From SPARQL Working Group
Revision as of 15:10, 27 July 2010 by Apollere2
Please justify/explain all edits to this page, in your "edit summary" text.
14:00:34 <RRSAgent> RRSAgent has joined #sparql 14:00:34 <RRSAgent> logging to http://www.w3.org/2010/07/27-sparql-irc 14:00:36 <Souri> Souri has joined #sparql 14:00:39 <LeeF> rrsagent, make logs world 14:00:41 <AxelPolleres> AxelPolleres has joined #sparql 14:00:52 <Zakim> +Sandro 14:01:05 <Zakim> +Souri 14:01:17 <Zakim> +MattPerry 14:01:28 <Zakim> +??P17 14:01:31 <Zakim> + +1.617.553.aaaa 14:01:36 <LeeF> zakim, aaaa is me 14:01:36 <Zakim> +LeeF; got it 14:01:45 <AxelPolleres> trackbot, start meeting 14:01:48 <trackbot> RRSAgent, make logs world 14:01:48 <NicholasH> zakim, ??P17 is me 14:01:48 <Zakim> +NicholasH; got it 14:01:50 <trackbot> Zakim, this will be 77277 14:01:50 <Zakim> ok, trackbot; I see SW_(SPARQL)10:00AM scheduled to start now 14:01:51 <trackbot> Meeting: SPARQL Working Group Teleconference 14:01:51 <trackbot> Date: 27 July 2010 14:02:04 <sandro> zakim, this is sparql 14:02:04 <Zakim> ok, sandro; that matches SW_(SPARQL)10:00AM 14:02:10 <AxelPolleres> chair: Axel Polleres 14:02:35 <AxelPolleres> scribe: NicholasHumfrey 14:02:44 <AxelPolleres> scribenick: NicholasH 14:02:44 <AxelPolleres> regrets: Chimezie Ogbujie, Gregory Williams 14:02:48 <sandro> zakim, mute AndyS 14:02:56 <pgearon_> I can't tell who I am 14:02:56 <AxelPolleres> agenda: http://www.w3.org/2009/sparql/wiki/Agenda-2010-07-27 14:02:57 <Zakim> AndyS should now be muted 14:03:16 <sandro> AndyS, you were the source of some wind/.breath noise. I muted you for the moment. 14:03:23 <pgearon> pgearon has left #sparql 14:03:25 <AndyS> zakim, unmute me 14:03:35 <sandro> zakim, who is on the call? 14:03:36 <Zakim> AndyS should no longer be muted 14:03:45 <NicholasH> Axel: some disccussion on the mailing list and will be talking about the open ISSUES 14:03:46 <Zakim> On the phone I see AndyS, Sandro, Souri, MattPerry, NicholasH, LeeF, +1.540.412.aabb, AxelPolleres, +3539149aacc 14:03:50 <SteveH> is the uk line still down? 14:04:00 <AxelPolleres> PROPOSED: Approve minutes at http://www.w3.org/2009/sparql/meeting/2010-07-20 14:04:02 <NicholasH> SteveH: working for me 14:04:07 <pgearon_> Zakim, aabb is me 14:04:07 <Zakim> +pgearon_; got it 14:04:15 <AndyS> seconded 14:04:18 <sandro> SteveH, sorry, yeah, I think the UK line turns out to only have two circuits. We're trying to provision more. 14:04:26 <SteveH> sandro, doh :( 14:04:32 <AxelPolleres> RESOLVED: Approve minutes at http://www.w3.org/2009/sparql/meeting/2010-07-20 14:04:33 <AndyS> two !!??!! 14:04:40 <ivan> zakim, dial ivan-voip 14:04:40 <Zakim> ok, ivan; the call is being made 14:04:41 <Zakim> +Ivan 14:04:56 <AlexPassant> Zakim: +353149aacc is me 14:05:03 <AlexPassant> Zakim, +353149aacc is me 14:05:06 <NicholasH> Axel: still some difficulties on the UK phone line 14:05:09 <Zakim> sorry, AlexPassant, I do not recognize a party named '+353149aacc' 14:05:15 <AlexPassant> Zakim, +3539149aacc is me 14:05:21 <AxelPolleres> sandro: we might switch providers for UK number again 14:05:22 <NicholasH> Axel: looking at chaning to a new provider due to there only being two phone lines 14:05:23 <Zakim> +AlexPassant; got it 14:05:42 <NicholasH> Axel: next regular meeting will be in one week 14:06:13 <AxelPolleres> don't miss the Update semantics call this friday 14:06:34 <AndyS> @16:00 WEST 14:06:40 <SteveH> Zakim, what's the phone number 14:06:40 <Zakim> I don't understand 'what's the phone number', SteveH 14:06:53 <NicholasH> Axel: looked through the issues lists and there are some old issues relating to Protocol that needs to be looked at 14:07:12 <AxelPolleres> lee, you there? 14:07:14 <NicholasH> Axel: anyone interested in a seperate phone call to disccuss them? 14:07:20 <AxelPolleres> Zakim, who is on the phone? 14:07:20 <Zakim> On the phone I see AndyS, Sandro, Souri, MattPerry, NicholasH, LeeF, pgearon_, AxelPolleres, AlexPassant, Ivan 14:07:45 <AndyS> Maybe better to start some material to discuss first ?? 14:08:03 <AxelPolleres> ACTION: axel to evaluate with LEe necessity of protocol dedicated TC, and in case send out doodle poll 14:08:04 <trackbot> Created ACTION-284 - Evaluate with LEe necessity of protocol dedicated TC, and in case send out doodle poll [on Axel Polleres - due 2010-08-03]. 14:08:12 <AndyS> Maybe better to start with some material to discuss first ?? even if just a reasonably complete issues list 14:08:49 <AxelPolleres> topic: shortcuts in update 14:08:51 <NicholasH> Axel: look at issues that don't have an obvious issue and put them on the aggenda for seperate call 14:08:56 <Zakim> +[IPcaller] 14:09:03 <SteveH> Zakim, [IPcaller] is me 14:09:03 <Zakim> +SteveH; got it 14:09:42 <AxelPolleres> alex: suggestion for 3 for moving data between graphs (mv,cp, cat) 14:10:02 <AndyS> MOVE, COPY, ADD 14:10:49 <NicholasH> alex: will make it easier to teach 14:11:00 <AxelPolleres> alex: concerns raised, shortcuts for language not yet existing, increased learning curve ... I still think that these operations are so common that shortcuts are justified. 14:11:04 <AxelPolleres> q? 14:11:38 <NicholasH> axel: issues relating to the HTTP protocol document 14:12:06 <NicholasH> if we agree on the shortcuts then they should also be in the HTTP document 14:12:31 <NicholasH> should also be available for use with the semicolon operators 14:12:32 <SteveH> I think it's a bad idea 14:12:50 <AxelPolleres> ack SteveH 14:13:03 <NicholasH> SteveH: it is a new language, alsomost nobody is using it in nager 14:13:03 <LeeF> I feel pretty strongly about this, but I will not object 14:13:07 <AlexPassant> agreed on your 2 comments re. HTTP protocol + use of �[C; 14:13:18 <NicholasH> SteveH: we don't know what operations those verbs should map to 14:13:35 <NicholasH> SteveH: move could map to almost anything 14:13:51 <NicholasH> SteveH: can't guaranteee atomisity 14:14:06 <NicholasH> SteveH: syntactic sugar crazyness 14:14:20 <Souri> s/atomisity/atomicity/ 14:15:03 <NicholasH> Axel: wouldn't not change anything, because we already have multiple statement problems 14:15:48 <AxelPolleres> SteveH: we introduce a new problem, because we could have engines that can deal with single statement requests atomically, but not the shortcuts 14:15:58 <NicholasH> SteveH: the shortcut could be an alias, so no guarantee of atomicity, confusing to users 14:16:01 <AlexPassant> SteveH: is your issue (besides atomicity) related to MOVE only or to the 3 proposals ? 14:16:47 <AndyS> My position has changed - 14:17:00 <NicholasH> Alex: Steve and AndyS have concerns but not objections 14:17:37 <NicholasH> AndyS: initially I wasn't a support but not I think that it is intuative to what users want to do 14:17:47 <SteveH> MOVE is clearly not intuative 14:18:18 <NicholasH> AndyS: it is intuative to me 14:18:19 <ivan> s/intuative/intuitive/ 14:18:19 <AlexPassant> SteveH: it's mapped to unix mv 14:19:16 <Souri> minor comment: Do we really need MOVE? or is it just a RENAME? 14:19:24 <SteveH> exactly 14:19:25 <NicholasH> Alex: it is not clear if people would object to it 14:19:28 <LeeF> exactly indeed 14:19:40 <LeeF> we don't know what we need because this language doesn't exist yet 14:19:42 <NicholasH> Alex: rename would have more support than move? 14:19:51 <AlexPassant> s/Alex/axel 14:19:57 <NicholasH> sorry :( 14:20:13 <LeeF> SteveH++++ 14:20:19 <NicholasH> SteveH: language isn't old enough to need shortcuts 14:20:39 <NicholasH> SteveH: not know if it is needed yet. And can't change seantics in 12 months time 14:21:03 <NicholasH> Axel: too controversal right now 14:21:30 <NicholasH> Axel: put +1 if you are in favour of shortcuts, -1 if you are against 14:21:31 <SteveH> -1 14:21:31 <AxelPolleres> strawpoll, +1 for shortcuts, 0 neutral, -1 if against 14:21:32 <NicholasH> -1 14:21:36 <AxelPolleres> 0 14:21:37 <LeeF> Straw poll: Including MOVE, COPY, ADD shortcuts in update 14:21:46 <LeeF> -1 14:21:47 <sandro> +1 14:21:47 <AndyS> +1 (for editors to draft some text) 14:21:51 <pgearon> 0 14:21:53 <ivan> 0 14:21:55 <MattPerry> -1 14:21:57 <Souri> 0.1 14:21:57 <AlexPassant> +1 14:22:56 <SteveH> there are other things I would prefer the update editors work on 14:23:03 <SteveH> given limited time 14:23:27 <SteveH> q+ 14:23:35 <NicholasH> Axel: agrees that we would like to see other issues worked on more than shortcuts 14:23:50 <NicholasH> Axel: should prioritise other issues 14:24:18 <NicholasH> SteveH: reserve right to change his mind, if spec doesn't get published within 9 months 14:24:45 <NicholasH> SteveH: concern is that it is too soon to add 14:25:16 <AxelPolleres> ISSUE: Shall we add shortcuts for update as proposed in http://lists.w3.org/Archives/Public/public-rdf-dawg/2010JulSep/0053.html 14:25:16 <trackbot> Created ISSUE-59 - Shall we add shortcuts for update as proposed in http://lists.w3.org/Archives/Public/public-rdf-dawg/2010JulSep/0053.html ; please complete additional details at http://www.w3.org/2009/sparql/track/issues/59/edit . 14:25:19 <NicholasH> Axel: will open an issue to re-look at it later 14:25:28 <LeeF> ack SteveH 14:25:38 <NicholasH> Axel: thanks for the disussion 14:26:00 <NicholasH> Axel: made a summary of the issues 14:26:02 <AxelPolleres> topic: issues list 14:26:03 <AxelPolleres> http://www.w3.org/2009/sparql/track/issues/open 14:26:53 <LeeF> I saw Andy's latest mail and agree with it. 14:27:08 <NicholasH> Axel: There has been some disscussion on ISSUE-1 already 14:27:45 <NicholasH> Axel: should name it as an extension 14:28:17 <AxelPolleres> ACTION: dig out results on previous discussion on ISSUE-1 14:28:17 <trackbot> Sorry, couldn't find user - dig 14:28:37 <NicholasH> Axel: next ISSUE-16 14:28:56 <NicholasH> Axel: had a lot of discussion at the last face-to-face 14:29:06 <NicholasH> Axel: left open to monitor it 14:30:15 <NicholasH> Axel: ISSUE-18 is strongly connected to the disccussion about atomicity and transactions 14:30:38 <NicholasH> Axel: any implementation SHOULD guarantee atomicity 14:30:50 <AxelPolleres> ISSUE-26 Conjunction of operation vs atomocity, transactions 14:30:53 <pgearon> I believe that we've addressed several of issues for Update, but the issues have remained open 14:31:26 <AxelPolleres> PROPOSED: to close ISSUE-18 and ISSUE-26 with the insight that we require that any compliant implementation SHOULD treat every HTTP request atomically, and that we don't want to go any further in specifying transacionality and concurrency 14:31:59 <SteveH> +1 14:32:05 <pgearon> +1 14:32:06 <ivan> 1 14:32:09 <NicholasH> +1 14:32:10 <AxelPolleres> +! 14:32:16 <AxelPolleres> +1 14:32:18 <MattPerry> +1 14:32:23 <AlexPassant> +1 14:32:24 <AndyS> +1 14:32:28 <Souri> +1 14:32:38 <AxelPolleres> RESOLVED: to close ISSUE-18 and ISSUE-26 with the insight that we require that any compliant implementation SHOULD treat every HTTP request atomically, and that we don't want to go any further in specifying transacionality and concurrency 14:32:46 <AxelPolleres> close ISSUE-18 14:32:46 <trackbot> ISSUE-18 Concurrency in SPARQL/update closed 14:32:54 <AxelPolleres> close ISSUE-26 14:32:55 <trackbot> ISSUE-26 Conjunction of operation vs atomocity, transactions closed 14:33:13 <NicholasH> Axel: next ISSUE-19 14:33:30 <AxelPolleres> ISSUE-19 Security issues on SPARQL/UPdate 14:33:57 <pgearon> The current text is: "Exposing RDF data for update creates many security issues which any deployment must be aware of, and consider the risks involved. This submission does not describe such issues." 14:34:14 <NicholasH> Axel: should there be a seperate section on Security? 14:34:19 <AndyS> c.f. http://www.w3.org/TR/rdf-sparql-query/#security 14:34:52 <AxelPolleres> http://lists.w3.org/Archives/Public/public-rdf-dawg/2010JulSep/0141.html 14:35:16 <SteveH> -1 to being vague here 14:35:23 <SteveH> that doesn't sound like a good thing to do 14:35:31 <AlexPassant> proposal is "the specification does not address security concerns related to SPARQL/Update and that implementers and users MUST be aware of security concerns when allowing SPARQL/Update on their dataset" 14:35:33 <AxelPolleres> "the specification does not address security concerns related to SPARQL/Update and that implementers and users MUST be aware of security concerns when allowing SPARQL/Update on their dataset" from Alex' mail 14:35:33 <NicholasH> Axel: should not attempt to list all the secutiry issues, but should outline some of the high level problems 14:35:35 <AxelPolleres> q? 14:35:39 <SteveH> q+ 14:35:46 <AlexPassant> SteveH: my concern is why mention issue X and not issue Y 14:36:10 <AlexPassant> q+ 14:36:20 <SteveH> ack me 14:36:24 <NicholasH> SteveH: there are a number of very serious security issues which should be addressed, such as putting server in DMZ 14:36:48 <AxelPolleres> ack AlexPassant 14:37:00 <SteveH> we should mention all of them! 14:37:08 <NicholasH> Alex: my concern is that if you list some issues but not others, it might confuse people 14:37:39 <AndyS> Something like: "there are security issues, that include, but are not limited to, .... (some important ones) ..." 14:37:42 <NicholasH> Axel: just say "some issues include" 14:37:46 <AxelPolleres> Axel: we could say" issue include, but are not limited to" 14:37:50 <Souri> s/secutiry/security/ 14:37:55 <pgearon> +q 14:38:14 <AndyS> .... and if you want something listed, you get to propose text. 14:38:14 <NicholasH> Axel: need someone to take an action to summarise these issues 14:38:30 <NicholasH> p 14:38:48 <NicholasH> pgearon: not trying to enumerate the issues but gives people a starting point 14:38:50 <SteveH> +1 to pgearon 14:39:03 <AxelPolleres> Paul: a central section to collect issues in the draft may actually be a good idea. 14:39:23 <NicholasH> pgearon: in favour creating a security section 14:40:03 <NicholasH> pgearon: I am the Update editor, it should be me 14:40:16 <AxelPolleres> Action: Paul to collect update security issues and report back to mailinglist regarding ISSUE-19 14:40:16 <trackbot> Created ACTION-285 - Collect update security issues and report back to mailinglist regarding ISSUE-19 [on Paul Gearon - due 2010-08-03]. 14:40:51 <AxelPolleres> ISSUE-22 Support of SOAP/WSDL in protocol for SPARQL/Update 14:41:46 <NicholasH> Axel: think this issue just need closing after checking minor changes 14:41:58 <AxelPolleres> PROPOSED: to close ISSUE-22 with the understanding that SPARQL1.1 only standardizes HTTP bindings, and will specify these in WSDL2.0. 14:42:52 <NicholasH> AndyS: if it is defined in WSDL2, what is the difference between that and SOAP 14:43:33 <ivan> q+ 14:43:41 <pgearon> q- 14:44:35 <NicholasH> ivan: What ever was defined for SOAP in SPARQL Query 1.0 will that be supported in SPARQL 1.1? (for query?) 14:45:24 <NicholasH> ivan: if am a user of SPARQL 1.0 with SOAP, will I still be able to do that in a years time? 14:45:57 <NicholasH> ivan: are we dropping a feature from SPARQL 1.0? 14:46:06 <AxelPolleres> ivan: are we chopping a feature of SPARQL1.0 here? 14:46:25 <AxelPolleres> ... and how large is the usage of SPARQL1.0 SOAP? 14:46:27 <pgearon> +q 14:46:46 <ivan> ack ivan 14:47:02 <AlexPassant> should we poll the community re. SOAP ? 14:47:04 <AxelPolleres> lee, are you in a position to say somthing about the SOAP issue? 14:47:17 <NicholasH> pgearon: when this issue came up before, nobody on the working group was using the SOAP bindings 14:47:49 <NicholasH> LeeF: procedurally this has to go into the draft 14:47:59 <NicholasH> pgearon: should also ask on the mailing lists 14:48:14 <AxelPolleres> s/LeeF/sandro/ 14:48:18 <pgearon> q- 14:48:25 <NicholasH> anyone who has a problem with this, please let us know 14:49:04 <ivan> q+ 14:49:16 <AndyS> sec 5 : "Removed the section on SOAP bindings, and referred to other WSDL bindings in general" 14:49:21 <sandro> sandro: it should be a big, red, editor's note saying "Hey, we're dropping this because we think no one's using it. If you are using it, please let us know!" 14:49:34 <AndyS> and some text mention of SOAP ... so to Sandro's Q: no. 14:50:50 <AxelPolleres> ACTION: LeeF add a note on dropping of SOAP binding to next WD of protocol11 and explicitly solicit feedback on usage of SOAP in SPARQL 14:50:50 <trackbot> Created ACTION-286 - Add a note on dropping of SOAP binding to next WD of protocol11 and explicitly solicit feedback on usage of SOAP in SPARQL [on Lee Feigenbaum - due 2010-08-03]. 14:51:12 <ivan> [[[For all new features, backwards compatibility with the current version of SPARQL is of great importance. All queries, that are valid in the January 2008 version of SPARQL, should remain valid in the new version and should produce identical results. For each new feature, if there is doubt or a perceived problem with respect to this, the guideline should be to not include the feature in the set of additions. 14:51:13 <ivan> ]] 14:51:14 <AxelPolleres> leef, hope that's ok. 14:51:47 <NicholasH> ivan: sorry for going very formal. I want back to the charter. It doesn't not given clear text on if we are allowed to drop the SOAP binding, if you like it or not 14:52:29 <NicholasH> Axel: but we don't have to add update suppport for the SOAP bindings 14:53:05 <NicholasH> Axel: if the SPARQL protocol, says that just one of the other needs to be supported 14:53:29 <NicholasH> Axel: I don't think that the old spec says that both have to be supported 14:53:57 <NicholasH> Axel: I will add this to the protocol issue that we have for seperate telecon 14:54:17 <NicholasH> Axel: will keep this open for the moment 14:54:20 <AxelPolleres> we can't close this issue for now 14:54:48 <NicholasH> ISSUE-23 Content negotiation/switch for mediatype 14:54:50 <AxelPolleres> ISSUE-23 Content negotiation/switch for mediatype 14:56:29 <NicholasH> Axel: could close this issue with the insight that the Content negotiation text in the protocol document is good enough 14:56:31 <AxelPolleres> PROPOSED: Close ISSUE-23 with the insight that the current content HTTP negotiation mechanism (as discussed in sparql11-protocol) is sufficient. 14:57:11 <AndyS> q+ 14:57:18 <ivan> ack ivan 14:57:37 <LeeF> FYI I've explicitly sought SOAP experiences in the past and completely failed to find any 14:57:47 <NicholasH> AndyS: slight issue with the wording, we don't do Content negotiation, the HTTP protocol does 14:58:03 <AxelPolleres> PROPOSED: Close ISSUE-23 with the insight that HTTP negotiation mechanism (as discussed in sparql11-protocol) is sufficient. 14:58:07 <AndyS> q- 14:58:17 <AndyS> seconded 14:58:22 <AxelPolleres> +1 14:58:24 <ivan> 0 14:58:25 <NicholasH> +1 14:58:35 <pgearon> +1 14:58:36 <ivan> +0 14:58:37 <pgearon> +1 14:58:59 <AxelPolleres> RESOLVED: Close ISSUE-23 with the insight that HTTP negotiation mechanism (as discussed in sparql11-protocol) is sufficient. 14:59:08 <AxelPolleres> close ISSUE-23 14:59:08 <trackbot> ISSUE-23 Content negotiation/switch for mediatype closed 14:59:17 <Zakim> -LeeF 14:59:19 <ivan> zakim, drop me 14:59:19 <Zakim> Ivan is being disconnected 14:59:21 <Zakim> -Ivan 14:59:30 <Zakim> -Sandro 14:59:32 <Zakim> -pgearon_ 14:59:32 <AxelPolleres> adjourned 14:59:33 <Zakim> -AlexPassant 14:59:36 <Zakim> -MattPerry 14:59:37 <AxelPolleres> (time's up) 14:59:38 <Zakim> -SteveH 14:59:42 <Zakim> -AndyS 14:59:48 <AndyS> ADJOURNED 14:59:49 <Zakim> -Souri 15:00:28 <AxelPolleres> reminder ... TC on Update formal semantics ... Fri 30th July 4pm UK time, 11am Eastern time 15:00:29 <Zakim> -NicholasH 15:00:32 <AxelPolleres> bye all 15:00:33 <NicholasH> wow, that was hard! 15:00:58 <AxelPolleres> Nicholas, thanks for scribing, I will take care ofpublishing the minutes! 15:01:14 <AxelPolleres> rrsagent, make records, public 15:01:14 <RRSAgent> I'm logging. I don't understand 'make records, public', AxelPolleres. Try /msg RRSAgent help 15:01:25 <AxelPolleres> rrsagent, make records public # SPECIAL MARKER FOR CHATSYNC. DO NOT EDIT THIS LINE OR BELOW. SRCLINESUSED=00000288