14:29:09 RRSAgent has joined #dawg 14:29:09 logging to http://www.w3.org/2006/02/21-dawg-irc 14:29:47 zakim, this will be dawg 14:29:47 ok, LeeF; I see SW_DAWG()9:30AM scheduled to start in 1 minute 14:30:19 SW_DAWG()9:30AM has now started 14:30:26 +??P4 14:30:33 zakim, ??P4 is me 14:30:33 +AndyS; got it 14:30:40 +[IBMCambridge] 14:30:43 +Libby_Miller 14:30:54 zakim, IBMCambridge is temporarily me 14:30:54 +LeeF; got it 14:31:23 +Sven_Groppe 14:31:49 LeeF has changed the topic to: "DAWG Weekly 21Feb. scribe: AndyS http://www.w3.org/2001/sw/DataAccess/weekly-agenda" 14:31:50 EliasT has joined #dawg 14:32:00 +Kendall_Clark 14:32:00 oh yeah 14:32:01 :) 14:32:20 +EricP 14:32:20 +Elias_Torres 14:32:27 Zakim, Elias_Torres is me 14:32:27 +EliasT; got it 14:33:21 +DanC 14:33:44 agenda + Convene ... http://www.w3.org/2001/sw/DataAccess/weekly-agenda 14:34:04 agenda + upcoming ftf 2-3Mar 14:34:05 Zakim, mute me por favor 14:34:05 I don't understand 'mute me por favor', EliasT 14:34:11 agenda + Toward updated QL WD 14:34:13 Zakim, please mute me 14:34:13 EliasT should now be muted 14:34:19 agenda + comments related to rdfSemantics, owlDisjunction 14:34:22 zakim, mute me 14:34:22 EricP should now be muted 14:34:26 agenda + Toward CR 14:34:33 agenda + test suite maintenance 14:34:38 Zakim, take up item 1 14:34:38 agendum 1. "Convene ... http://www.w3.org/2001/sw/DataAccess/weekly-agenda" taken up [from DanC] 14:34:59 Scribe: AndyS 14:35:06 Agenda: http://lists.w3.org/Archives/Public/public-rdf-dawg/2006JanMar/0409.html 14:35:32 Regrets: JosD, JeenB, EnricoF 14:35:39 Zakim, who's on the phone? 14:35:39 On the phone I see AndyS, LeeF, Libby_Miller, Sven_Groppe, Kendall_Clark, EliasT (muted), EricP (muted), DanC 14:36:03 Chair: DanC 14:36:39 -> http://lists.w3.org/Archives/Public/public-rdf-dawg/2006JanMar/att-0400/14-dawg-minutes.html minutes 14 Feb 14:36:54 2nd 14:37:11 RESOLVED: to approve minutes 14 Feb 14:37:30 Meeting: DAWG Weekly 14:37:44 agenda + JSON serialization draft 14:37:50 Next meeting: F2F (W3C All Groups Meeting, Cannes, 2/3 March) 14:37:55 Zakim, next item 14:37:55 agendum 2. "upcoming ftf 2-3Mar" taken up [from DanC] 14:37:59 ack ericp 14:38:09 ack ericP 14:38:29 Eric: Agenda in-progress 14:38:49 q+ re observers 14:38:50 Eric: 24 people registeres inc observers (14 observers) 14:38:56 +PatH 14:40:01 EricP: More of a workshop style for part of the meeting 14:40:15 q+ ask if it makes sense to join with swig for some dawgshows 14:40:22 q+ to ask if it makes sense to join with swig for some dawgshows 14:40:25 DanC: noted [test-dodo] items 14:41:01 -> http://esw.w3.org/topic/SwigAtTp2006 SwigAtTp2006 14:41:21 (afternoon of which day?) 14:42:06 hmm, I believe my availability to dial-in has been reduced to something approximating zero. 14:42:11 Next steps discussion: afternoon of Thursday 14:42:53 not sure if i need to send regrets for dialing in? 14:43:02 or maybe Friday (SWBPD WG meeting Friday - "fairly informal") 14:43:02 no more than you have 14:43:12 'k 14:43:18 KC sends regrets for F2F 14:43:54 PatH sends regrets for F2F 14:44:39 Zakim, next item 14:44:39 agendum 3. "Toward updated QL WD" taken up [from DanC] 14:44:49 zakim, mute me 14:44:49 EricP should now be muted 14:44:59 DanC: Please publish message sent 14:45:00 Alabama. :> 14:45:04 zakim mute me 14:45:11 zakim, mute me 14:45:11 Libby_Miller should now be muted 14:45:17 DanC: may tweak status section 14:45:28 ...should appear any time now 14:45:50 EricP did "EricP to fix /0 stuff in rq23" 14:46:07 Commets due 3 weeks yesterday (13 March) 14:47:13 Zakim, next item 14:47:13 agendum 4. "comments related to rdfSemantics, owlDisjunction" taken up [from DanC] 14:47:29 Continue action Enrico 14:47:39 ACTION: Enrico to explain definitions in reply to LeeF's questions 14:47:58 ACTION: patH to respond to PFPS comments (pending LC publication) 14:48:39 patH has joined #dawg 14:49:21 Is a *very* minor editorial comment on the current QL draft in order? 14:49:37 No :-) Oh all right 14:49:54 Zakim, next item 14:49:54 agendum 5. "Toward CR" taken up [from DanC] 14:50:25 ACTION: KC to update protocol references to [SRD] and WSDL 14:50:47 ACTION: KendallC to edit proto-wd w.r.t. decision: in case of syntax error [including extension], must return malformed query or query request refused 14:51:04 agenda + WSD WG comments 14:51:07 Zakim, take up item 8 14:51:07 agendum 8. "WSD WG comments" taken up [from DanC] 14:51:42 -> http://lists.w3.org/Archives/Public/public-rdf-dawg-comments/2006Feb/0009.html RE: Belated comments on SPARQL Protocol for RDF 25 January 2006 LC WD 14:52:49 hugo's WSDL converter does not do 2.0 to 1,1 14:53:32 Aye 14:53:33 DanC: request someone to feed our WSDL to feed woden 14:53:37 Sure 14:53:58 ACTION Leef: try SPARQL WSDL files with Woden validator, report results. 14:54:15 LeeF, be sure to include CVS version info from our WSDL files when you report results. 14:54:18 I'll wait until it's checked in, and once checked in I'll also update the WSDL 1.1 stuff 14:54:24 http://incubator.apache.org/woden/userguide.html 14:55:12 ACTION KC: update SPARQL WSDL query operation to change name to query-request (though it may change back) 14:55:35 Zakim, close item 8 14:55:35 agendum 8, WSD WG comments, closed 14:55:36 I see 3 items remaining on the agenda; the next one is 14:55:37 5. Toward CR [from DanC] 14:56:12 # ACTION: DanC to respond to major technical: underspecified errors [CONTINUES] 14:56:18 ACTION: DanC to respond to major technical: underspecified errors [CONTINUES] 14:56:19 ACTION: DanC to respond to major technical: underspecified errors 14:56:33 ACTION: -8 14:56:37 -> http://www.w3.org/2001/sw/DataAccess/crq349 CR request, in progress 14:57:48 RRSAgent, drop action 9 14:57:54 RRSAgent, drop action 8 14:58:03 http://www.w3.org/2001/sw/DataAccess/crq349 search for @@hmm 14:59:26 I can also imagine it going to a LC and attracting a lot of comments (of the form Kendall just mentioned) :-) 15:00:00 Can accept and postpone straight away - input to SPARQL-2 15:03:16 SPARQL Protocol And RDF Query Language 15:03:57 DanC: currently minded not to take requirments to rec. 15:05:27 ACTION DanC: noodle on SOAP exit criteria 15:06:14 DanC: I do not expect to chair through the whole of the CR phase 15:08:37 ack ericp 15:10:03 REQUEST FOR TESTCASE: something that uses SPARQL media type in POST 15:10:46 never mind. it uses x-www-urlencoded 15:11:09 zakim, mute me 15:11:09 EricP should now be muted 15:13:16 ack ericp 15:13:33 zakim, mute me 15:13:33 EricP should now be muted 15:13:34 i've heard that from people at Nokia and at my own lab, actually. :> 15:14:30 s/satsifaction of the community./satisfaction of a critical mass of the community/ 15:15:42 zakim, mute me 15:15:42 Kendall_Clark should now be muted 15:16:38 erp, i'm actually convinced by those objections now. :< 15:17:52 kendallclark, yeah, you noted an interest in 4-ary querying in your SPARQLNextSteps response 15:18:11 yep, that's exactly what I'm convinced by now. 15:18:46 FROM/FROM NAMED seem a far too coarse way of doing that when we have a formalism for doing it more finely grained that does *not* involve web-arch issues. 15:18:51 (review of CR draft) 15:19:06 i note that this came *from* timbl's comment 15:20:24 Hayes: "defns are unnecessarily complicated" 15:20:25 zakim, unmute me 15:20:25 Kendall_Clark should no longer be muted 15:22:59 Zakim, next item 15:22:59 agendum 5. "Toward CR" taken up [from DanC] 15:23:04 Zakim, close item 5 15:23:04 agendum 5, Toward CR, closed 15:23:05 I see 2 items remaining on the agenda; the next one is 15:23:06 6. test suite maintenance [from DanC] 15:23:30 ack ericP 15:23:38 ACTION: JosD to make test case from [19]Sergio's basic query patterns examples. 15:23:42 http://www.w3.org/mid/20060118004002.GC12293@w3.org 15:23:48 ACTION: EricP to fix test schema to match manifest with negative tests [recorded in [21]09/27-dawg-minutes.html#action16] 15:23:57 ACTION: DanC to follow up re optional test based on op:dateTime triple 15:24:05 ACTION: DaveB add to test suite the temperature case from comment on truth tables in [22]commentor's message 15:24:12 Zakim, take up item 2 15:24:12 agendum 2. "upcoming ftf 2-3Mar" taken up [from DanC] 15:24:28 Friday PM : test suite maintenance 15:25:07 I'd like to join that, hopefully we can have one schema for both language and protocol. 15:25:22 Zakim, unmute me 15:25:22 EliasT should no longer be muted 15:25:31 I will be there. 15:26:01 Libby is on the queue 15:26:02 No? 15:26:08 ack libby 15:26:09 libby, you wanted to ask if it makes sense to join with swig for some dawgshows 15:26:12 Oh 15:26:13 :) 15:26:15 Nevermind 15:26:18 sorry I'm muted! 15:26:21 q- 15:26:27 ADJOURN. 15:26:45 RRSAgent, please draft minutes 15:26:45 I have made the request to generate http://www.w3.org/2006/02/21-dawg-minutes.html DanC 15:26:51 did we talk about JSON? 15:26:52 RRSAgent, make logs world-access 15:26:54 patH has left #dawg 15:26:58 agenda? 15:27:15 Zakim, take up item 7 15:27:15 agendum 7. "JSON serialization draft" taken up [from DanC] 15:27:23 http://www.mindswap.org/~kendall/sparql-results-json/ 15:27:28 AndyS, http://www.w3.org/2006/02/21-dawg-minutes is now public 15:27:57 Got it. 15:28:12 http://sparql.org/sparql?query=PREFIX+books%3A+++%3Chttp%3A%2F%2Fexample.org%2Fbook%2F%3E%0D%0APREFIX+dc%3A++++++%3Chttp%3A%2F%2Fpurl.org%2Fdc%2Felements%2F1.1%2F%3E%0D%0ASELECT+%3Fbook+%3Ftitle+WHERE+%7B+%3Fbook+dc%3Atitle+%3Ftitle+%7D%0D%0A&default-graph-uri=http://www.w3.org/2001/sw/DataAccess/proto-tests/data/select/simple-data.rdf&accept=application/sparql-results%2Bjson 15:28:32 Please add ARQ/Joseki to the impls. 15:29:57 HP is omniscient 15:34:19 i like both ideas, elias, precisely for the 'small, puny client' design space 15:35:12 there's an obvious bridge there though, i.e. dereferencing the name using a well-understood protocol handler 15:35:15 example: FROM 15:35:21 bingo 15:35:43 Maybe able to add that. 15:35:48 http://en.wikipedia.org/wiki/Data:_URL 15:36:12 [[ 15:36:14 The FROM NAMED syntax suggests that the IRI identifies the corresponding graph, but actually the relationship between a URI and a graph in an RDF dataset is indirect. the IRI identifies a resource, and the resource is represented by a graph (or, more precisely: by a document that serializes a graph). 15:36:18 patH has joined #dawg 15:36:19 ]] 15:36:35 safe URL characters :) 15:36:47 date.replace(" ","") 15:36:59 escape(data) 15:37:01 .replace("%20", " ") 15:37:12 encodeURIComponent(data) 15:37:26 -Kendall_Clark 15:37:57 AndyS, is data: support coming on ARQ/Jena? 15:38:10 soon :-) 15:38:17 Not yet. 15:38:34 I might have to code a quick LocatorData :-) 15:38:37 with arbitrarily long URIs, right, Elias? :p 15:38:42 The cost is ... related to "beer" (or wine, as we are invFrance) 15:39:00 or the pay? cough. cough. 15:39:20 AndyS, we need tokens to bypass the 50K limit. 15:39:21 "Small beer" - I'm paid in the UK :-) 15:39:21 { sparql:names ?G } <=> { webarch:identifies [ webarch:representation ?G ] } 15:40:17 I'll try to add it ASAP. Not sure Locator will hack it - yes it will - returns a stream to its own bytes. What a hack! 15:40:36 -Libby_Miller 15:40:46 To add data: support to Jena 15:41:08 Why over 50K? (not in a data URI please!) 15:42:24 DanC, on the right side of the <=> i see three nodes. which of them goes into SELECT * FROM NAMED ??? ? 15:43:06

and ?G 15:43:26 thanks, AndyS 15:43:27 either of them works? that's odd 15:43:32 -DanC 15:43:35 huh? no; each of them has a position 15:43:36 -EricP 15:43:40 -EliasT 15:43:42 -PatH 15:43:44 -Sven_Groppe 15:43:48 SvenGroppe has left #dawg 15:44:01 SELECT * FROM NAMED # results in a dataset with (, G1) in its dataset 15:44:10 AndyS has left #dawg 15:44:14 -AndyS 15:44:38 where G1 is some graph derived from a representation [ is webarch:identifies of "http://...erics-foaf" ] 15:44:50 -LeeF 15:44:51 SW_DAWG()9:30AM has ended 15:44:52 (the <=> rule above has a use/mention bug) 15:44:53 Attendees were AndyS, Libby_Miller, LeeF, Sven_Groppe, Kendall_Clark, EricP, EliasT, DanC, PatH 15:45:44 and which of , G1 will be the graph identifier in the GRAPH ??? graph patterns? 15:46:00 it sounds like G1 15:46:25 but that leaves an odd conundrum when writing the queries: 15:47:01 SELECT * FROM NAMED WHERE { GRAPH G1 { ?s ?p ?o } } 15:47:07 "http://...erics-foaf" will be the identifier in GRAPH ?? patterns. 15:48:42 i think i meant G1, per 15:48:44 2006-02-21T15:44:01Z SELECT * FROM NAMED # results in a dataset with (, G1) in its dataset 15:49:26 rather: with ("http://...erics-foaf", G1) in its dataset 15:51:49 let { ?URI sparql:names ?GRAPH } mean: SELECT * FROM NAMED ?URI results in a dataset including (?URI, ?GRAPH). Then { ?URI sparql:names ?GRAPH } <=> { [ log:uri ?URI] log:semantics ?GRAPH }. 15:52:25 where log:uri is, roughly, owl:inverseOf webarch:identifies 15:52:43 and log:semantics is, roughly, webarch:representation (+ RDF parsing) 15:53:29 does that make sense? (starting with let { ...) 16:04:26 DanC, need to run off. read back and continue in 1.5 hr 16:04:58 ok. hasta. 16:05:22 now i need to find the chairs of SWIG and SWBP so i can send out a proposed agenda ebfore i leave 16:05:34 2 chairs for SWIG, 1 for SWBP, iirc 16:06:37 only danbri listed for SWIG. weak 17:08:49 EliasT has joined #dawg 17:12:28 ugh. bunch of broken links in rq23 17:49:59 Serious breaks? 17:55:37 serious enough to stop publication. You're copied on the webmaster's reply of 21 Feb 2006 11:55:00 -0500 17:55:55 http://www.w3.org/TR/2006/WD-rdf-sparql-query-20060220/defn_SPARQLQueryString 17:55:55 What to do: The link is broken. Fix it NOW! 17:55:55 Response status code: 404 17:56:06 I wonder if I introduced that one 17:56:16 http://www.w3.org/TR/2006/WD-rdf-sparql-query-20060220/parsers/ 17:56:16 What to do: The link is forbidden! This needs fixing. Usual suspects: a missing index.html or Overview.html, or a missing ACL. 17:56:16 ugh. 17:56:29 I'm pretty sure I didn't do this one: 17:56:29 http://www.w3.org/TR/2006/WD-rdf-sparql-query-20060220/rHEX 17:56:29 What to do: The link is broken. Fix it NOW! 17:56:49 I'll look at it now. 17:57:06 if you fix in rq23, I can apply patches to /TR/ 17:57:07 Shoudl I change rq23? 17:57:19 yes 17:57:37 And it has to be propagted to the publishable version by you? 17:58:27 It's just a missing # - one occurence 18:03:24 has to be propagated by somebody; I doubt you have CVS write access, though you're welcome to try. You could ask Matthew to apply some patches. I can probably get to it later today, though 18:04:03 Zakim has left #dawg 18:05:18 AndyS, where'd you get to on rq23 b0rken links? 18:05:49 One moment - running the validator 18:06:25 standing by to apply patches 18:08:24 There is an href="rHEX" which should be href="#rHEX" (one times) 18:08:40 Quite a few more - give me some time. 18:09:24 roger 18:09:41 href="#defn_SPARQLQueryString" - # missing (x1) 18:09:53 standing by in a doing-day-job-at-the-same-time way 18:11:21 WD-rdf-sparql-query-20060220/parsers/ needs an index.html - no - gets 403 so I presume its OK really 18:12:15 In CVS log: http://www.w3.org/mid/20051130065250.GC30250@w3.org - a message from you about xsd:string/simple literal 18:13:28 huh. will look for it 18:14:34 Don't understand "There are broken fragments which must be fixed." then long, long list of lines 18:15:48 below there should be a list of the fragments in question 18:18:38 They don't align. 18:19:42 Can you fix rHEX, defn_ and your email then rerun the validator? I'll do an HTML validation in case the link checker is confused by errors. 18:20:29 on it... 18:20:44 xsd:string link should be http://www.w3.org/mid/20051130065250.GC30250@w3.org 18:20:57 sorry, http://www.w3.org/mid/20051130195040.GB31232@w3.org 18:21:25 I'll change rq23. 18:22:32 If rq23 == the WD draft? 18:22:40 Is rq23 byte== the WD draft? 18:23:20 ahh, doubt it 18:24:21 i tend to edit in place (ensuring that WD 1.1 byte== rq23 1.???) but danc likes to keep rq23 from lying about what it is 18:24:39 It's XHTML strict 18:24:46 i will do a CVS diff of all the fixes and apply it to the WD 18:25:12 Linkchecking rq23 ... 18:25:31 since I'm doing this publication, indeed, rq23 will not lie about itself. 18:26:00 3 patches so far: 18:26:14 1/ href="rHEX" => href="#rHEX" 18:26:31 2/ href="defn_SPARQLQueryString" => href="#defn_SPARQLQueryString" 18:26:56 3/ CVS log at xsd:string/ ==> http://www.w3.org/mid/20051130195040.GB31232@w3.org 18:27:29 . 18:27:42 roger. will try to apply the cvs diffs 18:27:46 "Broken fragments and their line numbers" - the refs seem to be right - is this a linkcheck effect? 18:28:33 ew... line number mismatches. what a pain 18:28:41 Seem to refer to valid # refs e.g. http://www.w3.org/1999/02/22-rdf-syntax-ns#nil 18:28:54 most of the link problems should show up if you run the linkchecker on rq23 itself 18:30:31 ACL problem : fixed : http://www.w3.org/2001/sw/DataAccess/rq23/sparql-defns.html 18:30:41 i recall something funny about #nil from the last pub 18:30:59 s/recall something/recall that there was something/ 18:31:42 checking : defn_BGPequiv ... 18:31:58 Need to merge the two as : 18:32:05

Definition: Basic Graph Pattern equivalence

18:35:06 Rechecking ... 18:40:31 No red in summary. Just "Broken fragments and their line numbers: They need to be fixed!" which I don't understand as it looks right (and does not appear in the table) 18:44:10 OK - assuming I'm done now. 18:44:25 Will be online later on 18:56:58 updated. off to buy beer. work on rq23 upon successful completion of that mission 19:22:48 AndyS has joined #dawg 19:24:20 Back - how are those #refs from the link checker? 19:34:20 AndyS, I see ericP claimed victory. Hmm... I'm 4 min late for my next telcon. Maybe I'll pass on that one. 19:34:55 claimed victory? 19:36:48 updated. off to buy beer. work on rq23 upon successful completion of that mission 19:36:58 oh... I read "successful completion" too quickly 19:37:06 yes, i have beer 19:37:15 I think I'll join you... brb... 19:37:19 and i can prove it 19:38:17 now *there*s a turing test... prove you have beer, over IRC 19:38:39 now jump up and down. what's your stomach feel like? 19:39:02 how intelligent does a person have to be to pass a turing test? 19:39:30 I bet my beer's cheaper than yours. 19:39:48 Milwaukee's best. aka the beast. $5.50/12-pack 19:40:25 but mine has more trappist monk elbow greese in it 19:41:28 any reason not to add a trailing slash to http://www.w3.org/TR/2001/NOTE-vcard-rdf-20010222 ? 19:41:45 i can't think of any, but wonder if someone had a good reason for leaving it off 19:41:45 because the title page of that note gives a uri with no trailing / 19:41:53 aha. see? 19:42:17 it doesn't matter a whole lot. but I prefer that if a doc says "this is my URI" that we take it at its word 19:44:36 i think the linkchecker gives an error on http://www.w3.org/1999/02/22-rdf-syntax-ns#nil because it doesn't know how to resolve application/rdf+xml fragments 19:45:09 yeah... why do we link to that? 19:45:29 [[ 19:45:30 The form () is an alternative for the IRI rdf:nil 19:45:31 which is http://www.w3.org/1999/02/22-rdf-syntax-ns#nil. 19:45:34 ]] 19:46:08 seems like just mentioning that URI would have been enough, but OK 19:46:17 it's a rendered (in your face) URL in the text 19:46:39 yes, the in-your-face-ness is necessary. the link... is ok, I guess 19:48:26 $ patch Overview.html <~/w3ccvs/WWW/2001/sw/DataAccess/rq23/,link-fixes 19:48:31 Hunk #6 FAILED at 6270. 19:49:12 ah... that's the CVS log 19:49:56 i just commited 1.651 . is that in your patch set? 19:50:07 hmm... if we were using one of the modern version control systems (hg in particular), this would be easier. 19:50:44 no, I'm only up to 1.650 so far... 19:51:49 oops, 651 includes http://www.w3.org/TR/2001/NOTE-vcard-rdf-20010222/ , which i had written down before you noted that they elide the trailing link in the spec 19:51:59 worth another pass? 19:52:12 saw that trailing / ; applied it. 19:52:39 other than that, it's set to go (once we convince that http://www.w3.org/1999/02/22-rdf-syntax-ns#nil is a valid uriref) 19:53:02 I'll note #nil in mail to webreq 19:53:14 can you stand by while I note issues#rdfSemantics in SOTD? 19:54:06 sure 19:54:16 i noted #nil in the last go-round, btw 19:56:05 so... if somebody stopped you in the hall and asked you why we're doing this WD, what would you say? 19:58:30 http://www.w3.org/TR/2006/WD-rdf-sparql-query-20060220/ v v 1.8 2006/02/21 19:57:49 19:58:33 pls review SOTD 19:58:51 oh... say... what did you guys do about an index.html for /parsers/ ? 19:58:55 anything? 19:59:24 hmm... no changes in rq23/parsers/ 19:59:44 It looks good to me, DanC 19:59:58 v1.8 that is, Leef? 20:00:01 LeeF 20:00:24 double checking 20:00:33 yes 20:01:23 i see no index.html 20:01:41 $ (echo

why am I getting: bash: syntax error near unexpected token `;' 20:02:25 aha... 20:03:01 quote the first

also, you're missing a

:-) 20:03:35 I'm gonna pipe it into tidy 20:03:48 $ (echo "