Chatlog 2010-12-21

From SPARQL Working Group
Revision as of 16:15, 21 December 2010 by Apollere2 (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.

14:57:46 <RRSAgent> RRSAgent has joined #sparql
14:57:46 <RRSAgent> logging to
14:57:52 <LeeF> RRSAgent, make logs world
14:57:54 <bglimm> bglimm has joined #sparql
14:57:56 <LeeF> zakim, this will be SPARQL
14:57:56 <Zakim> ok, LeeF; I see SW_(SPARQL)10:00AM scheduled to start in 3 minutes
14:58:06 <Zakim> SW_(SPARQL)10:00AM has now started
14:58:08 <LeeF> Meeting: SPARQL Working Group Weekly Teleconference
14:58:10 <LeeF> Chair: LeeF
14:58:10 <LeeF> regrets: Alex
14:58:12 <LeeF> Scribe: Axel
14:58:14 <Zakim> + +44.208.439.aaaa
14:58:19 <LeeF> Scribenick: AxelPolleres
14:58:21 <SteveH_> Zakim, aaaa is me
14:58:21 <Zakim> +SteveH_; got it
14:58:28 <LeeF> Date: 2010-12-20
14:58:45 <Zakim> +??P3
14:58:45 <MattPerry> MattPerry has joined #sparql
14:58:48 <LeeF> hmm
14:58:49 <LeeF> that's not the date
14:58:51 <LeeF> Date: 2010-12-21
14:58:54 <LeeF> Agenda: Agenda: 
14:58:55 <cbuilara> zakim, ??P3 is me
14:58:55 <Zakim> +cbuilara; got it
14:58:57 <bijan> bijan has joined #sparql
14:59:00 <LeeF> Agenda:
14:59:06 <SteveH_> cbuilara, you have mic problems
14:59:20 <Zakim> +??P4
14:59:23 <cbuilara> ok I will solve them, just a sec
14:59:24 <LeeF> zakim, code?
14:59:24 <Zakim> the conference code is 77277 (tel:+1.617.761.6200 tel:+ tel:+44.203.318.0479), LeeF
14:59:29 <bglimm> Zakim, ??P4 is me
14:59:29 <Zakim> +bglimm; got it
14:59:31 <Zakim> + +1.310.729.aabb
14:59:38 <kasei> Zakim, aabb is me
14:59:38 <Zakim> +kasei; got it
14:59:50 <AxelPolleres> AxelPolleres has joined #sparql
14:59:50 <Zakim> +Sandro
14:59:54 <SteveH> Zakim, SteveH_ is me
14:59:54 <Zakim> +SteveH; got it
15:00:05 <Zakim> + +1.617.553.aacc
15:00:08 <LeeF> zakim, aacc is me
15:00:08 <Zakim> +LeeF; got it
15:00:21 <bglimm> Zakim, mute me
15:00:21 <Zakim> bglimm should now be muted
15:00:22 <bijan> Having some trouble dialing in
15:00:24 <LeeF> zakim, who's on the phone?
15:00:24 <Zakim> On the phone I see SteveH, cbuilara, bglimm (muted), kasei, Sandro, LeeF
15:00:27 <Zakim> +AxelPolleres
15:00:32 <LeeF> bijan, which # are you trying?
15:00:35 <chimezie> chimezie has joined #sparql
15:00:39 <bijan> It's on my end :)
15:00:41 <bijan> But cambridge
15:00:43 <AndyS> zakim is not getting to the second part of the message.
15:00:49 <chimezie> Zakim, what is the pass code?
15:00:51 <Zakim> the conference code is 77277 (tel:+1.617.761.6200 tel:+ tel:+44.203.318.0479), chimezie
15:00:59 <bglimm> Bijan, UK is mostly not working, I usually try France
15:01:03 <bijan> Workign on it
15:01:05 <bglimm> +
15:01:06 <bijan> I'm in the states
15:01:10 <SteveH> I got in on the london number ok
15:01:11 <bglimm> Ah, ok
15:01:14 <bijan> Hence not with my normal setup
15:01:20 <Zakim> +??P12
15:01:26 <AndyS> zakim, ??P12 is me
15:01:26 <Zakim> +AndyS; got it
15:01:32 <Zakim> +Chimezie_Ogbuji
15:01:36 <AxelPolleres> Lee, as mentioned, I can srcibe (alex regrets...)
15:01:39 <AxelPolleres> ok?
15:01:41 <LeeF> zakim, who's speaking?
15:01:50 <LeeF> AxelPolleres, yes, thanks - i already set you up as scribe with zakim/rrsagent
15:01:52 <Zakim> LeeF, listening for 10 seconds I heard sound from the following: AxelPolleres (11%), LeeF (31%), AndyS (57%), Chimezie_Ogbuji (5%)
15:01:54 <AxelPolleres> scribe: AxelPolleres
15:02:06 <Zakim> +??P14
15:02:12 <chimezie> Zakim, mute me
15:02:12 <Zakim> Chimezie_Ogbuji should now be muted
15:02:22 <MattPerry> zakim, P14 is me
15:02:22 <Zakim> sorry, MattPerry, I do not recognize a party named 'P14'
15:02:27 <LeeF> zakim, ??P14 is MattPerry
15:02:27 <Zakim> +MattPerry; got it
15:02:51 <LeeF> zakim, who's on the phone?
15:02:51 <Zakim> On the phone I see SteveH, cbuilara, bglimm (muted), kasei, Sandro, LeeF, AxelPolleres, AndyS, Chimezie_Ogbuji (muted), MattPerry
15:02:56 <Zakim> +??P15
15:03:01 <bijan> zakim, ??p15 is me
15:03:01 <Zakim> +bijan; got it
15:03:07 <bijan> zakim, mute me
15:03:07 <Zakim> bijan should now be muted
15:03:13 <EFranconi> EFranconi has joined #sparql
15:03:27 <AxelPolleres> topic: admin
15:03:28 <LeeF> topic: Admin
15:03:33 <LeeF> PROPOSED: Approve minutes at
15:03:39 <LeeF> PROPOSED: Approve minutes at
15:03:46 <AxelPolleres> Lee: let's approve old minutes...
15:03:53 <Zakim> + +1.540.841.aadd
15:04:24 <AndyS> +1 to minutes
15:04:28 <LeeF> zakim, aadd is pgearon
15:04:28 <Zakim> +pgearon; got it
15:04:38 <LeeF> RESOLVED: Approve minutes at
15:04:42 <pgearon> pgearon has joined #sparql
15:04:43 <LeeF> resolved: Approve minutes at
15:04:48 <LeeF> RESOLVED: Approve minutes at
15:04:58 <LeeF> Next regular meeting: 2011-01-05 @ 15:00 UK / 10:00 EST (scribe: Olivier? AndyS?) 
15:05:19 <SteveH> Jan 4th?
15:05:38 <AxelPolleres> Leef: next meeting will be Jan 4th ,no meeting next week, enjoy your holidays
15:05:44 <LeeF> Next regular meeting: 2011-01-04 @ 15:00 UK / 10:00 EST (scribe: Olivier? AndyS?) 
15:05:52 <LeeF> topic: Last Call
15:06:22 <LeeF>
15:06:27 <AxelPolleres> Lee: last call will be "standing topic" for the following TCs
15:06:59 <Zakim> -SteveH
15:07:00 <AxelPolleres> .... wiki page has for each doc WG decisions needed and open issues,
15:07:22 <AxelPolleres> ... as well as editorial tasks
15:07:34 <Zakim> +SteveH
15:07:53 <AxelPolleres> ... please all editors, complete that wiki page, would be extremly helpful
15:08:04 <Zakim> +??P2
15:08:13 <AxelPolleres> ... also volunteering for review can be done on that wiki page
15:08:18 <Zakim> -SteveH
15:08:50 <AxelPolleres> ... all editors, if you believe your doc is ready for LC, record that as well
15:09:01 <Zakim> +SteveH
15:09:36 <AndyS> No test doc?
15:09:58 <AxelPolleres> ... all WG members, feel free to add yourselves volunteering for a comprehensive review
15:10:30 <AxelPolleres> Lee: testdoc not yet critical for LC, let's focus on that as we have those ready
15:10:50 <LeeF> zakim, who's on the phone?
15:10:50 <Zakim> On the phone I see cbuilara, bglimm (muted), kasei, Sandro, LeeF, AxelPolleres, AndyS, Chimezie_Ogbuji (muted), MattPerry, bijan (muted), pgearon, ??P2, SteveH
15:10:52 <AxelPolleres> s/those/those nine docs/
15:10:57 <LeeF> zakim, ??P2 is EFranconi
15:10:57 <Zakim> +EFranconi; got it
15:11:30 <AndyS> I'm a bit worried about having too few tests (not the test doc) - tests test the spec and check we have done the details.  Already finding small issues with functions.
15:11:53 <AxelPolleres> Lee: two main topics today... enrico's proposal for new OWL entailment regime, and open issues on SPARQL uniform HTTP Protocol, we'll try to limit each one to 25min.
15:12:00 <SteveH> +1 to AndyS 
15:12:37 <AxelPolleres> ... would like to focus on conrete querstions rather than rehashing discussions carried out on the list already.
15:12:45 <AndyS> Topic: OWL Direct entailment regime with non-distinguished variables
15:12:59 <Zakim> -SteveH
15:13:14 <AxelPolleres> enrico: propose to add entailment regime that differs from current OWL DS proposal.
15:13:27 <Zakim> +??P22
15:13:29 <LeeF> (AndyS, I share the concern. Not sure how to clone our WG resources. :/)
15:13:33 <SteveH> Zakim, ??P22 is me
15:13:33 <Zakim> +SteveH; got it
15:14:32 <AxelPolleres> .... would change that bnodes would not be in the answer set (birte said that this is not maintaining lower entailment regimes results), and allow non-distinguished variables.
15:15:03 <AxelPolleres> ... negative remark I got was that two deviating ent regimes might be confusing (bijan).
15:15:51 <AxelPolleres> ... why I still want to have it: is that this would fit the needs of people doing 20 years of DB+ontologies research, and allow a richer way to query ontologies. 
15:16:16 <AxelPolleres> ... without this extension SPARQL becomes useless for OWL-QL.
15:16:50 <AxelPolleres> ... for applications that don't necesarily all exist now, but I see huge potential for DB+ontologies.
15:16:59 <bijan> zakim, unmute me
15:16:59 <Zakim> bijan should no longer be muted
15:17:09 <bglimm> I cant really say much, my cold made me loose my voice
15:17:12 <AxelPolleres> Lee: bijan/birte wanna express your concerns? 
15:17:30 <AxelPolleres> bijan: not just confusing to have more than once, but other concerns.
15:18:34 <AxelPolleres> ... we have already confusion with OWL profiles explaining our users. We try to make that work for our users to query SW data, DB scenario is also important
15:18:48 <bijan> zakim, mute me
15:18:48 <Zakim> bijan should now be muted
15:18:51 <AxelPolleres> ... but we don't want to deviate on RDF data from other ent regimes.
15:19:03 <AxelPolleres> q+
15:19:08 <LeeF> ack AxelPolleres
15:19:33 <LeeF> AxelPolleres: We should discuss the two points separately - bnodes in answers, and non-distinguished variables
15:19:38 <bijan> Let me add that non-distinguished variables complicate the implementation even in the polynomial fragments.
15:20:08 <EFranconi> q+
15:20:18 <sandro> q+ to ask about who will do the writing and produce test cases and when
15:20:20 <LeeF> ack EFranconi
15:20:56 <AxelPolleres> Axel: there might be cases for bnodes allowed, but also non-distinguished variables.
15:21:29 <AxelPolleres> Enrico: disagrees... if nondist-variables, we shouldn't have bnodes in answers, both issues go together.
15:21:29 <LeeF> ack sandro
15:21:29 <Zakim> sandro, you wanted to ask about who will do the writing and produce test cases and when
15:21:55 <AxelPolleres> sandro: does enrico volunteer to do the editing work?
15:22:22 <AxelPolleres> enrico: can be done, don't know about the details, that's a minor point.
15:22:38 <LeeF> q?
15:22:40 <AxelPolleres> lee: given our existing schedule and resources, that's a challenge
15:22:46 <bglimm> Well, there have to be implementations and test cases, which is more work then actually just adding another regime to the spec
15:22:58 <SteveH> doesn't the point about returning "existential variables" apply equally to vanilla SPARQL?
15:23:18 <AxelPolleres> ... another point is whether we would've implementations, enrico put a few implementers in touch with us on the list.
15:24:01 <AxelPolleres> ... one potential option would be to add it under the "at risk" label.
15:24:21 <bijan> Two independent implementations
15:24:28 <AxelPolleres> enrico: we'd need an implementation which complies to the standard, right?
15:24:36 <AxelPolleres> lee: we need two
15:24:58 <AxelPolleres> enrico: quonto, a system from pisa, clark&parsia, ...
15:25:30 <AxelPolleres> ... that to nd-variables, none of them planned to be compliant with sparql (mostly because there was no such entailment regime)
15:26:12 <AxelPolleres> ... wrappers around existing systems should be possible/opportunity
15:26:54 <AxelPolleres> ... but probably not gonna happen within the next 4 months.
15:27:14 <EFranconi> q+
15:27:20 <LeeF> ack EFranconi
15:27:34 <AxelPolleres> lee: ??? (didn't catch that)
15:28:03 <AxelPolleres> enrico: market hasn't pushed so far, since they weren't aware
15:28:35 <AxelPolleres> lee: then not clerar to me whether they have a reason to move to sparql or whether happy with current conjunctive queries they have
15:28:38 <LeeF> q?
15:28:50 <bijan> q+
15:28:53 <SteveH> q+
15:28:54 <bijan> zakim, unmute me
15:28:54 <Zakim> bijan should no longer be muted
15:28:56 <LeeF> ack bijan
15:28:56 <AxelPolleres> enrico: just a matter of (standard?) syntax
15:29:43 <AxelPolleres> bijan: one thing that came out is that a lot of use case for that is analysis rather than end user queries
15:30:38 <bijan> zakim, mute me
15:30:38 <Zakim> bijan should now be muted
15:30:38 <LeeF> q?
15:30:40 <LeeF> ack SteveH
15:31:43 <AxelPolleres> steve: there are already situations where implementations can't handle all of the syntax of SPARQL (quad stores vs. triple stores, e.g.) 
15:32:09 <EFranconi> Q+
15:32:20 <AxelPolleres> sandro: wonder if the market will end up misusing the existing SPARQL/OWL ent regime, if we don't cater for them
15:32:22 <LeeF> ack EFranconi
15:33:05 <AxelPolleres> enrico: I think this will happen, some implementers already say "we want SPARQL syntax/protocol"
15:33:45 <Zakim> -kasei
15:34:05 <Zakim> +kasei
15:34:08 <AndyS> misusing? experimenting with future possibilities?
15:34:21 <SteveH> returning an error under certain circumstances would be complaint
15:34:22 <AxelPolleres> sandro: you say that this group would make a parallel standard, but not "claim" they are compliant to the spec.
15:34:23 <bijan> I.e.,entailment regimes are an extension point
15:34:30 <LeeF> q?
15:34:49 <AxelPolleres> lee: I see three options.
15:36:10 <AxelPolleres> .... 1) we don't have enough consensus to cater for this in this round, we hope that this group of implementers will go ahead and specify their ent regime and can be added in next standardisation role (clearly the least satisfying to enrico, and could slow adoption in some market, but leat impact to our schedule)
15:36:17 <AxelPolleres> 2) include the regime
15:36:30 <AxelPolleres> 3) include the regime "at risk"
15:36:50 <AxelPolleres> both 2)+3) would require text and test cases
15:37:15 <AxelPolleres> .... would like to go for strawpoll.
15:37:27 <bijan> +1 to sandro; at risk is the default to help the schedule
15:37:29 <AxelPolleres> sandro: I would want to make it at risk anyways, if we do it
15:37:37 <AxelPolleres> lee: ok let's drop 2)
15:37:44 <bijan> Those seem like the options :)
15:37:54 <AxelPolleres> ... any more comments?
15:38:37 <AndyS> Would current ent regime for OWL2-DS name need to change?
15:38:50 <AxelPolleres> enrico: in sparql1.0 it was anyways legal to add new ent. regimes
15:38:56 <bijan> AndyS: No
15:38:57 <AxelPolleres> lee: this is still the case
15:39:05 <bglimm> We don't say that the specified regimes are the only ones forever after
15:39:10 <bijan> +1 to Lee; Entailment regime are an extension point
15:39:40 <AxelPolleres> sandro: we're definitly open to more ent. regimes in the future
15:39:55 <AxelPolleres> (enrico can you type in what you just said)
15:40:01 <AndyS> Leaving a postponed issue is the way to do that
15:40:32 <AxelPolleres> lee: strawpoll...
15:40:40 <chimezie> Yes, if we choose 1, we probably should mark it as a postponed issue 
15:40:45 <LeeF> straw poll: (1) leave regimes as-is / (2) include regime with non-distinguished variables
15:41:07 <SteveH> 1, with a postponed issue
15:41:09 <bglimm> 1
15:41:10 <kasei> 0
15:41:11 <cbuilara> 0
15:41:13 <bijan> 1
15:41:14 <LeeF> 0
15:41:15 <sandro> 2, but not a strong preference
15:41:17 <AndyS> 0
15:41:17 <chimezie> 1, with postponed issue
15:41:18 <EFranconi> 2
15:41:19 <MattPerry> 0
15:41:25 <pgearon> 0
15:41:33 <AxelPolleres> 2 only if we have someone committing to edit/spec it, 1 otherwise
15:42:13 <AxelPolleres> (count me as a 0)
15:42:53 <AxelPolleres> lee: I'll set up an issue for this...
15:43:19 <LeeF> PROPOSED: Close ISSUE-64 with no change for now, and include ISSUE-64 on a list of postponed issues for a future WG to consider
15:44:06 <Zakim> -bglimm
15:45:07 <chimezie> Zakim, unmute me
15:45:07 <Zakim> Chimezie_Ogbuji should no longer be muted
15:45:08 <bijan> Seconded
15:45:53 <AxelPolleres> seconded, so
15:45:54 <Zakim> +??P1
15:45:57 <chimezie> Zakim, mute me
15:45:57 <Zakim> Chimezie_Ogbuji should now be muted
15:46:07 <bglimm> Zakim, ??P1 is me
15:46:07 <Zakim> +bglimm; got it
15:46:11 <LeeF> RESOLVED: Close ISSUE-64 with no change for now, and include ISSUE-64 on a list of postponed issues for a future WG to consider
15:46:19 <bglimm> Sorry, my line dropped
15:46:50 <LeeF>  ISSUE-64: Closed with no action as of now, but postponed for consideration by a future WG
15:47:32 <LeeF> ACTION: Lee to close ISSUE-64 and place ISSUE-64 on a list of postponed issues
15:48:07 <chimezie> Zakim, unmute me
15:48:07 <Zakim> Chimezie_Ogbuji should no longer be muted
15:48:28 <AxelPolleres> lee: trackbot gone, need to add issues/actions manually after call
15:48:37 <LeeF>
15:48:44 <AxelPolleres> topic: http udate protocol issues
15:49:00 <AxelPolleres> lee: first one ISSUE-56
15:49:22 <Zakim> -bglimm
15:49:42 <AxelPolleres> .... question whether somthing about PATCH needs to be said in http-rdf-update 
15:50:32 <AxelPolleres> chime: we had 2 comments about this, as a result about that thread, my suggestion was to keep PATCH not normative but add some more words about it
15:50:57 <chimezie>
15:51:17 <AxelPolleres> ... most recent comment from andy, see url just pasted.
15:51:27 <LeeF>
15:51:36 <AxelPolleres> lee: current version reflects your proposal?
15:51:40 <AxelPolleres> chime: yes
15:52:00 <Zakim> -bijan
15:52:46 <AxelPolleres> ... I am happy with that informative note.
15:52:53 <AxelPolleres> ... anyone objects?
15:53:36 <LeeF> PROPOSED: Close ISSUE-56 with the inclusion of non-normative only text addressing how PATCH should be used with the Uniform HTTP Protocol
15:53:52 <LeeF> PROPOSED: Close ISSUE-56 with the inclusion of the existing non-normative only text addressing how PATCH should be used with the Uniform HTTP Protocol
15:54:17 <LeeF> RESOLVED: Close ISSUE-56 with the inclusion of the existing non-normative only text addressing how PATCH should be used with the Uniform HTTP Protocol
15:54:23 <LeeF> ACTION: Lee to close ISSUE-56
15:54:58 <LeeF>
15:55:07 <AxelPolleres> Lee: next issue on the agenda is  the one around the OPTION method 
15:56:31 <AxelPolleres> chime: ... if you send an OPTION or GET to a service you get the SD document.
15:56:57 <AxelPolleres> lee: that matches greg's understanding?
15:57:10 <AxelPolleres> greg: need to look that up, but sounds fine to me
15:57:29 <AxelPolleres> lee: looks like what we had decided.
15:58:01 <SteveH> can we discuss the xml:base issue?
15:58:08 <AxelPolleres> ... any alternatives to suggest? If not, chime please take this as consensus to the group.
15:58:20 <LeeF>
15:58:24 <Zakim> -EFranconi
15:59:02 <Zakim> -MattPerry
16:00:54 <AxelPolleres> chime: summarising issue about BASE URI
16:01:26 <SteveH> that captured my viewpoint
16:01:39 <AxelPolleres> Lee: is there any authority we can ask about that interpretation of XML base?
16:01:46 <AxelPolleres> sandro: can try to find out
16:02:28 <AxelPolleres> lee: steve/andy, if we can confirm that this is a valid interpretation of BASE, would your concerns still stand?
16:02:55 <AxelPolleres> andy: still some concerns, leaves too much open(?)
16:03:53 <AxelPolleres> steve: ... (similar concerns?) ... in 4store we banned relative URIs 
16:03:56 <pgearon> +q
16:04:22 <LeeF> ack pgearon
16:04:40 <AxelPolleres> just one thing for the records: new comment also affects http-update-protocol
16:05:04 <SteveH> note this only applies to ?graph=[URI]
16:05:10 <SteveH> in the HTTP update protocol
16:05:43 <AxelPolleres> paul: I am working with RIF and relative URIs at the moment, similar issues. I would agree with banning them here
16:06:11 <AxelPolleres> chime: handled properly doesn't leave much ambiguity
16:06:18 <AxelPolleres> sandro: (disagrees)
16:06:25 <AxelPolleres> s/sandro/steve/
16:06:35 <AndyS>  Example: whether it's a N-triple doc or an RDF/XML doc for the graph makes a difference but the same RDF data.
16:08:57 <AndyS> I believe XML base says that it's only in-content: 
16:09:44 <SteveH> chimezie, what would you do with <doc><el xml:base="A"></el><el xml:base="B"></el></doc>?
16:10:06 <AxelPolleres> lee: sandro would you be willing to pursue on that...
16:10:23 <AxelPolleres> (sandro and chime to figure out details after call)
16:10:36 <SteveH> bye all
16:10:38 <Zakim> -LeeF
16:10:39 <Zakim> -SteveH
16:10:42 <Zakim> -pgearon
16:10:56 <LeeF> ACTION: sandro to pursue answer to whether the base of a URI can come from inside the request's content (e.g. xml:base)
16:11:08 <AxelPolleres> zakim, attendees?
16:11:08 <Zakim> I don't understand your question, AxelPolleres.
16:11:10 <Zakim> -kasei
16:11:12 <AxelPolleres> rrsagent, make records public