Chatlog 2011-11-15

From SPARQL Working Group
Revision as of 16:04, 15 November 2011 by Sandro (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
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.

14:51:24 <RRSAgent> RRSAgent has joined #sparql
14:51:24 <RRSAgent> logging to
14:51:26 <trackbot> RRSAgent, make logs world
14:51:26 <Zakim> Zakim has joined #sparql
14:51:28 <trackbot> Zakim, this will be 77277
14:51:28 <Zakim> ok, trackbot; I see SW_(SPARQL)10:00AM scheduled to start in 9 minutes
14:51:29 <trackbot> Meeting: SPARQL Working Group Teleconference
14:51:29 <trackbot> Date: 15 November 2011
14:52:09 <AndyS> Query doc isn't finished yet -- see PosLastCall page.
14:52:19 <AndyS>  wiki:PostLastCall
14:52:24 <AxelPolleres_> agenda: (essentially the same points, we'll run through status/progress made and PostLastCall page)
14:52:31 <AxelPolleres_> chair: AxelPolleres
14:52:59 <AxelPolleres_>
14:53:38 <kasei> ok, thanks AndyS 
14:55:24 <AxelPolleres_> regrets:BirteGlimm
14:58:35 <AxelPolleres_> As regards "future work", lee has started a new page:
14:59:25 <MattPerry> MattPerry has joined #sparql
14:59:53 <AxelPolleres_> Open issue from last time: assign reviewers for protocol
15:00:21 <kasei> Zakim, I hate you.
15:00:21 <Zakim> I don't understand 'I hate you', kasei
15:00:24 <Zakim> SW_(SPARQL)10:00AM has now started
15:00:31 <Zakim> +kasei
15:00:43 <Zakim> + +1.603.897.aaaa
15:00:45 <kasei> wow. that seems to have worked. need to curse Zakim more often.
15:00:56 <MattPerry> zakim, aaaa is me
15:00:56 <Zakim> +MattPerry; got it
15:01:03 <Zakim> + +49.897.aabb
15:01:19 <AxelPolleres_> Zakim, aabb is probably me
15:01:23 <Zakim> +[Sophia]
15:01:25 <Zakim> +AxelPolleres; got it
15:02:02 <AxelPolleres_> Zakim, who is on the phone?
15:02:04 <LeeF> Zakim, I hate you.
15:02:05 <Zakim> On the phone I see kasei, MattPerry, AxelPolleres, [Sophia]
15:02:09 <Zakim> I don't understand 'I hate you', LeeF
15:02:13 <Zakim> + +1.617.553.aacc
15:02:15 <LeeF> sweet, works like a charm
15:02:17 <Zakim> +??P12
15:02:18 <LeeF> zakim, aacc is me
15:02:19 <Zakim> +LeeF; got it
15:02:22 <AndyS> zakim, ??P12 is me
15:02:22 <Zakim> +AndyS; got it
15:02:23 <kasei> there's no agenda online?
15:02:44 <pgearon> pgearon has joined #sparql
15:02:45 <kasei> ah, ok.
15:03:14 <Zakim> +pgearon
15:03:15 <AxelPolleres_> Zakim, who is on the phone?
15:03:15 <Zakim> On the phone I see kasei, MattPerry, AxelPolleres, [Sophia], LeeF, AndyS, pgearon
15:04:21 <chimezie> chimezie has joined #sparql
15:05:06 <AxelPolleres_> Zakim, who is on the phone?
15:05:06 <Zakim> On the phone I see kasei, MattPerry, AxelPolleres, [Sophia], LeeF, AndyS, pgearon
15:05:12 <Zakim> + +1.216.368.aadd
15:05:35 <chimezie> Zakim, +1.216.368.aadd is me
15:05:35 <Zakim> +chimezie; got it
15:05:50 <chimezie> Zakim, mute me
15:05:50 <Zakim> chimezie should now be muted
15:06:25 <Zakim> +Sandro
15:06:45 <AxelPolleres_> sandro, can you scribe?
15:06:51 <bglimm> bglimm has joined #sparql
15:06:54 <Zakim> -Sandro
15:07:07 <AxelPolleres_> scribe: AxelPolleres
15:07:14 <AxelPolleres_> topic: admin
15:07:18 <Zakim> +sandro
15:07:32 <AxelPolleres_> scribe: sandro
15:07:44 <chimezie> I'm getting caught up with my ACTION items outstanding (unable to last week)
15:08:06 <sandro> axel: Agenda is reviewing status of documents, checking on reviews; see where we are toward the next publication round.
15:08:24 <AxelPolleres_> PROPOSED: Approve minutes at
15:08:56 <AxelPolleres_> RESOLVED: Approve minutes at
15:09:47 <sandro> axel: (looking for next scribe)
15:10:13 <chimezie> sure
15:10:23 <AxelPolleres_> scribe for next week: chime
15:10:44 <sandro> topic: liaison with RDF WG
15:10:50 <sandro> axel: some email;   comment Andy?
15:11:20 <sandro> andy: No current decisions by RDF WG affecting us.    Just working through consequences of decisions earlier on, eg datatype changes
15:11:25 <AxelPolleres_> discussion on effects of
15:11:33 <sandro> andy: There will be something about RDF Graphs, but not yet.
15:11:40 <sandro> axel: You did some changes based on datatype decision?
15:12:36 <sandro> andy: Because all literals will have datatypes, the DATATYPE( ) fn in sparql is affected.  It didn't cover lits with lang tags.     plain lits without lang tags were already returned as xs:string.   Now it returns rdf:LangString for lang tagged strings.
15:13:17 <sandro> axel: So we're blending in some RDF 1.1 features there.   Shall we discuss this as a group, since we already decided to stick to RDF 1.0, with only commentary about RDF 1.1.
15:13:28 <sandro> axel: any opinions?
15:13:35 <sandro> (silence)
15:13:52 <sandro> +1 I'm fine with proceeding like this.
15:14:11 <kasei> i'm hesitant to change it at this point.
15:14:23 <sandro> axel: So, we'll try to keep as close as we can to RDF 1.1 without disrupting the test suite.
15:14:47 <pgearon> Olivier had a question about loading RDF data where data from 1.1 may define multiple graphs, but Update 1.1 currently specifies a particular graph
15:15:03 <sandro> kasei: I'm wary of making changes that potentially change the results of SPARQL 1.1 queries, on something that's not a spec (REC) yet.
15:15:25 <sandro> axel: this dt is in RDF 1.1?
15:15:28 <sandro> sandro: yes
15:15:34 <sandro> axel: so we have to wait for it?
15:15:52 <AxelPolleres_> I am a bit worried processwise
15:15:57 <LeeF> We wouldn't need to wait because we're not leaning on RDF, we're just defining things ourselves
15:16:09 <LeeF> at least, that's how i understood what Andy did
15:16:15 <AndyS> +1 to LeeF
15:16:25 <sandro> sandro: we can just define what we need ourselves
15:16:33 <sandro> sandro: it's half a sentence.....
15:16:47 <sandro> axel: Does any of this affect D-Entailment and the fixed datatype map, Birte?
15:16:50 <kasei> won't this change results of in-the-wild 1.0 queries?
15:17:11 <sandro> axel: I'll ask Birte in email
15:17:39 <sandro> kasei: won't this change results of in-the-wild 1.0 queries?
15:18:09 <sandro> andy: Yes, all the RDF 1.1 changes have that potential.  Particularly with datatypes, which is one of our extension points.
15:18:13 <LeeF> Was it an error in SPARQL 1.0 or an empty sting?
15:18:19 <LeeF> s/sting/string
15:18:36 <sandro> andy: THe change to simple literals, having xs:string is also like this.
15:18:39 <AndyS> error
15:19:13 <sandro> axel: two observable changes.    plain literals with and without language tags.
15:19:15 <LeeF> I'm comfortable making a change that goes from error --> proper semantics -- I think that's consistent with SPARQL's extension mechanism in general
15:19:43 <Olivier_> The problem is that rdf:LangString is not a definitive decision in a final REC, so it may change ?
15:19:45 <chimezie> I would concur to those who understand this better than me
15:19:52 <AndyS> zakim, who is on the phone?
15:19:52 <Zakim> On the phone I see kasei, MattPerry, AxelPolleres, [Sophia], LeeF, AndyS, pgearon, chimezie (muted), sandro
15:20:02 <chimezie> s/concur/defer
15:20:02 <pgearon> +q
15:20:05 <LeeF> Olivier_, I do share that concern... it's a little risky
15:20:51 <sandro> lee: there is some risk of RDF WG changing its mind
15:21:05 <sandro> andy: either way we're a bit stuck.    
15:21:12 <sandro> andy: it's a no win situation
15:21:38 <AxelPolleres_> Can we ask the RDF WG at least to make some formal resolution for the URI, or is that alrady there?
15:21:43 <sandro> lee: We could just put a note in about what we expect.
15:21:52 <kasei> if we were divergent from the eventual rdf1.1 spec, though, implementors could still do extensions to align.
15:22:34 <sandro> axel: we seem fine with xs:string...
15:22:37 <sandro> andy: that's not a change
15:22:50 <sandro> andy: It was already in SPARQL 1.9
15:22:58 <sandro> s/1.9/1.0/
15:23:14 <AxelPolleres_> Alternative 1: return an error 
15:23:36 <LeeF> DATATYPE("foo"@en)
15:23:38 <AxelPolleres_> Alternative 2: return rdf:LangString 
15:23:47 <sandro> 2
15:23:48 <LeeF> 2
15:24:05 <LeeF> 2 (with note is good)
15:24:23 <sandro> andy: Don't hide it.   Put a note in the spec about the change.
15:24:27 <LeeF> Don't think somewhere else would be "hiding", but still happy to just include it
15:24:27 <AndyS> Alternative 3: as alt 2 with a note about the change
15:24:43 <sandro> axel: We could use "AT RISK" so we can drop it during CR.
15:24:54 <AxelPolleres_> 2 (mark explicitly at risk)
15:24:58 <MattPerry> 1
15:24:58 <chimezie> which alternative is at risk?
15:25:17 <chimezie> 2
15:25:19 <AndyS> 2 or 3
15:25:22 <Olivier_> 2
15:25:27 <kasei> 0
15:25:40 <sandro> sandro: Yeah, I think "AT RISK" can say datatype URI may change.
15:26:24 <sandro> MattPerry: This is a mild preference, toward letting implementors do the extension, but I wouldn't lie in the road over this, no.
15:26:31 <AxelPolleres_> direction seems to be to go for Alternative 2 with an AT RISK note.
15:26:41 <sandro> axel: sound like:  we'll go with alternative 2, with an AT RISK note.
15:26:50 <Zakim> +??P2
15:27:24 <sandro> axel: Any more about Query document?    How close is it to review-ready.
15:27:29 <AxelPolleres_> topic: query
15:27:32 <sandro> andy: I'm working through the list of work items.
15:27:49 <sandro> steve: someone spotted an order consistency problem
15:28:04 <sandro> steve: mismatch between my head and [something].
15:28:18 <sandro> axel: Reviewable by next telecon?
15:28:26 <sandro> steve: I'm aiming for it, but ... 50/50.
15:28:36 <AndyS> Unlikely on my part.
15:28:45 <AxelPolleres_> we aim at reviewable state by next Telco?
15:28:58 <bglimm> I plan to review the non-algebra sections tomorrow. Is that ok (sorry, only IRC)
15:29:00 <sandro> axel: Birte will review in parallel, in order to have a chance to vote within 2 weeks.  Okay?
15:29:26 <pgearon> q-
15:29:26 <kasei> yes
15:29:26 <AndyS> bglimm - doc not finished, hopefully local changes only but several to do.
15:29:40 <bglimm> ok, then I better wait
15:29:47 <sandro> axel: Greg, can you start reviewing?
15:29:52 <sandro> kasei: Whenever it's ready
15:30:01 <bglimm> or you send a list naming the sections that are ready
15:30:14 <swh> swh has joined #sparql
15:30:27 <swh> Zakim, who's on the phone?
15:30:27 <Zakim> On the phone I see kasei, MattPerry, AxelPolleres, [Sophia], LeeF, AndyS, pgearon, chimezie (muted), sandro, ??P2
15:30:33 <swh> Zakim, ??P2 is me
15:30:33 <Zakim> +swh; got it
15:30:53 <sandro> topic: Overview, Fed pubs
15:31:01 <sandro> axel: Docs are ready.
15:31:55 <sandro> sandro: I think we can publish today.  I might need a little help.
15:32:20 <sandro> sandro: I'll do my best.
15:32:36 <sandro> axel: announcing them?
15:33:28 <sandro> axel: wait for it on, then re-announce as you like
15:34:01 <chimezie> Zakim, unmute me
15:34:01 <Zakim> chimezie should no longer be muted
15:34:05 <sandro> topic: graph store protocol
15:34:25 <sandro> axel: I think this is ready?     we just need feedback from all the commenters.
15:34:49 <sandro> chimezie: I only need to followup with people with non-substantive comment, and inconclusive comments.
15:35:16 <sandro> chimezie: working through them, then it'll be ready for review
15:35:33 <sandro> axel: Send mail to list when it's ready to review.  Then Andy and Sandro will review
15:35:41 <sandro> chimezie: okay
15:36:21 <sandro> topic: Service Description
15:36:34 <sandro> axel: I didn't find time to do my review yet.   
15:37:07 <AxelPolleres> AxelPolleres has joined #sparql
15:37:07 <sandro> (axel having IRC problems.)
15:37:26 <sandro> topic: ENtailment Regimes
15:37:38 <sandro> axel: Birte says second review by Markus Kr.
15:37:44 <sandro> topic: Update
15:37:52 <bglimm>  Yes, Markus will review
15:37:52 <AxelPolleres>
15:37:54 <sandro> axel: post-last-call page says some things have been done
15:38:22 <bglimm> I've send out replies to Michael Schneider last week, but haven't received feedback yet
15:39:20 <sandro> axel: have one review; document is close to reviewable.
15:39:31 <sandro> axel: second reviewer = Matt.
15:39:40 <sandro> MattPerry: I'll wait for the next version
15:39:49 <sandro> topic: Protocol
15:39:53 <AndyS> pgearon - comments RC-4 and DB-4 -- have had time to look/complete the update parts of the draft response?
15:40:12 <sandro> lee: Greg and I have done most of what needs to be done; document is ready to review.
15:40:39 <AxelPolleres> ACTION: AndyS to review protocol
15:40:40 <trackbot> Created ACTION-559 - Review protocol [on Andy Seaborne - due 2011-11-22].
15:40:52 <sandro> lee: there are three open protocol issues, on tracker -- once reviewers have looked over them, we can close those issues next week, based on what's in current document.
15:41:05 <sandro> axel: reviews by Andy, and ...?
15:41:25 <pgearon> AndyS: not sure. I've done a number of things lately, and I don't recall if those are in the mix. Will get back to you as soon as I'm done with the review from Olivier
15:41:30 <chimezie> Zakim, mute me
15:41:30 <Zakim> chimezie should now be muted
15:41:43 <swh> what's the deadline?
15:42:03 <AndyS> pgearon - I'll see updates to the wiki via RSS.
15:42:23 <AxelPolleres> Olivier, Carlos, any chance foryou to take a second review on board?
15:42:38 <Olivier_> Yes
15:42:38 <sandro> SteveH: I can't morally do it until I'm done with Query.  Maybe I can rope someone here into doing it.
15:42:41 <LeeF> ACTION: Steve to make a best effort attempt to review or delegate a review for protocol
15:42:41 <trackbot> Created ACTION-560 - Make a best effort attempt to review or delegate a review for protocol [on Steve Harris - due 2011-11-22].
15:42:53 <LeeF> ACTION: Olivier to review SPARQL protocol document
15:42:54 <trackbot> Created ACTION-561 - Review SPARQL protocol document [on Olivier Corby - due 2011-11-22].
15:43:15 <sandro> lee: Greg and I will send an email later today saying when it's totally ready for review.
15:43:28 <sandro> topic: JSON Results Format
15:43:43 <sandro> axel: We said this doesn't need review.   ready for CR.
15:43:49 <sandro> topic: Comments
15:44:06 <sandro> axel: Nothing substantial new.   Please update the wiki page as you handle comments.
15:44:25 <sandro> topic: Any Other Business
15:45:17 <sandro> axel: Paul, was there anything else you needed to discuss?
15:45:53 <sandro> pgearon: One thing that came up was the LOAD operation, which Olivier asked about.   If RDF 1.1 has a syntax for multiple graphs, will the LOAD command need that.
15:46:00 <AxelPolleres>
15:46:27 <sandro> axel: which comment...?
15:46:37 <AxelPolleres> Olivier's review
15:47:58 <AndyS> Fuseki (will be) implementing loading quads if LOAD <foo> and no INTO -- user request and expectation.
15:48:06 <Zakim> -chimezie
15:48:09 <sandro> pgearon: Not taking into account this direction of RDF 1.1 is a cause for concern for me
15:48:17 <sandro> axel: This would rather go into a note
15:48:45 <sandro> pgearon: Also, the additional copy/move shortcuts.  At Risk.
15:48:50 <AxelPolleres> COPY, ADD MOVE... still marked as AT RISK
15:49:07 <LeeF> I feel the same way I felt all along
15:49:08 <sandro> AxelPolleres: Only feedback was to keep it.
15:49:16 <LeeF> Keep it AT RISK and see who does / does not implement it
15:49:43 <sandro> LeeF: If people implement it, we'll keep it.
15:50:21 <sandro> lee: "At Risk" is about implementation exerpeince, during CR.
15:50:49 <sandro> AndyS: We've had feedback already; why not decide.
15:51:16 <sandro> lee: We're waiting to see how it shows up in implementation report during CR.
15:51:27 <sandro> SteveH: We've implemented it, as has Andy.
15:51:27 <AxelPolleres> Andy, Steve have implemented it.
15:51:31 <sandro> Lee: We have not.
15:51:58 <AxelPolleres> (Discussion whether or not to leave the AT RISK note in)
15:52:11 <AxelPolleres> sandro: let's strive for only as many AT RISK notes as we need.
15:52:47 <pgearon> +1 to remove AT RISK note
15:52:58 <AndyS> Note it is the grammar in rq25
15:53:01 <swh> +1 -- IFF there's 3+ implementations, -1 otherwise
15:53:04 <MattPerry> +1
15:53:05 <LeeF> -1
15:53:09 <Olivier_> +1
15:53:10 <AndyS> +1
15:53:13 <kasei> -1
15:53:14 <sandro> 0 insufficient data
15:53:19 <AxelPolleres> strawpoll: +1 to remove the at risk note at
15:53:29 <AxelPolleres> 0 
15:53:50 <swh> pgearon, have you implemented it?
15:54:03 <pgearon> no. Have only just started on Update implementation
15:54:11 <pgearon> yes
15:54:11 <swh> then I'm -1
15:55:28 <AxelPolleres> Has sesame implemented it?
15:55:31 <LeeF> I think Anzo does implement it
15:56:14 <LeeF> 3 on the call
15:56:22 <LeeF> including 1 who thinks it's a bad idea anyway :D
15:56:50 <sandro> axel: Is that an objection Lee?
15:57:15 <sandro> Lee: Turns we did implement it, so that's three impls on the call now.
15:57:29 <sandro> lee: no objection
15:58:13 <AxelPolleres> PROPOSED: remove the AT RISK note for COPY, ADD, MOVE in Update
15:58:15 <swh> seconded
15:58:17 <sandro> +1
15:58:22 <AxelPolleres> +1
15:58:23 <Olivier_> +1
15:58:24 <kasei> 0
15:58:27 <MattPerry> +1
15:58:35 <AndyS> +1
15:58:38 <LeeF> 0
15:59:07 <pgearon> +1
15:59:31 <AxelPolleres> RESOLVED: remove the AT RISK note for COPY, ADD, MOVE in Update (7 pro, 2 abstentions, no objections)
15:59:55 <AxelPolleres> adjourned
15:59:55 <sandro> ADJOURNED
15:59:56 <Zakim> -LeeF
15:59:57 <MattPerry> bye
15:59:58 <Zakim> -[Sophia]
15:59:59 <Zakim> -swh
16:00:00 <Zakim> -kasei
16:00:03 <Zakim> -sandro
16:00:04 <Zakim> -pgearon
16:00:09 <Zakim> -AxelPolleres
16:00:10 <Zakim> -MattPerry
16:00:26 <AxelPolleres> rrsagent, make records public