13:59:18 RRSAgent has joined #sparql 13:59:18 logging to http://www.w3.org/2010/09/14-sparql-irc 13:59:33 Zakim has joined #sparql 13:59:57 Zakim, this is sparql 14:00:02 trackbot, start meeting 14:00:04 RRSAgent, make logs world 14:00:05 ok, kasei; that matches SW_(SPARQL)10:00AM 14:00:06 Zakim, this will be 77277 14:00:07 Meeting: SPARQL Working Group Teleconference 14:00:07 Date: 14 September 2010 14:00:11 + +33.4.92.38.aaaa 14:00:15 ok, trackbot; I see SW_(SPARQL)10:00AM scheduled to start now 14:00:18 chair: Axel Polleres 14:00:24 Zakim, who is on the phone? 14:00:26 Zakim, aaaa is me 14:00:33 I notice SW_(SPARQL)10:00AM has restarted 14:00:33 agenda: http://www.w3.org/2009/sparql/wiki/Agenda-2010-09-14 14:00:35 On the phone I see AxelPolleres, kasei, +33.4.92.38.aaaa 14:00:39 +OlivierCorby; got it 14:00:41 +??P18 14:01:00 scribenick: kasei 14:01:00 scribe: Greg Williams 14:01:07 zakim, dial ivan-voip 14:01:07 ok, ivan; the call is being made 14:01:09 +Ivan 14:01:12 zakim, ??P18 is me 14:01:12 +AndyS; got it 14:01:26 Zakim, who is on the phone? 14:01:32 On the phone I see AxelPolleres, kasei, OlivierCorby, AndyS, Ivan 14:01:37 + +1.603.897.aabb 14:01:48 zakim, aabb is me 14:02:09 +MattPerry; got it 14:02:53 Souri has joined #sparql 14:03:02 topic: admin 14:03:10 PROPOSED: Approve minutes at http://www.w3.org/2009/sparql/meeting/2010-09-07 14:03:26 AxelPolleres: minutes from last time? 14:03:40 RESOLVED: Approve minutes at http://www.w3.org/2009/sparql/meeting/2010-09-07 14:04:22 +??P42 14:04:22 ... next meeting. Both chairs gone next week. 14:04:28 + +1.617.245.aacc 14:04:28 ... need to find a chair. 14:04:30 + +1.603.897.aadd 14:04:36 -??P42 14:04:37 Zakim, ??P42 is me 14:04:41 ivan: sandro may be around, but unsure. 14:04:44 I could scribe next time 14:04:46 ACTION: Axel to clarify chairing next time 14:04:47 Created ACTION-309 - Clarify chairing next time [on Axel Polleres - due 2010-09-21]. 14:05:13 +[IPcaller] 14:05:13 topic: status of documents 14:05:13 s/unsure/I am not sure/ 14:05:39 I already had ??P42 as Guest P42 26632, AlexPassant 14:05:44 AxelPolleres: which documents are ready for publication. what issues remain? 14:05:47 q+ 14:05:50 ... want to assign reviewers. 14:05:55 ... start with query document. 14:06:16 Reviews by end of this call? :-) 14:06:20 ivan: good to know the times involved in reviewing 14:06:41 -[IPcaller] 14:06:47 ack ivan 14:06:49 +??P33 14:06:50 zakim, aacc is me 14:07:00 AxelPolleres: Query document. Andy sent a list of which sections are ready for review. 14:07:06 +LeeF; got it 14:07:09 AndyS: sent list of sections that had changed from 1.0 14:07:22 http://lists.w3.org/Archives/Public/public-rdf-dawg/2010JulSep/0376.html 14:07:58 q+ 14:08:02 AxelPolleres: which sections should we assign for review? Can we review some sections now? 14:08:25 ... are there stable parts reviewable now? 14:08:39 AndyS: thought we were aiming for working draft publication out. If so, then the whole document is reviewable. 14:08:43 + +44.186.528.aaee 14:08:52 Zakim, +44.186.528.aaee is me 14:08:52 +bglimm; got it 14:09:13 AxelPolleres: with query, we probably need another WD publication. will anything happen before we publish/review? 14:09:30 ack ivan 14:09:44 ivan: I thought we were talking about LC reviews. 14:09:50 Zakim, mute me 14:09:50 bglimm should now be muted 14:09:54 ... now I hear something different. what are we talking about? 14:10:21 http://www.w3.org/2009/sparql/meeting/2010-08-24#publication_schedule 14:10:29 http://www.w3.org/2009/sparql/meeting/2010-08-10#document_publishing_status__2f_plans 14:10:56 AxelPolleres: I understood in our meeting on Aug 24 that we suggested the way forward is to have editors put missing pieces in, go with one more round of WD 14:11:16 ... we'll incorporate reviews from outside. 14:11:23 ... try to get the documents out, and in parallel get LC ready. 14:11:29 ... is that just my understanding? 14:11:58 ivan: that's fine if it's the plan. If we are looking for LC reviewers, then this should happen only when the whole document is ready. 14:12:08 ... reviewing just one part of a document should not happen. 14:13:01 AxelPolleres: idea was to have missing pieces in [by now?], but that hasn't happened yet. 14:13:13 ... could have WG reviewers for the parts that are done. 14:13:28 ... want an indication of what parts are finished, what's unstable. 14:13:46 ... the list AndyS sent around is what has changed, not what is ready. 14:13:57 AndyS: indication to people what sections are relevant. 14:14:05 AxelPolleres: can you provide a list of what's ready/stable? 14:15:00 ... for going forward, we can assign reviewers to go over it in the next 1/2 weeks? 14:15:27 AndyS: isn't clear to me if the reviews right now are for LC or just WD? 14:15:47 AxelPolleres: can we get 2 people to review document in current state? 14:15:52 I can 14:15:55 I can too 14:16:08 AxelPolleres: Greg and Matt will review. 14:16:39 AxelPolleres: looking for having the documents ready to publish. we should also think about the date we want to publish. 14:16:53 ... is one week possible? 14:16:59 2 week is better 14:17:03 I can probably do 1, but 2 better. 14:17:12 ... let's discuss in 2 weeks. 14:17:30 ... we'll try for 2 weeks for all documents. 14:17:55 ACTION: greg to review queryt for WD publication 14:17:55 Created ACTION-310 - Review queryt for WD publication [on Gregory Williams - due 2010-09-21]. 14:18:10 ACTION: matt to review query for WD publication 14:18:10 Sorry, couldn't find user - matt 14:18:26 AxelPolleres: Update document. 14:18:49 AlexPassant: the formal model is not complete yet. 14:18:55 ... review could start by Monday next week. 14:19:10 ... formal model hasn't been reviewed yet. maybe 2 reviewers plus one just for the model. 14:19:34 AxelPolleres: is the document ready from your point of view? are all issues fixed? 14:19:51 AlexPassant: no, not yet. issues will be ready by Monday. will be able to say on monday if ready for LC or if more time is needed. 14:20:15 AxelPolleres: we should have at least some reviews to get out a WD. 14:20:30 ... AlexPassant can inform reviewers when it is ready for review. 14:20:48 ... volunteers for reviewing update? 14:20:51 ACTION: axel to review update 14:20:51 Could not create new action (failed to parse response from server) - please contact sysreq with the details of what happened. 14:20:51 Could not create new action (unparseable data in server response: local variable 'd' referenced before assignment) - please contact sysreq with the details of what happened. 14:21:23 ACTION: AndyS to review update formal model 14:21:23 Created ACTION-312 - Review update formal model [on Andy Seaborne - due 2010-09-21]. 14:21:36 ACTION: axel to review Update 14:21:36 Created ACTION-313 - Review Update [on Axel Polleres - due 2010-09-21]. 14:21:57 (assumes Alex has had time to add more formal model) 14:22:00 ACTION: alex to inform andy/axel for review readyness 14:22:00 Created ACTION-314 - Inform andy/axel for review readyness [on Alexandre Passant - due 2010-09-21]. 14:22:12 AxelPolleres: Entailment document 14:22:15 Zakim, unmute me 14:22:20 bglimm should no longer be muted 14:22:28 bglimm: the document is ready for review. 14:22:36 q+ 14:22:40 ... want to add example for RDF-based semantics, but that's an informative section so doesn't require waiting. 14:23:07 AxelPolleres: made progress with sandro on issues around RIF. 14:23:26 ... progress, but will not have it in final shape for this round of WD. 14:23:42 ... should mark these points in the document, and will have to review again before LC. 14:23:58 ... can we assign reviewers for OWL and RDF parts? 14:24:19 ... does Chime still have open issues in the document? 14:24:31 ivan: still have to agree on URIs for the namespace document. 14:25:04 ... one issue is rdf import. another is what URI to use for entailment. 14:25:15 ... not sure where we are on that. 14:25:22 q- 14:25:25 ... everything leads to Chime. 14:26:00 ACTION: Birte to follow up with Chime on review readiness of entailment 14:26:01 Created ACTION-315 - Follow up with Chime on review readiness of entailment [on Birte Glimm - due 2010-09-21]. 14:26:23 I can review 14:26:28 AxelPolleres: does it make sense to look for entailment reviewers, to be ready whenever issues are dealt with? 14:26:48 ACTION: Olivier to review entailment whenever he gets ok from Chime/Birte 14:26:48 Created ACTION-316 - Review entailment whenever he gets ok from Chime/Birte [on Olivier Corby - due 2010-09-21]. 14:26:57 ... Olivier can review. Anybody else? 14:27:14 ... Zakim can pick a victim. 14:28:00 AxelPolleres: need reviewers from the WG. 14:28:03 Zakim, pick a victim 14:28:03 Not knowing who is chairing or who scribed recently, I propose AndyS 14:28:04 ... we'll try Zakim. 14:28:29 Zakim, pick a victim 14:28:29 Not knowing who is chairing or who scribed recently, I propose bglimm 14:28:39 Zakim, pick a victim 14:28:39 Not knowing who is chairing or who scribed recently, I propose AndyS 14:29:07 Zakim, pick Axel 14:29:07 I don't understand 'pick Axel', AndyS 14:29:13 (Zakim is being uncooperative.) 14:29:19 AxelPolleres: LeeF, can you review? 14:29:34 LeeF: yes, but don't feel qualified for some of it. 14:29:41 ACTION: Lee to review entailment 14:29:41 Created ACTION-317 - Review entailment [on Lee Feigenbaum - due 2010-09-21]. 14:29:57 AxelPolleres: HTTP Update 14:30:03 ... Chime isn't around, so can't make a decision today. 14:30:07 AxelPolleres: Service Description 14:30:24 ... any volunteers? 14:30:24 I volunteer for SD 14:30:36 ACTION: alex to review SD 14:30:36 Created ACTION-318 - Review SD [on Alexandre Passant - due 2010-09-21]. 14:31:43 AxelPolleres: is it ready for review? 14:32:02 I can review 14:32:10 kasei: yes, content is all reviewable. open issues for adding cross-links to other documents (e.g. protocol) 14:32:12 ACTION: birte to review SD 14:32:12 Created ACTION-319 - Review SD [on Birte Glimm - due 2010-09-21]. 14:32:19 AxelPolleres: Alex and Birte will review SD. 14:32:27 AxelPolleres: Protocol document 14:32:33 ... still have to figure out the status. 14:32:39 Uothing new, i think we shouldn't poublish it this time around 14:32:54 ... will leave it out for this round. 14:32:58 protocol to be left out for this round. 14:33:09 AxelPolleres: Test Suite 14:33:27 ... question about how we publish it. 14:33:43 ... does it make sense to put the test suite in rec-track? 14:33:52 ... or sufficient to have it as a note? 14:34:07 ... will check our previous discussions. 14:34:19 AndyS: I don't see how we can have feedback on implementations unless test suite is rec-track. 14:35:10 AndyS: is this about the test case document, or the tests themselves? 14:35:20 AxelPolleres: we have to check whether those we've dicussed are marked as approved. 14:35:34 ... I would publish the whole document structure, but need a separate overview document. 14:35:41 ... only have the readme document right now. 14:35:57 ... doesn't make sense to publish this alone. maybe we should hold back. 14:36:14 ... on the other hand, we're losing time if we don't do it now. 14:36:32 ... I won't be able to work on it in the next 2 weeks. 14:36:44 ... I can promise to do it in 3 weeks, then we revisit. 14:37:12 ACTION: Axel to bring test suite in shape for FPWD within 3 weeks 14:37:12 Created ACTION-320 - Bring test suite in shape for FPWD within 3 weeks [on Axel Polleres - due 2010-09-21]. 14:37:26 AxelPolleres: Overview document 14:37:36 ... can probably put it out as FPWD 14:37:42 ... any opinions? 14:37:56 link in the agenda doesn't work 14:38:01 http://www.w3.org/2009/sparql/wiki/Overview-Document 14:38:32 http://www.w3.org/2009/sparql/wiki/Overview-Document 14:38:48 maybe just for me 14:38:54 q+ 14:39:23 ack kasei 14:39:28 I don't think the overview document is ready to be published. 14:39:53 kasei: more comfortable if FPWD on overview if there was content in the major sections 14:40:04 ivan: not intended for a rec document, right? 14:40:10 ... timing is different than for other docs. 14:40:20 ... no urgency to publish the overview right now. 14:40:56 Axel: Overview is not REC track, it's a NOTE 14:41:01 no urgency for Overview document at the moment, if we publish as note, no hurry 14:41:06 pgearon_ has joined #sparql 14:41:29 AxelPolleres: that's it for the agenda. hoping next week we can discuss LET/BIND. 14:41:42 q? 14:41:43 ... any other issues? 14:41:51 AndyS: are we actually having a meeting next week? 14:42:02 AxelPolleres: who is not around next week? 14:42:03 I am at risk 14:42:05 regrets for next week? 14:42:32 I am around. Seems to be only me. Decisions, decisions. 14:42:53 I am 14:43:01 AxelPolleres: hard to have a chair next week. ok to skip next week? 14:43:19 ivan?, axel, lee, souri 14:43:25 ivan: propose you wait for sandro to come back and ask him if he can chair. 14:43:38 Steve is still on holiday next week, right? 14:43:43 could be a very productive meeting :) 14:44:06 AndyS: propose chairs send out email for next week so progress can be made even without a telecon. 14:44:35 pgearon_ has left #sparql 14:45:21 zakim, drop me 14:45:21 Ivan is being disconnected 14:45:22 -Ivan 14:45:23 -LeeF 14:45:25 -??P33 14:45:26 - +1.603.897.aadd 14:45:27 adjourned 14:45:27 -bglimm 14:45:28 -MattPerry 14:45:28 adjourned 14:45:32 -OlivierCorby 14:45:38 -AndyS 14:46:04 -kasei 14:46:07 axel: please try to get the promised reivews in in time, possibly earlier, such that we can decide to publish WDs in 2, latest 3 weeks. 14:46:15 link to the minutes publishing stuff? 14:46:28 rrsagent, make records public 14:46:39 kasei, will you do the commonscribe thing? 14:46:46 to publish the minutes? 14:46:51 yes, but need a link to it 14:47:14 AndyS: still around? 14:47:16 AndyS, are you still there? I meant to ask how you run tests with ARQ and whether I can do that too, just using my modified BGP evaluation. That would spare me the time of writing code for parsing tests and producing the right output format 14:47:26 kasei, yes 14:47:30 http://www.w3.org/2010/09/14-sparql-irc 14:47:34 thanks! 14:47:41 wonder if you'd seen my question from yesterday about type promotion? 14:47:51 -AxelPolleres 14:47:52 SW_(SPARQL)10:00AM has ended 14:47:54 Attendees were AxelPolleres, kasei, +33.4.92.38.aaaa, OlivierCorby, Ivan, AndyS, +1.603.897.aabb, MattPerry, +1.617.245.aacc, +1.603.897.aadd, [IPcaller], LeeF, bglimm 14:48:19 AxelPolleres: where's the commonscribe publishing stuff located? 14:48:52 bglimm, should be possible to reuse all the test code but IIRC adding in a different BGP engine is not possible externally. 14:49:39 for modifed codebase, put in your BGP code and run from command line with "qtest manfest.ttl" - there is a helper script qtest 14:50:03 kasei, what was the question? Slipped my mind or eyesight. 14:50:25 AndyS: right above the defn of Sum in 10.2.2, there's an example about summing 1, 2.0e0, and 3.0. the doc says it should result in 6.0 (decimal), but I expected 6e0 (float). 14:50:31 Ok, I'll see what I have to do to make it work with my code, thanks 14:50:36 since float can't be 'promoted' to decimal, only to double. 14:50:50 Looking .... 14:52:39 http://www.w3.org/2009/CommonScribe/panel/ 14:53:23 Either float or double - can't remember which OTTOMH 14:53:41 but definitely not decimal, right? 14:54:04 not decimal 14:54:55 ok. 14:55:20 Seems to be float (http://www.w3.org/TR/xpath20/#promotion) 14:55:27 I suppose for the non-GROUPBY agg tests I added, I could modify them so that the numeric testing happens inside an ASK, just like the GROUP_CONCAT tests... 14:56:41 Devious - I was going to (re)write result set testing and include value testing so I don't mind either way. 14:57:13 Guess it's a Q of what's easier for the non-WG impls. 14:57:40 well, for the GROUP BY tests, I think keeping the numeric values in the srx makes sense, otherwise the tests end up having enormous FILTERs. 15:00:13 bglimm, the test suite executor will execute whatever the rest of the code does (IIRC it has to switch off some ARQ-ism to be strict). Installing a QueryExecutionFactory should just (!) work. The code of course uses Jena to read the manifest so slightly incestuous but it avoids using SPARQL. 15:02:25 I still get: No such tests jar, and try to figure out what the script actually want, but that's test.sh because I can't find qtest... 15:03:01 Ah, just found qtest 15:07:29 Bacjk in a moment IRC client has "lost" the cusor and cursor keys. 15:07:39 AndyS has joined #sparql 15:08:00 Better. Can see where I'm typing now. 15:08:43 bglimm, if you're working from Eclipse, qtest will add on the classes/ directory so you get the development version. 15:09:24 Yes, I just noticed that. I think, I'll get it to work and if not, I'll get back to you. 15:11:38 Great - drop me email. Are you using the QueryExecutionFactory? 15:11:43 Yes 15:12:43 Great. Should make reading the manifest and executing tests separate. 15:14:55 Nice, now it runs at least in your ARQ (not my BGP matching), I had it set to Java 1.5, but it requires 1.6 now and didn't compile after I did an SVN update (I had not updated for a long time). 15:15:18 Tests = 10 : Successes = 10 : Errors = 0 : Failures = 0 15:23:24 Yea! 16:21:38 LeeF has joined #sparql 16:57:56 Zakim has left #sparql 18:45:05 pgearon_ has joined #sparql