARQL WD: timing, reviewers 14:37:47 Anyone attending XML 2004 in DC next week? 14:37:53 DanC: can you move Zakim onto new agendums? 14:38:03 -JosD 14:38:12 Zakim, take up agendum 1 14:38:12 agendum 1. "Convene, take roll, review agenda http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0244.html" taken up [from DanC] 14:38:33 Zakim, next agendum 14:38:33 agendum 1 was just opened, DanC 14:38:36 Zakim, close agendum 1 14:38:36 agendum 1 closed 14:38:37 I see 8 items remaining on the agenda; the next one is 14:38:38 2. Use Cases [from DanC] 14:38:39 Zakim, take up agendum 2 14:38:39 agendum 2. "Use Cases" taken up [from DanC] 14:38:50 Zakim, who's talking 14:38:50 I don't understand 'who's talking', DanC 14:38:58 Zakim, who is making noise? 14:39:08 DanC, listening for 10 seconds I heard sound from the following: SimonR (0%), Farrukh (33%), AndyS (10%), DanC (56%) 14:39:16 +JosD 14:39:59 EricP went to .. workshop got use case, not appeared so far 14:40:08 kendall work on stuff in 0205 14:40:12 Background here should be less 14:40:47 EricP action done 14:40:50 Zakim, next agendum 14:40:50 agendum 3. "ftf 4" taken up [from DanC] 14:41:14 re 2 - not much wg excitement re use case 14:41:58 TomAdams has joined #dawg 14:42:04 zakim, mute me 14:42:04 Kendall_Clark should now be muted 14:42:29 bummer: next f2f is same day as Bush's inauguration, which I was hoping to protest. Suck! 14:42:33 ACTION DanC: Make a dawg f2f4 registration form 14:42:39 zakim, unmute me 14:42:39 Kendall_Clark should no longer be muted 14:42:52 zakim, Tucana is TomAdams 14:42:52 sorry, TomAdams, I do not recognize a party named 'Tucana' 14:42:54 +[Tucana] 14:42:56 zakim, Tucana is TomAdams 14:42:56 +TomAdams; got it 14:43:05 JanneS to elaborate further on currency, other details in email to lsit 14:43:16 welcome TomAdams 14:43:50 what's the weather - could be 0C, or -20C says Jannes - bring a good jacket 14:43:54 whew! 14:44:10 helsinki to espoo - in rush hour, could be 20m 14:45:22 great to know 14:45:28 +10 14:45:30 slight pref to Helsinki, I heard from JanneS - walking around, things to see 14:45:35 Zakim, next agendum 14:45:35 agendum 4. "ftf 5" taken up [from DanC] 14:45:59 continue EricP action: 14:46:05 ACTION EricP: find logistics re F2F5 at tech plenary in Boston March 28 Feb - 4 Mar (2 days). Some WG preference to Mon/Tue of that week. some progress: 14:47:38 notes only TomA on both swbpd wg and da wg 14:47:42 also JosD, PatH 14:47:51 scheduling nightmare 14:48:11 Tom is happy to have an overlap, Dave Wood, Tucana's CTO is co-chair of SWBP 14:48:25 AndyS - swig meeting say about sparql 14:48:29 DanC - also on plenary day 14:48:31 Zakim, next agendum 14:48:31 agendum 5. "Protocol, Interoperability Demonstration Services" taken up [from DanC] 14:49:15 protocol draft 14:49:20 see http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0255.html 14:49:32 http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0255.html http://monkeyfist.com/kendall/sparql-protocol/ 14:49:58 too long to read now, for me 14:50:48 KC - didn't want to specify a min protocol with 1 method, got 7 methods, all orthoganal 14:51:00 trying to make it possible to get consensu on diffefrrent methods without affecting the others 14:51:29 reflects things we (mindlab) wants to do with rdf queries 14:52:26 abstract protocol (5) and a concrete one (7) with http 14:52:31 with examples for all the methods in http 14:52:36 at http://monkeyfist.com/kendall/sparql-protocol/#http 14:53:20 KC - deadset against hardocidng query parametsr for a number of reasons 14:53:48 [GRAPH_ID] is standing in for some param, not actually using that particular word 14:55:02 .. and probes an ally for bare wires 14:55:50 ACTION TomA, JanneS: read the protocol draft http://monkeyfist.com/kendall/sparql-protocol/ and email a review to the WG before next telcon 14:57:16 Zakim, next agendum 14:57:16 agendum 6. "SPARQL update, issues" taken up [from DanC] 14:57:38 jos: heh :> 14:58:45 eric's actions continued 14:59:01 ACTION EricP: supply definitions for SELECT (vars ordered or not?) ala http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0151.html 14:59:10 ACTION PatH review SPARQL def'ns post-publication 14:59:20 ACTION PatH: review SPARQL def'ns post-publication 14:59:25 ddd 14:59:32 ACTION AndyS: to write a test cases which use ? and $ in the same query 15:00:13 DanC asks for somebody else to help on comments list reading, responsing 15:00:31 AndyS - missing some things for recording for future work 15:00:32 see http://lists.w3.org/Archives/Public/public-rdf-dawg-comments/ 15:00:42 TomA happy to do that, but I'm half a slacker ;) 15:01:01 The work is bursty - current going quiet - next WD ... 15:01:07 ... then LC. 15:01:14 I can help with the next burst, I suspect. 15:01:23 TomC is the comment reading, responding person - will get poked by DanC if necessary :) 15:01:32 Zakim, next agendum 15:01:32 agendum 7. "SOURCE Issue" taken up [from DanC] 15:01:39 my msg on SOURCE http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0241.html 15:02:01 Zakim, who's on the phone 15:02:01 I don't understand 'who's on the phone', DanC 15:02:03 Zakim, who's on the phone? 15:02:06 On the phone I see SimonR, Kendall_Clark, DanC, AndyS, Farrukh, JanneS, DaveB, JosD, TomAdams 15:02:55 DanC read it, endorses it 15:03:06 AndyS read it, some questions and want a grip on the detail 15:03:13 s/a/to get a/ 15:03:19 questions to the list 15:04:17 Farrukh - there are use cases where the source of the data is significant; not read this email yet 15:04:28 JanneS - not digested yet 15:05:40 ACTION JanneS: review http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0256.html and check what happens when no source support is implemented 15:05:58 JosD - read it, issue when source is unbound 15:07:09 got issue when there is no binding 15:07:19 DB - need to see an example 15:08:09 KendallC - quickly reading it 15:09:45 my nomenclature "Operation Target Set" 15:09:52 since I have non-query operations 15:10:00 suggestion to change Query Graph 15:10:03 DB - to Data Graph 15:11:09 suggestions Federated Graph 15:11:12 for QG 15:11:12 "Aggregated" is the term we used in UC&R discussions 15:11:14 or Set 15:12:42 I have an implementation that (first pass) seems close but a little different 15:13:05 ACTION DaveB: update the test repository to record or amend tests to correspond to the http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0241.html proposal 15:13:36 ACTION DanC: suggest formal definitions for SOURCE 15:13:49 ... perhaps using larch tools 15:14:17 DaveB ACTION is done 15:14:34 Zakim, next agendum 15:14:34 agendum 8. "Disjunction issue" taken up [from DanC] 15:14:59 ^- above action is the agenda action for Daveb item5, done 15:15:14 SimonR been reading things... 15:15:32 and now LEFT OUTER JOIN - can't figure out what it means 15:16:44 with disj you generate NULLS - this var is not bound in this soln 15:16:46 SQL LEFT OUTER JOIN presumably 15:16:57 for a LOJ, you generate NULLS to preserve values which have 0 answers 15:17:04 - no idea what those NULLs mean 15:17:14 unless they mean the same things as those in disj(unction) 15:17:27 so maybe proving optional can't be used to implement disjunction 15:17:53 SimonR - cannot answer the last line as yes/no yet 15:18:17 as AndyS has written, has defined things as unbound 15:18:41 ericp, my action regards terms like "queried graph" and "graph name" and such 15:18:43 AndyS - is that not related to the optional scoping rule 15:19:05 LOJ is not defined algebratically 15:19:13 so is hard to combine with other things such as disj 15:19:19 danc, gotcha 15:19:20 ^-SR 15:19:44 SR - if I have an optional result, I can't look at each soln and interpret what it means. 15:19:56 For everything else, a solution like ?x=foo ?y=bar 15:20:05 with a LOJ, hard to work out what it means 15:20:46 having problem forming an example, or help in the literature to find an interpretation 15:21:59 ACTION Simon: explain how much of disjuction can be done with optionals, nested or otherwise. Point to references, problems with LEFT OUTER JOIN in the literature. 15:22:20 AndyS suggests looking at JosD's tests 15:22:36 JosD - keep the variable name in the soln 15:23:06 ... herberand universe ... 15:26:39 record is sketchy here, not following it 15:26:57 ACTION SteveH: own (i.e. propose resolution to) disjunction issue 15:27:03 ACTION SteveH: owns issue 'nested optionals' 15:27:10 Zakim, next agendum 15:27:10 agendum 9. "next SPARQL WD: timing, reviewers" taken up [from DanC] 15:27:26 The main suspicion I have is that the nulls generated by the left outer join sort-of-mean "there are no values for this variable", whereas the "unconstrained" markers generated by disjunction mean almost exactly the opposite, that the variable could take any value without invalidating the proposition. 15:28:39 simon, i'm curious how those differ 15:28:44 (practicallt) 15:28:48 (practically) 15:29:14 EricP: Well, imagine how equality works between them, joining, etc.... 15:30:03 schedule thoughts 15:33:49 simon, i've not seen a practical problem with SQL in which NULL != NULL 15:35:00 timeliness 15:35:58 Farrukh says around April 2005 would be a great time to have a CR sparql query. earliest. 15:36:10 Farrukh: the project I'm involved in, an open source implementation of EBXML, isn't likely to integrate SPARQL until April 2005... so a Candidate Recommendation at that point would be good. 15:36:33 ACTION DanC: notify SemWeb CG of EbXML possible sync point in April 15:36:54 DaveB, AndyS, SteveH, EricP potentially doing www2005 tutorial, may 2050 15:36:55 2005 15:37:02 i have same conference concerns, FWIW 15:37:27 Zakim, next agendum 15:37:27 I do not see any non-closed agenda items, DanC 15:37:44 (belated regrets from HowardK) 15:37:56 q+ To talk about spec quickly... 15:38:03 AndyS - UNSAID 15:38:12 ... want to raise the barrier to features coming in 15:38:52 not clear there is yet requirement for unsaid. WD1 said likely to be dropped 15:38:59 ... may become more likely 15:39:15 DanC will track this decision via an issue 15:39:41 ack TomAdams 15:39:50 TomC - spec question on going to CR 15:40:09 Kowari release early Dec, could start impl early Jan 15:40:15 -JosD 15:40:28 hmm, /me adds .AU to Expatriation Target Set :) 15:40:31 (Tucana) Dave wanted to know when things stable 15:40:40 UK likes long vacation over XMas as well. Not for the beach though 15:41:01 US *likes* long holiday over Xmas, it's just that our Evil Overlords don't permit it! 15:41:40 Timepoint: W3C TechPlenary 15:41:41 TA - April/May would suite for a CR, new software relase 15:45:12 yes, toys to play with is a big deal in getting feedback 15:46:10 ADJOURN. 15:46:11 reported my toy http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0174.html 15:46:14 -Farrukh 15:46:15 -JanneS 15:46:19 believes that Jan 2005 last call will be hard (is new language and not like 5 year old N3) 15:46:29 -DanC 15:46:30 -DaveB 15:47:08 Simon? 15:47:18 cheers all 15:47:24 AndyS: Yes? 15:47:38 EricP: See you later. 15:47:41 We said we'd talk about interoperablity 15:47:54 thanks much for the thoughts on the schedule, everybody. I was starting to worry alone, and starting to get grumpy as a result. 15:48:02 Sure. On the phone, or just IRC? 15:48:22 Phone (if Tom will allow it) 15:48:24 FarrukhNajmi has left #dawg 15:48:43 AndyS: We can queue up. :) 15:48:48 q+ 15:48:56 ack TomAdams 15:48:56 TomAdams, you wanted to talk about spec quickly... 15:49:42 Cough, cough 15:51:00 ack AndyS 15:51:16 q+ To hassles Simon to review the Kowari paper ;) 15:53:28 DaveB has joined #dawg 15:53:31 -Kendall_Clark 15:53:49 later, guys. i need lunch soon. :) 15:54:18 Good luck guys ;) Simon, if you're not too tired, please take a look at that Kowari paper I sent you. It needs to be submitted in a few hours. Thanks! 15:54:23 -TomAdams 15:56:29 http://www.sparql.org/ 15:57:00 http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0220.html 16:17:32 -SimonR 16:17:34 -AndyS 16:17:35 SW_DAWG()9:30AM has ended 16:17:36 Attendees were Kendall_Clark, DanC, AndyS, Farrukh, SimonR, JanneS, DaveB, +3249456aaaa, JosD, [Tucana], TomAdams 16:40:46 RRSAgent, make logs world-access 16:40:56 RRSAgent, bye 16:40:56 I see 12 open action items: 16:40:56 ACTION: DanC to Make a dawg f2f4 registration form [1] 16:40:56 recorded in http://www.w3.org/2004/11/09-dawg-irc#T14-42-33 16:40:56 ACTION: EricP to find logistics re F2F5 at tech plenary in Boston March 28 Feb - 4 Mar (2 days). Some WG preference to Mon/Tue of that week. some progress: [2] 16:40:56 recorded in http://www.w3.org/2004/11/09-dawg-irc#T14-46-05 16:40:56 ACTION: EricP to supply definitions for SELECT (vars ordered or not?) ala http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0151.html [3] 16:40:56 recorded in http://www.w3.org/2004/11/09-dawg-irc#T14-59-01 16:40:56 ACTION: PatH to review SPARQL def'ns post-publication [4] 16:40:56 recorded in http://www.w3.org/2004/11/09-dawg-irc#T14-59-20 16:40:56 ACTION: AndyS to to write a test cases which use ? and $ in the same query [5] 16:40:56 recorded in http://www.w3.org/2004/11/09-dawg-irc#T14-59-32 16:40:56 ACTION: JanneS to review http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0256.html and check what happens when no source support is implemented [6] 16:40:56 recorded in http://www.w3.org/2004/11/09-dawg-irc#T15-05-40 16:40:56 ACTION: DaveB to update the test repository to record or amend tests to correspond to the http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0241.html proposal [7] 16:40:56 recorded in http://www.w3.org/2004/11/09-dawg-irc#T15-13-05 16:40:56 ACTION: DanC to suggest formal definitions for SOURCE [8] 16:40:56 recorded in http://www.w3.org/2004/11/09-dawg-irc#T15-13-36 16:40:56 ACTION: Simon to explain how much of disjuction can be done with optionals, nested or otherwise. Point to references, problems with LEFT OUTER JOIN in the literature. [9] 16:40:56 recorded in http://www.w3.org/2004/11/09-dawg-irc#T15-21-59 16:40:56 ACTION: SteveH to own (i.e. propose resolution to) disjunction issue [10] 16:40:56 recorded in http://www.w3.org/2004/11/09-dawg-irc#T15-26-57 16:40:56 ACTION: SteveH to owns issue 'nested optionals' [11] 16:40:56 recorded in http://www.w3.org/2004/11/09-dawg-irc#T15-27-03 16:40:56 ACTION: DanC to notify SemWeb CG of EbXML possible sync point in April [12] 16:40:56 recorded in http://www.w3.org/2004/11/09-dawg-irc#T15-36-33