14:30:23 RRSAgent has joined #dawg 14:30:24 +??P5 14:30:31 zakim, ??P5 is AndyS 14:30:31 +AndyS; got it 14:30:42 scribe: EricP 14:30:51 Zakim, take up agendum 1 14:30:51 agendum 1. "Convene, take roll, review records and agenda http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0412.html" taken up [from DanC] 14:31:04 There is the matter of untrusted graphs (re Source) of course. 14:31:19 Alberto/regrets 14:31:32 HowardK/regrets 14:31:40 regrets Pat Hayes 14:31:48 Hiroyuki Sato, regrets 14:32:06 zakim, who is here? 14:32:06 On the phone I see EricP, DanC, SimonR, TomAdams, SteveH, Kendall_Clark, Kevin, AndyS 14:32:31 regrets Dave Beckett 14:33:03 Farrukh/regrets 14:33:12 30 Nov http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0394.html 14:33:37 resolved to accept http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0394.html as a true record 14:33:59 zakim, please take up agendum 1 14:33:59 agendum 1. "Convene, take roll, review records and agenda http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0412.html" taken up [from DanC] 14:34:09 Zakim, pick a scribe 14:34:09 Not knowing who is chairing or who scribed recently, I propose AndyS 14:34:26 next meeting Dec 14 14:34:34 chair: danc 14:34:40 scribe: tomA 14:34:49 ACTION KendallC: expose our walking tour data to SPARQL querying clients 14:34:49 ETA before F2F4 14:35:02 agenda + ftf5 in March 2005 14:35:43 zakim, agenda? 14:35:43 I see 7 items remaining on the agenda: 14:35:44 1. Convene, take roll, review records and agenda http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0412.html [from DanC] 14:35:47 2. ftf4 2005-01-19/2005-01-20 in Helsinki, Finland [from DanC] 14:35:48 3. Test Cases [from DanC] 14:35:49 4. Protocol [from DanC] 14:35:50 5. SPARQL SOURCE Issue [from DanC] 14:35:51 6. SPARQL update, issues [from DanC] 14:35:52 7. ftf5 in March 2005 [from DanC] 14:35:55 Zakim next agendum 14:35:58 Zakim, next agendum 14:35:58 agendum 2. "ftf4 2005-01-19/2005-01-20 in Helsinki, Finland" taken up [from DanC] 14:37:13 [DanC forwards mail to WG re what to wear in Finland] 14:37:27 Janne/regrets 14:39:10 JosD has joined #dawg 14:39:21 DanC: expect 12 at f2f4 14:39:27 Zakim, take up agendum 7 14:39:27 agendum 7. "ftf5 in March 2005" taken up [from DanC] 14:39:39 http://esw.w3.org/topic/SwigAtTp2005 14:40:19 +JosD 14:41:00 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. CONTINUES 14:42:41 kendall: is it possible to meet 4 days? I feel I get a lot done at ftf meetings 14:42:49 KendallK proposes 4 day meeting 14:42:50 DanC: I'd be too exhaused to chair all that. 14:43:02 (kendall said "I'm not sure I'm proposing that" actually, ericp) 14:43:12 (right) 14:43:30 (favorable to the idea of the 4 day meeting) 14:43:35 EricP: will dillute the meeting some given ohter people's meeting oblications 14:43:58 SimonR: +1 (will be able to attend) 14:44:35 DanC: can chair two days. want to go to interest group meeting and play with toys 14:45:02 AndyS: can share pain with SWPBD for IG attendance 14:45:20 ... we should present or at least be available to the IG to get feedback 14:45:32 DanC: want to show clients interoperating 14:45:40 http://esw.w3.org/topic/DawgShows <- http://esw.w3.org/topic/SwigAtTp2005 14:45:41 ... cf DawgShow 14:45:46 http://esw.w3.org/topic/DawgShows 14:45:46 Zakim, next agendum 14:45:46 agendum 2. "ftf4 2005-01-19/2005-01-20 in Helsinki, Finland" taken up [from DanC] 14:45:52 Zakim, close agendum 2 14:45:52 agendum 2 closed 14:45:53 I see 4 items remaining on the agenda; the next one is 14:45:54 3. Test Cases [from DanC] 14:45:55 Zakim, next agendum 14:45:55 agendum 2 was just opened, DanC 14:46:17 DONE: ACTION Jos: Write email about alternative result format using 14:46:17 collections - avoid use/mention issues 14:46:21 Zakim, next agendum 14:46:21 agendum 2 was just opened, ericP 14:46:32 jos's msg http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0415.html 14:46:41 zakim, take up agendum 3 14:46:41 agendum 3. "Test Cases" taken up [from DanC] 14:47:01 Jos: confused about use/mention issue. the answer is just in the graph itself 14:47:30 DanC: use/mention: Mary is a girl vs. Mary is a four letter word starting with "M" 14:47:53 zakim, mute me 14:47:53 Kendall_Clark should now be muted 14:47:56 ... no discriminating use caess found yet. 14:48:01 s/use/test/ 14:48:06 tx 14:48:13 DanBri's URI munging test cases seem relevant here 14:48:17 zakim, unmute me 14:48:17 Kendall_Clark should no longer be muted 14:48:23 ah... good point, kendall 14:48:51 EricP, the <> in http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0420.html 14:49:00 DanC: another use/mention issue is regexps on URIs 14:49:13 ... not adding to issues yet... 14:49:25 The property URI and the resources URI cases are slightly different 14:49:43 ISSUE: use/mention issues in regex operators, test result format 14:50:22 DanC: gets to name the issue 14:50:29 s/:// 14:50:57 . ACTION: SteveH is willing to adapt his testing infrastructure to generate input/output RDF/XML and typed nodes into manifest file 14:50:58 DnaC: no sweeping changes on the test materials based on Jos's action 14:51:11 ACTION: SteveH is willing to adapt his testing infrastructure to generate input/output RDF/XML and typed nodes into manifest file. CONTINUES. 14:52:30 DanC: README file suggested a link to the minutes which would be useful to me 14:52:43 "dawgt:approval dawgt:Approved" -- tests/data/simple/manifest.n3 14:52:55 SimonR: some parts of the test area are forbidden 14:53:00 ACTION EricP: make sure all dawg tests are world-readable now and going forward 14:53:16 Minor request - no tabs in files please (I have them set as 4 for N3 in emacs) 14:53:21 (note to self, default ACL) 14:54:51 DanC: no specification for the "non-XML" syntax 14:55:19 ACTION EricP: come up with a specification for the "non-XML" syntax (may be turtle) 14:55:29 ? 14:55:32 SteveH: I'm using a turtle parser 14:55:50 DanC: should be integrated with the README 14:56:07 SimonR: next four test cases make sense to me 14:56:14 ... willing to approve them. 14:56:28 these are http://www.w3.org/2001/sw/DataAccess/tests/#sparql-query-example-2-1a 14:56:29 ... need to hack kowari to run them 14:56:36 JosD: i've run them 14:56:39 http://www.w3.org/2001/sw/DataAccess/tests/#sparql-query-example-2-2a 14:56:49 http://www.w3.org/2001/sw/DataAccess/tests/#sparql-query-example-2-3a 14:56:56 http://www.w3.org/2001/sw/DataAccess/tests/#sparql-query-example-2-4a 14:57:42 i see nothing under http://www.w3.org/2001/sw/DataAccess/tests/...,access... that isn't world-readbale 14:58:26 RESOLVED to approve the four tests (above) 14:58:38 Abstentions: KendallC 14:58:45 Objections: -none- 14:58:51 (I've run them, approve them) 14:59:01 [discussion of fifth] 14:59:07 zakim, mute me 14:59:07 Kendall_Clark should now be muted 14:59:16 Is Turtle (N3-data, whatever) going to "standardized" in W3C (c.f. N-Triples). Should we do it? 14:59:16 DanC: has a "<" in it. not sure support it. 14:59:32 considering http://www.w3.org/2001/sw/DataAccess/tests/#sparql-query-example-3 14:59:46 zakim, unmute me 14:59:46 Kendall_Clark should no longer be muted 15:00:36 PROPOSED to approve http://www.w3.org/2001/sw/DataAccess/tests/#sparql-query-example-3 15:00:57 SimonR: can you put scalar constraints in the graph? 15:01:40 "?price + 3 < ?x + ?y" did I hear that right, AndyS? interesting test case. 15:01:47 "Can you put scalar constraints into the same syntax as graph constraints?" 15:03:42 op:numeric-less-than 15:03:54 op: = http://www.w3.org/2001/sw/DataAccess/operations 15:03:56 zakim, mute me 15:03:58 Kendall_Clark should now be muted 15:04:01 EricP: specifying that they are supported in the graph constraint... 15:04:35 $Revision: 1.47 $ of $Date: 2004/12/07 16:31:41 $ has broken link http://www.w3.org/2001/sw/DataAccess/rq23/#extendedtests 15:04:37 [discussing standard XML-Schema comparitors in the graph constraint] 15:05:46 do we support this ?: 15:05:47 :book1 dc:title "SPARQL Tutorial" . 15:05:47 :book1 ns:price :price . 15:05:48 :price foo:lessThan 42. 15:06:09 can we query that differently than testing 15:06:18 :book1 dc:title "SPARQL Tutorial" . 15:06:21 :book1 ns:price 41 . 15:06:23 ? 15:06:44 s/foo/op/ over the last 8 lines 15:07:00 Zakim, who's on the phone? 15:07:00 On the phone I see EricP, DanC, SimonR, TomAdams, SteveH, Kendall_Clark (muted), Kevin, AndyS, JosD 15:07:04 pat? 15:07:53 What about op:lessThan "42"^^xsd:byte (0verloading) 15:09:04 ACTION: SimonR to sketch tests around comparitors in the grpah constraint 15:09:08 ACTION: EricP to sketch tests around comparitors in the grpah constraint 15:09:17 ACTION SimonR: propose test case for scalar constraint with syntax uniform to graph constraints 15:09:27 action -6 15:09:43 Steve - does that work for 3Store? Do you store values or lexical forms? 15:10:04 AndyS, lexical forms, but we store multiple ones, and do live cating to values 15:10:16 AndyS, future versions will (probably) store values too 15:10:27 cating == casting 15:10:55 +1 to simon's point 15:10:59 Our users expect round-tripping : xsd:byte in, xsd:byte out - need lexcial forms 15:11:51 ">" should be bound to a property and be treated as syntactic sugar for it. 15:12:46 ericP: that is distinct from whether those operators are available (and mean the same thing) in the graph constraints 15:12:48 zakim, unmute me 15:12:48 Kendall_Clark should no longer be muted 15:13:11 pointer to draft of that XSCH note, josd? 15:14:19 AndyS: XSCH SWBPD task force is addressing XSD data types in RDF. 15:14:41 ... value space of double and value space of integer are not necessarily the same 15:14:58 DanC, pointer is http://www.w3.org/2001/sw/BestPractices/XSCH/xsch-sw/ 15:14:59 ... can't necessarily compare them 15:15:17 Do we want to put some urgency behind this and aim at a yea/nay vote on ex3 for the next meeting? 15:15:19 Zakim, who's on the phone? 15:15:19 On the phone I see EricP, DanC, SimonR, TomAdams, SteveH, Kendall_Clark, Kevin, AndyS, JosD 15:15:34 http://www.w3.org/2001/sw/DataAccess/tests/#sparql-query-example-3 15:15:45 JosD: +1 to put scalar constraints in the graph constraints 15:16:15 DanC: who's happy wiht test3? 15:16:21 ericP: like the syntax 15:16:27 Danc: would approve 15:16:34 danc would abstain 15:16:36 tx 15:16:48 SimonR does't like syntax 15:16:57 Jose: abstain 15:16:57 I would approve it 15:17:07 SteveH: approve 15:17:17 i don't know how many i missed 15:17:26 i didn't XOFF 15:17:43 DanC: not putting the question now 15:17:53 TomAdams, doesn't like syntax, abstain 15:17:56 Andy's next example: 2*?x < ?y+3 15:18:01 (after 4-ish in favor and multiple abstantions) 15:18:05 Zakim, move to next agendum 15:18:05 agendum 4. "Protocol" taken up [from DanC] 15:18:38 26 Nov binding http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0364.html 15:18:46 Zakim, who's on the phone? 15:18:46 On the phone I see EricP, DanC, SimonR, TomAdams, SteveH, Kendall_Clark, Kevin, AndyS, JosD 15:18:47 DanC: updated veriable binding results format 15:19:35 EricP: i htink we should use an RDF-compitable but regular syntax 15:19:51 DanC: neutral 15:20:13 Jos sketched an RDF format for returning variable bindings pretty recently, as I recall. 15:20:15 SimonR: +1 RDF, approve list of vars at the head. 15:20:24 ... similar to our SOAP impl. 15:20:35 ... in retrospect, i wouldn't hvae done it that way 15:20:52 ... a var X will always have the smae time 15:21:17 ... would like the vars recorded in an attribute or some such 15:21:32 an rdf[/xml] compatible format will be slightly more bulkier, need extra properties between and for example 15:21:40 TomA: +1 Simon. 15:21:52 ... like them not in attributes 15:22:21 DanC: we could wrap this up in two weeks or so. good idea? bad idea? 15:22:30 "wrap this up" = publish as a WD 15:22:41 TomA: +1 wrap up 15:22:54 SteveH: like it -- need to play wiht it more 15:23:03 I'm hunting for advice what to change here, please can people send email. 15:23:22 KendallC: don't know which format people are talking about 15:23:35 (I don't like representing the variable list as elements -- as an attribute variable="x y" or a processing instruction are alternatives.) 15:23:38 ... like result 1 (vars in element names) 15:24:07 To be more specific, I like format 1 more than 2 15:24:13 SimonR: what don't like? What's a technical difference/problem? 15:24:19 ... liked Rob Shearer's point about getting XML out of RDF 15:24:22 steveh: thanks 15:24:29 +1 to what kendall said, needs to be an XML format 15:24:34 KevinW: like result 1 15:24:51 +1 to andy's point re: con-neg 15:24:59 AndyS: I don't think we need it to be RDF-compatibl.e can use an RDF form. 15:25:09 ... there are folks who like to schema everything 15:25:18 ... wasn't just HP people. 15:25:30 JosD: +1 ericP for RDF format. 15:25:35 DaveB: The techincal problem is that the elements used to represent the variables in the variable list behave entirely differently from everywhere else -- all other instances are associated with some RDF node. 15:25:48 ... proposals are still compact 15:26:02 Yep, I corrected that for the scribe, DanC. 15:26:07 tx 15:26:17 SimonR: fair enough 15:26:28 DanC: would like this in the protocol draft 15:26:40 +1 for adding to the protocol draft 15:26:48 SimonR: +1, can't imagine a protocol draft without it 15:26:57 KendallC: think it's more work 15:27:00 -1 for the protocol draft - can use without protocol 15:27:25 scribe is confused about *what's* more work 15:27:37 probably 2 WDs 15:27:46 ericP: I meant more work *for me* as the potential protocol draft editor 15:27:54 roger 15:28:04 DanC: want published soon 15:28:14 DaveB: That I'd be quite favorable to -- the variables declaration is forced to occur first by the way XML attributes work too, which is rather elegant. 15:28:16 daveb, do you want to edit your design into WD format, or would you prefer somebody else do that? 15:28:18 daveb: do you want to edit a draft of the XML result format? 15:28:28 ericP, Danc: yes, I'll edit it 15:28:30 s/:/,/ 15:28:48 daveb, can we action you? 15:28:51 yes 15:28:53 tx 15:28:54 I'd like to get an action on Dave to write and on ?SOMEBODY to review 15:29:06 when can you deliver a draft, dave? when can ?SOMEBODY review it? 15:29:20 ACTION: DaveB to write an XML result format 15:29:31 is this a "few days" sorta thing, daveB? or a 2 weeks thing? 15:29:32 ACTIO: AndyS to review 15:29:36 few days 15:29:55 ACTIO: SteveH to review Dave's XML result set format 15:30:11 ACTION: AndyS to review Dave's XML result set format 15:30:17 ACTION: SteveH to review Dave's XML result set format 15:30:36 so I'm expecting to have a draft and some reviews in my hand when I prepare next week's agenda. let me know if any risks/problems pop up 15:30:45 I shall try to make some sort of xml schema (RNG, weak xsd) for the result1 format 15:30:49 ok 15:31:17 I have no beliefs about the matter. 15:31:58 DanC: result set format thread issues... 15:32:17 # Issue with the result set format Seaborne, Andy (Monday, 22 November) http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0336.html 15:32:24 s/thread issues/issues thread/ 15:32:31 SimonR has joined #dawg 15:32:35 that was about listing the vars at the top 15:32:36 ... listing vars at top 15:32:39 ... will get worked out 15:33:05 http://monkeyfist.com/kendall/sparql-protocol-simplex/ 1.8 15:33:10 kendall, why "simplex"? 15:33:11 zakim, take up next agendum 15:33:11 agendum 5. "SPARQL SOURCE Issue" taken up [from DanC] 15:33:21 zakim, take up previous agendum 15:33:21 I don't understand 'take up previous agendum', ericP 15:33:24 oops 15:33:54 NOTE TO SCRIBE: still on protocol agendum 15:34:08 zakim, take up agendum 4 15:34:08 agendum 4. "Protocol" taken up [from DanC] 15:34:34 KendallC: comments were against the larger document/design. 15:34:48 ... haven't checked them against simplex 15:35:33 AndyS: I put comments against the simplex spec 15:36:12 DanC: if i were KendallC, i wouldn't be in audit mode now. 15:36:37 Kendall: i read everyhting but didn't necessarily resond 15:36:52 AndyS: why is getTriples a separate language 15:37:29 ... HTTP section doesn't allow you to specify an other QL 15:37:57 KendallC: agree other langs should be available. 15:37:58 ... probably a document bug 15:38:35 [getTriples is getGraph] 15:39:03 ... not in audit more while doc isn't a WG draft 15:39:42 http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0326.html talks about lang= 15:40:08 [meeting will go full 90 mins] 15:40:26 DanC: would like to see simplex become a WD in three weeks 15:40:38 +1 simplex to WD 15:40:53 +1 15:40:59 +1 15:41:05 +1 15:41:27 i've implemented some of it 15:41:41 inc XML results 15:43:21 PROPOSE that this simplex be published pending approval from KendallC and TomA 15:43:40 Impl: http://www.sparql.org/query.html :-) 15:43:45 KendallC: want AndyS as he has the most substantive issues 15:43:49 PROPOSE that this simplex be published pending approval from KendallC and TomA and AndyS 15:44:19 So RESOLVED. 15:44:19 Objections: 0 15:44:27 Abstentions: 0 15:44:36 s/ten/tan/ 15:45:05 I'd be particularly pleased to see DaveB's XML result format integrated into the concrete protocol. 15:45:21 KendallC: time frame? three days? three weeks? 15:45:36 I'll be using the xml result format sans protocol, as will others I guess 15:45:52 I will agree to anything to publish - agreement is not necessary for a WD as far as I'm concerned 15:46:14 "SPARQL Protocol for RDF" 15:46:25 DaveB: I agree. I don't really *want* it in the protocol draft; but I'd really like to have a document I can point to from the protocol doc. 15:46:31 ACTION: publish "SPARQL Protocol for RDF" pending approval from TomA and AndyS 15:46:53 daveb, I hope to put a similar question re results format next week 15:47:03 ack DanC 15:47:12 DaveB: how to convert the query results definitions from rq23 into a sequence of XML characters seems to part of protocol to me, even if it's useful elsewhere. 15:47:23 DanC: we need to figure out how to test the protocol 15:47:27 ...seems to be... 15:48:02 Zakim, next agendum 15:48:02 agendum 5. "SPARQL SOURCE Issue" taken up [from DanC] 15:48:23 http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0426.html 15:49:00 ACTION DaveB: update the dawg test repository to record or amend tests to correspond to the http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0241.html 15:49:03 CONTINUED 15:49:11 # TKS/Kowari named graph usage Tom Adams (Tuesday, 7 December) http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0426.html 15:49:15 ACTION TomA: write email to list about how their customers are using named graphs 15:49:16 discharges Tom's action 15:49:16 DONE 15:49:31 ACTION JanneS: propose text for hard failure in the protocol draft 15:49:45 CONTINUED 15:49:47 ACTION DanC: suggest formal definitions for SOURCE 15:50:19 CONTINUED 15:50:25 distracted by OPTIONAL 15:51:07 [ Proposal: querying untrusted graphs ] 15:51:14 -SteveH 15:51:25 DanC: hope this will resove the SOURCE issue 15:51:44 I see it as input to the SOURCE issue; hence I'd like a response from DaveB, SOURCE issue owner 15:51:45 tx 15:51:57 Zakim, next agendum 15:51:57 agendum 6. "SPARQL update, issues" taken up [from DanC] 15:52:16 http://www.w3.org/2001/sw/DataAccess/rq23/ $Revision: 1.47 $ of $Date: 2004/12/07 16:31:41 $ 15:52:43 ACTION: PatH to review editors working draft of rq23 15:52:58 CONTINUED 15:53:13 much progress 15:53:15 progress on pat's action: http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0419.html http://www.ihmc.us/users/phayes/SPARQL-PAT.html 15:53:56 ah... http://www.ihmc.us/users/phayes/SPARQLdefs.html 15:54:05 KendallC: PatH has a set of definitions in SPARQLdefs.html 15:54:39 s/CONTINUED/DONE/ 15:55:01 ACTION AlbertoR: owns issue 'DESCRIBE' 15:56:04 -TomAdams 15:56:34 action DONE. 15:57:09 ACTION DanC: owner of issue 'yes or no questions' (pending protocol doc) 15:57:43 [ DanC auctions his issue ] 15:57:59 EricP: may take it next week 15:58:02 CONTINUED 15:58:09 ACTION SteveH: own (i.e. propose resolution to) disjunction issue 15:58:33 AndyS: disjunction has text with no outstanding issues 15:58:55 ... issue is whether examples clearly illustrate 15:59:12 DanC: test cases? 15:59:38 EricP: I can migrate some Algae UnionDisjunciton test cases 16:00:09 AndyS: integrating PatH's feedback could push us past XMass 16:00:19 ... would like to publish as is. 16:01:02 JosD: would like to incorporate PatH's changes 16:01:54 SimonR, KendallC: +1 to waiting for PatH's changes 16:03:01 PROPOSE to adjourn 16:03:19 SimonR: 2nd 16:03:28 -Kendall_Clark 16:03:29 ADJOURN. 16:03:29 -JosD 16:03:30 -SimonR 16:03:41 KevinW has left #dawg 16:03:44 -DanC 16:04:00 -Kevin 16:04:09 ACTION EricP: migrate some Algae UnionDisjunciton test cases 16:05:56 XSD issues: 16:05:58 comparing dissimilar varlue spaces 16:06:17 which to make required 16:06:51 storage issues for lexical comparisons 16:08:19 separate libs for XSD (comparisons) 16:08:34 - ask HowardK, LiamQ 16:08:49 how do you compare NAN and -0 and that sort of thing 16:09:05 impls are not aiming to do all that work from scratch 16:24:11 -AndyS 16:24:14 -EricP 16:24:15 SW_DAWG()9:30AM has ended 16:24:16 Attendees were EricP, DanC, SimonR, [Tucana], TomAdams, SteveH, Kendall_Clark, Kevin, AndyS, JosD 16:24:48 AndyS has left #dawg 16:30:18 RRSAgent, make logs world-access 16:30:22 RRSAgent, bye 16:30:22 I see 20 open action items: 16:30:22 ACTION: KendallC to expose our walking tour data to SPARQL querying clients [1] 16:30:22 recorded in http://www.w3.org/2004/12/07-dawg-irc#T14-34-49 16:30:22