Chatlog 2011-08-23

From SPARQL Working Group
Jump to: navigation, search

See original RRSAgent log and preview nicely formatted version.

Please justify/explain all edits to this page, in your "edit summary" text.

14:00:41 <RRSAgent> RRSAgent has joined #sparql
14:00:41 <RRSAgent> logging to
14:00:43 <trackbot> RRSAgent, make logs world
14:00:43 <Zakim> Zakim has joined #sparql
14:00:45 <trackbot> Zakim, this will be 77277
14:00:45 <Zakim> ok, trackbot; I see SW_(SPARQL)10:00AM scheduled to start now
14:00:46 <trackbot> Meeting: SPARQL Working Group Teleconference
14:00:46 <trackbot> Date: 23 August 2011
14:00:48 <LeeF> zakim, this will be sparql
14:00:48 <Zakim> ok, LeeF, I see SW_(SPARQL)10:00AM already started
14:01:02 <Zakim> +pgearon
14:01:04 <kasei> Zakim, who is on the phone?
14:01:04 <Zakim> On the phone I see ??P5, kasei, Bert, MattPerry, pgearon
14:01:33 <Zakim> +LeeF
14:02:22 <cbuilara> I'm also on the phone
14:02:41 <LeeF> zakim, Bert is cbuilara
14:02:41 <Zakim> +cbuilara; got it
14:02:45 <LeeF> zakim, ??P5 is Olivier
14:02:47 <Zakim> +Olivier; got it
14:02:56 <LeeF> Chair: LeeF
14:03:00 <LeeF> Scribenick: MattPerry
14:03:46 <LeeF> Regrets: Birte, Axel, Alex, Andy, Sandro
14:04:17 <LeeF> Agenda:
14:04:35 <LeeF> topic: Admin
14:04:43 <LeeF> PROPOSED: Approve minutes at
14:05:11 <LeeF> RESOLVED: Approve minutes at
14:06:11 <MattPerry> LeeF: Carlos, can you scribe next week
14:06:15 <cbuilara> yes, I can
14:06:15 <LeeF> Next meeting is 30-August, scribe Carlos
14:06:18 <Zakim> + +1.216.368.aaaa
14:06:33 <kasei> Zakim, who is talking?
14:06:44 <Zakim> kasei, listening for 10 seconds I heard sound from the following: LeeF (90%)
14:06:45 <chimezie> chimezie has joined #sparql
14:06:52 <LeeF> zakim, aaaa is chimezie
14:06:52 <Zakim> +chimezie; got it
14:07:19 <SteveH__> SteveH__ has joined #sparql
14:07:31 <LeeF> regrets next week from: alex, axel, sandro
14:07:34 <MattPerry> LeeF: anyone that can't make it next week?
14:07:43 <SteveH__> probably
14:07:43 <MattPerry> silence
14:07:59 <SteveH__> Zakim, what's the code?
14:07:59 <Zakim> the conference code is 77277 (tel:+1.617.761.6200, SteveH__
14:08:07 <Zakim> +??P11
14:08:10 <SteveH__> Zakim, ??P11 is me
14:08:10 <Zakim> +SteveH__; got it
14:08:40 <LeeF> two things to add: dataset parameters in the protocol, and graph store protocol comments
14:09:00 <LeeF> topic: BASE & GRAPH & SPARQL 1.1 Update
14:09:14 <MattPerry> LeeF: question from SteveH on the mailing list
14:09:14 <LeeF>
14:09:28 <SteveH> q+
14:10:14 <MattPerry> SteveH: application sending data to triplestore ... wrap insert data URI around turtle, but it didn't work
14:10:41 <LeeF> ack SteveH
14:10:56 <MattPerry> SteveH: expected graph to change base URI, but it didn't
14:11:45 <MattPerry> LeeF: nothing says graph uri is part of base resolution chain
14:12:44 <LeeF> Relative IRIs
14:13:06 <MattPerry> LeeF: graph clause does not affect base URI
14:13:35 <MattPerry> LeeF: should we change this behavior ... what is the cost of the change
14:13:45 <MattPerry> SteveH: we shouldn't change it in query
14:14:25 <MattPerry> SteveH: it is worth explicitly mentioning this behavior in update
14:15:10 <MattPerry> pgearon: Yes. we could add an editorial note about this.
14:15:48 <LeeF> ACTION: Paul to add an editorial note to SPARQL update noting that GRAPH <uri> does not affect the base URI for triples within it for INSERT DATA (c. f.
14:15:48 <trackbot> Created ACTION-526 - Add an editorial note to SPARQL update noting that GRAPH <uri> does not affect the base URI for triples within it for INSERT DATA (c. f. [on Paul Gearon - due 2011-08-30].
14:16:00 <LeeF> q?
14:16:08 <LeeF> topic: dataset parameters in protocol
14:16:44 <MattPerry> LeeF: don't want to try to pass a resolution because Andy is not here
14:16:47 <LeeF>
14:17:00 <MattPerry> LeeF: any comments on this?
14:17:06 <SteveH> q+
14:17:10 <LeeF> ack SteveH
14:17:16 <kasei> I took a quick look. Looks OK to me, but don't have any real preference.
14:17:38 <MattPerry> SteveH: I can see use cases for this proposal and it looks sensible
14:17:47 <SteveH> …at a first glance
14:18:19 <MattPerry> LeeF: We can resolve the issue next week when Andy is here
14:18:27 <LeeF> topic: Comments
14:18:50 <LeeF>
14:19:34 <MattPerry> LeeF: one new editorial comment from Richard that Andy or SteveH can look at
14:19:58 <chimezie>
14:20:00 <MattPerry> LeeF: chimezie, you sent some comments right?
14:20:08 <MattPerry> chimezie: yes
14:20:16 <chimezie>
14:20:40 <MattPerry> chimezie: first response is to Ivan
14:21:34 <MattPerry> LeeF: this would be a substantial change
14:22:00 <MattPerry> ... is his suggestion a good idea?
14:22:05 <MattPerry> chimezie: yes
14:22:24 <MattPerry> LeeF: we need a group decision becuase this is a substantive change
14:22:58 <MattPerry> ... we can aim for decisions next week
14:24:18 <MattPerry> LeeF: for second one, I don't think there's any substantive change
14:24:28 <SteveH> we do the same sniffing, and consider it conformant
14:24:59 <MattPerry> LeeF: main question is should we support multi-part form
14:25:25 <MattPerry> ... the change would require another last call
14:25:45 <MattPerry> ... there are still documents not in last call though, so it's not quite as bad
14:25:45 <SteveH> if we are going for another last call I think we should consider dropping the direct POST feature, it's not very useful in our experience
14:26:44 <MattPerry> SteveH: direct POST feature is not very useful and a significant burden on developers
14:27:28 <SteveH> …as opposed the for form-based POST request (for Update)
14:27:51 <MattPerry> chimeze: SJ-1 would require substantive changes
14:28:02 <MattPerry> s/chimeze/chimezie
14:29:13 <kasei> i would oppose the suggested requirement for cache-related headers (SJ-1 point 6)
14:29:17 <MattPerry> chimezie: need to look at SJ-1 more thoroughly
14:30:34 <MattPerry> kasei: Point 6 is more far reaching ... I support it in theory, but it is a large burden on implementors. Should not use "MUST" wording
14:31:16 <MattPerry> LeeF: any more comment responses
14:31:28 <MattPerry> kasei: I have one open for David Booth
14:31:34 <MattPerry> ... hoping for some feedback
14:31:43 <MattPerry> ... DB6
14:33:47 <MattPerry> kasei: Point 3: issue with range of sd:endpoint property
14:34:22 <MattPerry> ... is this a legitimate issue?
14:34:49 <MattPerry> LeeF: I'm not concerned ... I would just clarify to David
14:36:36 <MattPerry> ... Point 4: IRI vs URI
14:36:52 <MattPerry> LeeF: answer should fall out of other relevant specs
14:37:44 <MattPerry> kasei: Point 7 r.e. testing ... only thing we can test is whether or not it returns RDF and does it have a certain triple
14:38:14 <MattPerry> LeeF: we could test optional parts to show levels of support
14:38:53 <MattPerry> kasei: Point 10 -- distinguish between read only and allows update
14:39:14 <AxelPolleres> AxelPolleres has joined #sparql
14:39:31 <MattPerry> LeeF: this only one of many such possible properties
14:39:58 <MattPerry> kasei: Point 11 is syntactic ... I mostly agree but am hesitant to change them
14:41:07 <MattPerry> ... Point 12 same as Point 3
14:41:44 <MattPerry> ... Point 15 -- use of "default dataset" ... it is not used anywhere else
14:42:23 <MattPerry> LeeF: anyone else have comments
14:42:26 <MattPerry> silence
14:42:42 <LeeF> topic: implementation report
14:42:49 <kasei>
14:43:35 <LeeF> zakim, who's on the phone?
14:43:35 <Zakim> On the phone I see Olivier, kasei, cbuilara, MattPerry, pgearon, LeeF, chimezie, SteveH__
14:43:44 <MattPerry> kasei: have reports for RDF::Query and Rasqal
14:43:55 <SteveH> sparql 1.1 tests: sorry, no
14:43:59 <MattPerry> LeeF: anyone else have a report
14:44:07 <MattPerry> chimezie: hope to but not yet
14:44:12 <MattPerry> pgearon: I haven't
14:44:59 <MattPerry> LeeF: we talked about using EricP's scripts from SPARQL 1.0 for SPARQL 1.1
14:45:13 <MattPerry> kasei: EricP's sripts do more stuff
14:45:29 <MattPerry> LeeF: I think it will be easier to use kasei's scripts
14:45:40 <MattPerry> kasei: I would be fine with moving my stuff to W3C
14:46:39 <MattPerry> LeeF: can you bundle your code up?
14:47:11 <MattPerry> kasei: there's probably quite a bit of pain to move it all over
14:47:28 <MattPerry> LeeF: we can leave the code where it is
14:48:13 <LeeF> ACTION: Gregory to let us know where the impl report code is in git and to put the generated report file somewhere in our CVS space
14:48:13 <trackbot> Created ACTION-527 - Let us know where the impl report code is in git and to put the generated report file somewhere in our CVS space [on Gregory Williams - due 2011-08-30].
14:48:36 <MattPerry> kasei: the entailment tests aren't marked as optional
14:50:02 <MattPerry> LeeF: I thought we would have multiple super manifests, so you could pick e.g. a query test suite
14:50:20 <MattPerry> ... right now we have only 1 super manifest
14:50:45 <MattPerry> chimezie: we should do this so the tests are more modular
14:51:18 <MattPerry> LeeF: we should at least have spec granularity
14:51:29 <MattPerry> kasei: I would be happy with that level of granularity
14:52:52 <AxelPolleres> AxelPolleres has joined #sparql
14:53:05 <MattPerry> LeeF: anyone not like this way of splitting up tests?
14:53:08 <MattPerry> silence
14:53:36 <Olivier> Olivier has joined #sparql
14:53:37 <LeeF> ACTION: Lee to create new "collective" manifests at the granularity of spec conformance (query, update, entailment, fed query, json results, ...)
14:53:37 <trackbot> Created ACTION-528 - Create new "collective" manifests at the granularity of spec conformance (query, update, entailment, fed query, json results, ...) [on Lee Feigenbaum - due 2011-08-30].
14:53:59 <kasei> q+
14:54:13 <LeeF> ack kasei
14:55:19 <MattPerry> kasei: we have currently be using mf:requires for optional things ... if we split into separate parts then these annotations may need to change
14:55:43 <MattPerry> ... it's useful to have those annotations on tests
14:56:29 <MattPerry> chimezie: what does it mean in terms of implementation reports for something to be optional
14:57:09 <MattPerry> LeeF: the optional tests don't count towards conformance
14:58:10 <MattPerry> LeeF: any more issues
14:58:21 <MattPerry> pgearon: Comment DB-5
14:58:48 <MattPerry> ... first item -- add virtual graph or keep copy, add, remove shortcuts
14:59:04 <MattPerry> ... those features are still marked as at risk
14:59:32 <MattPerry> LeeF: we should basically consider this as feeback to keep these features
15:00:03 <Zakim> -pgearon
15:00:04 <MattPerry> pgearon: I'd like to remove the at risk notes for these
15:00:18 <pgearon> sorry, got kicked off
15:00:21 <MattPerry> LeeF: I think we should leave at risk and wait for implementation reports
15:00:53 <LeeF> Adjourned.
15:00:54 <chimezie> bye
15:00:57 <Zakim> -chimezie
15:01:00 <LeeF>
15:01:12 <Zakim> -cbuilara
15:01:12 <Zakim> -LeeF
15:01:13 <Zakim> -Olivier
15:01:13 <Zakim> -SteveH__
15:01:15 <Zakim> -MattPerry
15:01:16 <pgearon> Thanks
15:01:16 <Zakim> -kasei
15:01:18 <Zakim> SW_(SPARQL)10:00AM has ended
15:01:20 <Zakim> Attendees were kasei, MattPerry, pgearon, LeeF, cbuilara, Olivier, +1.216.368.aaaa, chimezie, SteveH__