Chatlog 2011-02-08

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.

14:59:21 <RRSAgent> RRSAgent has joined #sparql
14:59:21 <RRSAgent> logging to
14:59:23 <trackbot> RRSAgent, make logs world
14:59:23 <Zakim> Zakim has joined #sparql
14:59:25 <trackbot> Zakim, this will be 77277
14:59:25 <Zakim> ok, trackbot; I see SW_(SPARQL)10:00AM scheduled to start in 1 minute
14:59:26 <trackbot> Meeting: SPARQL Working Group Teleconference
14:59:26 <trackbot> Date: 08 February 2011
14:59:32 <SteveH_> SteveH_ has joined #sparql
14:59:33 <LeeF> zakim, this will be SPARQL
14:59:33 <Zakim> ok, LeeF, I see SW_(SPARQL)10:00AM already started
14:59:35 <AndyS> zakim, who is on the phone?
14:59:35 <Zakim> On the phone I see ??P4, +1.603.897.aaaa
14:59:44 <AndyS> zakim, ??P4 is me
14:59:44 <Zakim> +AndyS; got it
14:59:52 <MattPerry> zakim, aaaa is me
14:59:52 <Zakim> +MattPerry; got it
14:59:54 <Zakim> + +1.310.729.aabb
15:00:00 <Zakim> + +44.186.528.aacc
15:00:01 <kasei> Zakim, aabb is me
15:00:04 <Zakim> +kasei; got it
15:00:06 <Zakim> +Lee_Feigenbaum
15:00:11 <bglimm> Zakim, +44.186.528.aacc is me
15:00:14 <Zakim> +bglimm; got it
15:00:29 <bglimm> Zakim, mute me
15:00:29 <Zakim> bglimm should now be muted
15:00:39 <cbuilara> cbuilara has joined #sparql
15:00:39 <Zakim> + +44.208.439.aadd
15:00:48 <SteveH_> Zakim, aadd is [Garlik]
15:00:48 <Zakim> +[Garlik]; got it
15:00:48 <LeeF> Chair: LeeF
15:00:53 <LeeF> Scribenick: MattPerry
15:00:54 <Zakim> +[IPcaller]
15:00:55 <SteveH_> Zakim, [Garlik] is temporarily me
15:00:55 <Zakim> +SteveH_; got it
15:01:04 <cbuilara> zakim, IPcaller is me
15:01:04 <Zakim> +cbuilara; got it
15:01:30 <Zakim> +pgearon
15:02:10 <LeeF> zakim, who's on the phone?
15:02:10 <Zakim> On the phone I see AndyS, MattPerry, kasei, bglimm (muted), Lee_Feigenbaum, SteveH_, cbuilara, pgearon
15:02:23 <LeeF> Agenda:
15:02:34 <LeeF> Regrets: Chimezie, Axel, Olivier
15:03:02 <alepas> alepas has joined #sparql
15:03:14 <Zakim> +??P22
15:03:32 <NickH> zakim, ??P22 is me
15:03:32 <Zakim> +NickH; got it
15:03:43 <LeeF> topic: Admin
15:03:49 <Zakim> + +3539149aaee
15:03:52 <LeeF> PROPOSED: Approve minutes at
15:03:52 <LeeF> PROPOSED: Approve minutes at 
15:03:54 <AlexPassant> Zakim, +3539149aaee is me
15:03:55 <Zakim> +AlexPassant; got it
15:04:55 <LeeF> RESOLVED: Approve minutes at
15:05:01 <LeeF> RESOLVED: Approve minutes at 
15:05:14 <LeeF> Next regular meeting: 2011-02-15 @ 15:00 UK / 10:00 EST (scribe: SteveH) 
15:05:53 <LeeF>
15:05:56 <MattPerry> Topic: comments
15:06:22 <MattPerry> LeeF: mainly administrative tasks today
15:07:29 <MattPerry> LeeF: Chime and Lee to work with on dataset protocol comments
15:07:50 <Zakim> +Sandro
15:08:01 <kasei> q+
15:08:06 <LeeF> ack kasei
15:08:06 <MattPerry> LeeF: does anyone not want to proceede with dataset protocol specification?
15:08:25 <AndyS> I think we *should* proceed with that doc.
15:08:55 <MattPerry> kasei: I'm not opposed to publishing it, but there are some complexity issues
15:09:17 <MattPerry> LeeF: reviewers are Andy, Steve and Nick
15:09:48 <NickH> I am going to get my comments to the list today
15:09:50 <MattPerry> LeeF: more targeted comments are helpful
15:10:22 <NickH> I support the publishing of the SPARQL 1.1 RDF Dataset HTTP Protocol
15:11:08 <MattPerry> LeeF: Jeen Brokestra posted feedback on implementing SPARQL 1.1
15:11:09 <kasei> Jeen's email did reach the comments list
15:11:28 <Zakim> -pgearon
15:12:05 <Zakim> +pgearon
15:12:12 <MattPerry> LeeF: Andy, Steve looked at post?
15:12:32 <MattPerry> AndyS: did look at it briefly ... comments are pretty high level
15:12:50 <MattPerry> AndyS: may want to ask for more specific comments
15:12:59 <kasei> I think giving Jeen a concrete example of where MINUS and NOT EXISTS give different answers would help.
15:13:16 <AndyS>
15:13:23 <MattPerry> SteveH: I look more closely at the aggregates comments
15:14:02 <MattPerry> SteveH: I [will] look more closely
15:14:17 <LeeF>
15:14:36 <MattPerry> LeeF: has Greg been able to look at this comment
15:15:03 <MattPerry> kasei: did look at it will depend on interaction between dataset protocol and service description
15:16:04 <AndyS> Jeen has already commented on MINUS in JB-2  That part is a dup and Axel is down to reply.
15:16:41 <MattPerry> LeeF: that's all the unassigned comments
15:16:57 <MattPerry> LeeF: we have a few comments that have been recently addressed
15:17:03 <AndyS> q+
15:17:29 <LeeF> ack AndyS
15:18:10 <MattPerry> AndyS: Jeen has comments already that cover most of his post
15:18:31 <MattPerry> AndyS: Axel has the Minus comment
15:18:49 <LeeF> topic: To Last Call
15:19:06 <LeeF>
15:19:30 <MattPerry> LeeF: Query has had some progress
15:19:54 <kasei> I'm editing to add new blocking issue on SD around the sd:url discussion.
15:20:09 <MattPerry> AndyS: need some text in each document that points to the overview document or the list of documents
15:21:02 <LeeF>
15:21:27 <LeeF>
15:21:30 <MattPerry> LeeF: Greg put this page together that lists new URIs defined ... these need to go into the namespaces document
15:21:57 <MattPerry> LeeF: Sandro can you do it?
15:22:02 <MattPerry> Sandro: Yes I can
15:22:22 <LeeF> ACTION: Sandro to update sparql namespace document with URIs for existing & new functions and aggregates as per
15:22:22 <trackbot> Created ACTION-389 - Update sparql namespace document with URIs for existing & new functions and aggregates as per [on Sandro Hawke - due 2011-02-15].
15:23:01 <MattPerry> LeeF: there are still some minor editorial work in query
15:23:12 <MattPerry> AndyS: Property Path issues still need to be sorted out
15:23:37 <MattPerry> AndyS: I will propose a change to reflect Matt's comment on node marking vs edge marking
15:23:57 <MattPerry> AndyS: has to do with test results
15:24:19 <MattPerry> LeeF: do we need a working group decision
15:24:23 <MattPerry> AndyS: probably
15:24:46 <MattPerry> AndyS: could start reviewing the stable parts of the doc now
15:25:15 <MattPerry> SteveH: Andy is right ... the doc is very large ... what sections?
15:25:40 <MattPerry> AndyS: Section in formal algebra is only one affected by Property Paths issue
15:25:59 <MattPerry> LeeF: can we enumerate the parts to be avoided?
15:26:16 <MattPerry> AndyS: review every section except 18
15:26:53 <Zakim> -AndyS
15:26:54 <MattPerry> SteveH: will change non-alebra part of aggregates
15:27:14 <MattPerry> LeeF: This is section 11
15:27:18 <Zakim> +??P1
15:27:24 <MattPerry> s/alebra/algebra
15:27:30 <AndyS> zakim, ??P1 is me
15:27:30 <Zakim> +AndyS; got it
15:27:34 <AndyS> ArbitraryLengthPath and ZeroLengthPath are areas of possible change 
15:27:44 <AndyS> (defn and eval)
15:27:52 <MattPerry> LeeF: will leave it up to reviewers to decide if they want to review it now
15:28:24 <MattPerry> LeeF: see one editorial change in Update and some comments around formal model
15:28:25 <LeeF> zakim, who's on the call?
15:28:29 <Zakim> On the phone I see MattPerry, kasei, bglimm (muted), Lee_Feigenbaum, SteveH_, cbuilara, NickH, AlexPassant, Sandro, pgearon, AndyS
15:29:04 <MattPerry> LeeF: what is the status of the formal model?
15:29:18 <MattPerry> AlexPassant: We are fixing the model now based on comments
15:29:44 <MattPerry> AlexPassant: Hoping to have a draft by the end of the week
15:30:33 <MattPerry> LeeF: will sequence of operations bit be in the draft?
15:30:34 <LeeF>
15:30:38 <MattPerry> AlexPassant: yes
15:31:10 <MattPerry> LeeF: Progress on Protocol?
15:31:47 <MattPerry> LeeF: Service Description is basically finished 
15:32:14 <MattPerry> LeeF: any issues in Service Description that need working group time?
15:33:00 <MattPerry> kasei: issue of wether or not to add dataset protocol stuff to it
15:33:40 <kasei> I should say that the one issue I added to the last call page for SD doesn't depend on the dataset protocol issue (I don't think)
15:34:04 <MattPerry> LeeF: some discussion with Kendall about Entailment review
15:34:33 <Zakim> + +539149aaff
15:34:41 <AlexPassant> Zakim, +539149aaff is me
15:34:41 <Zakim> +AlexPassant; got it
15:34:52 <bglimm> I didn't yet succeed to convince somebody from our group to do another review
15:34:56 <MattPerry> MattPerry: have some comments from Zhe Wu that I will send out
15:36:10 <MattPerry> topic: Test Status
15:36:39 <MattPerry> LeeF: I have reviewed select expression tests
15:37:51 <LeeF> topic: Action round-up
15:38:01 <LeeF>
15:38:30 <Zakim> -AlexPassant
15:38:52 <MattPerry> LeeF: will go through actions quickly ... stop me if we need to discuss anything
15:39:32 <LeeF> trackbot, close ACTION-291
15:39:33 <trackbot> ACTION-291 Work MIME type registration information into SPARQL Update document based on Sandro's suggestion for formatting etc. closed
15:41:35 <MattPerry> LeeF: Text needed in member submission for update to say it is not the standard
15:41:38 <LeeF>  ACTION-296: Take a look at OWL 1.1 documents for example text
15:41:38 <trackbot> ACTION-296 to suggest a wording for a disclaimer text for the SPARQL update member submission notes added
15:42:42 <LeeF> trackbot, close ACTION-345
15:42:42 <trackbot> ACTION-345 Check example in closed
15:43:20 <LeeF> trackbot, close ACTION-354
15:43:20 <trackbot> ACTION-354 Wiki-to-last-call doc regarding update closed
15:43:37 <kasei> that's done
15:43:43 <kasei> 356
15:43:58 <kasei> I've completed ACTION-370 on function tests.
15:44:39 <MattPerry> My property path action is pending resolution of the property path semantics issue
15:44:54 <MattPerry> kasei: I have looked at tests and agree with them
15:45:02 <LeeF> trackbot, close ACTION-370
15:45:02 <trackbot> ACTION-370 Look at function tests closed
15:45:22 <LeeF> trackbot, close ACTION-374
15:45:22 <trackbot> ACTION-374 Look at projected expression tests closed
15:45:29 <LeeF> trackbot, close ACTION-375
15:45:29 <trackbot> ACTION-375 Look at CONSTRUCT WHERE tests closed
15:46:39 <LeeF> topic: Issues round-up
15:46:41 <LeeF>
15:46:48 <Zakim> +bglimm.a
15:46:56 <MattPerry> LeeF: Security issues in update
15:46:59 <bglimm> Zakim, +member:bglimm.a is me
15:46:59 <Zakim> sorry, bglimm, I do not recognize a party named '+member:bglimm.a'
15:47:07 <LeeF>
15:47:19 <bglimm> Zakim, member:bglimm.a is  me
15:47:19 <Zakim> sorry, bglimm, I do not recognize a party named 'member:bglimm.a'
15:47:27 <bglimm> Zakim, a is me
15:47:27 <Zakim> sorry, bglimm, I do not recognize a party named 'a'
15:47:29 <MattPerry> LeeF: should leave issue open until doc has been reviewed
15:48:06 <MattPerry> LeeF: Issue 52: availability of unnamed graph in service description
15:48:24 <bglimm> Zakim, mute me
15:48:24 <Zakim> bglimm was already muted, bglimm
15:48:29 <MattPerry> kasei: I don't think this is really an issue
15:48:41 <LeeF> PROPOSED: Close ISSUE-52 with no change
15:48:42 <MattPerry> LeeF: I agree ... any concerns with closing issue
15:48:45 <LeeF>
15:48:53 <LeeF> PROPOSED: Close ISSUE-52 ( with no change
15:49:08 <kasei> +1
15:49:19 <LeeF> RESOLVED: Close ISSUE-52 ( with no change
15:49:26 <Zakim> -bglimm
15:49:26 <LeeF> trackbot, close ISSUE-52
15:49:26 <trackbot> ISSUE-52 Do we need the availability of an unnamed graph in SD? closed
15:49:59 <AndyS> q+
15:50:08 <LeeF> ack AndyS
15:50:10 <MattPerry> LeeF: Issue 65 needs some discussion
15:50:17 <MattPerry> AndyS: can we close Issue 58
15:50:52 <LeeF>  ISSUE-58: Update document has registration information; registration process being tracked via actions
15:50:52 <trackbot> ISSUE-58 Register a MIME type for SPARQL Update requests? notes added
15:50:57 <LeeF> trackbot, close ISSUE-58
15:50:57 <trackbot> ISSUE-58 Register a MIME type for SPARQL Update requests? closed
15:51:03 <AndyS> BTW - wasn't the extension going to be ".ru" - for RDf update c.f .rq redf query.  Says ".ud"
15:51:23 <LeeF> Next week: talk about ISSUE-65 during teleconference
15:51:53 <MattPerry> pgearon: I can change the extension in the doc
15:52:52 <MattPerry> LeeF: valuable to for someone to summarize discussion of service vs graph store vs ... on the mailing list
15:53:34 <MattPerry> AndyS: one of the issues is naming
15:54:15 <LeeF> AndyS: RDF Dataset HTTP Protocol says that doing a GET on a graph store URI returns the service description
15:55:26 <LeeF>
15:55:29 <MattPerry> LeeF: is this Section 5.8 in the doc?
15:55:47 <MattPerry> AndyS: yes
15:56:29 <MattPerry> LeeF: a GET on the endpoing gets the service description
15:56:48 <MattPerry> LeeF: there is confusion about whether or not that is the graph store URI
15:57:27 <kasei> this section 5.8 is a surprise to me.
15:57:44 <kasei> "Per section 2 of SPARQL 1.1 Service Descriptions, ..." doesn't seem relevant here.
15:57:52 <MattPerry> LeeF: is there somewhere that talks about doing a GET on a graph store URI
15:58:04 <MattPerry> LeeF: do we talk about graph store URI vs service URI
15:58:34 <MattPerry> AndyS: will send some more email
15:59:22 <MattPerry> kasei: depends on which "service" you are talking about
15:59:53 <LeeF>
16:00:03 <LeeF>  GET /rdf-graphs/employees?graph=http%3A// HTTP/1.1
16:00:47 <MattPerry> LeeF:  /rdf-graphs/employees?graph is ambiguous
16:01:02 <LeeF>   GET /rdf-graphs/services
16:01:14 <LeeF> """
16:01:15 <LeeF> identifies the HTTP service that implements this protocol. In order to dispatch requests to manage named or default graphs by embedding them in the query component of the service URL, the service URL will need to be known a priori. As will be discussed later in this document, both HTTP OPTIONS and GET requests can be sent to the service and the the 
16:01:15 <LeeF> response to such a request is a representation of a service description
16:01:16 <LeeF> """
16:02:42 <Zakim> -Lee_Feigenbaum
16:02:45 <NickH> Thanks LeeF!
16:02:47 <Zakim> -SteveH_
16:02:49 <Zakim> -AlexPassant
16:02:51 <Zakim> -bglimm
16:02:53 <Zakim> -MattPerry
16:02:57 <Zakim> -cbuilara
16:02:59 <Zakim> -kasei
16:03:01 <Zakim> -AndyS
16:03:09 <Zakim> -pgearon
16:03:16 <Zakim> -NickH
16:03:19 <Zakim> -Sandro
16:03:21 <Zakim> SW_(SPARQL)10:00AM has ended
16:03:23 <Zakim> Attendees were +1.603.897.aaaa, AndyS, MattPerry, +1.310.729.aabb, kasei, Lee_Feigenbaum, bglimm, +44.208.439.aadd, SteveH_, cbuilara, pgearon, NickH, AlexPassant, Sandro
16:03:40 <LeeF> I see that in the POST section (5.4), the document talks about posting to the Graph Store URI
16:04:50 <LeeF> I don't see anything else in the document that talks about the URI of a Graph Store at all
16:17:37 <kasei> can anyone glance over a MINUS vs. NOT EXISTS example for me?
16:17:50 <kasei> wanted a simple example to send along to Jeen
16:22:01 <kasei> (think, but not entirely sure, that this is correct)
16:24:18 <LeeF> "a simple example" hah!
16:27:48 <kasei> well, simple-ish :)
16:27:59 <kasei> the simplest case I could come up with that shows a difference.
16:34:19 <SteveH> { ?x ?y ?z } MINUS { ?a ?b ?c } and { ?x ?y ?z } FILTER( NOT EXISTS { ?a ?b ?c } ) are different
16:34:39 <SteveH> it depends on what your expectation of the operation is, e.g. binding v's non binding
16:34:48 <kasei> true. this is specifically to address Jeen's claim that they are the same when they have shared variables, though.
16:34:54 <kasei> sorry, should have mentioned that.
16:35:09 <kasei> so, simplest case with shared variables.
16:35:43 <SteveH> ah, sure
16:35:43 <SteveH> mn, m memory not doing well :)
16:36:52 <SteveH> I think that's a relative simple example...
16:37:15 <SteveH> ...and FILTER NOT EXISTS still messes with my head
16:37:43 <kasei> heh
16:39:19 <SteveH> kasei, are you sure that ?n is in scope there?
16:39:36 <kasei> where?
16:39:57 <kasei> in the MINUS?
16:40:15 <SteveH> I'm not sure that it isn't FWIW
16:40:23 <kasei> no, and that's exactly the point.
16:40:30 <SteveH> ah, I see
16:40:54 <SteveH> ah, gotcha, I misread the data
17:43:10 <AlexPassant> AlexPassant has joined #sparql
17:59:53 <SteveH> SteveH has joined #sparql
18:05:31 <Zakim> Zakim has left #sparql
18:14:04 <AndyS> AndyS has joined #sparql
19:51:08 <AxelPolleres> AxelPolleres has joined #sparql
20:03:27 <AndyS> kasei, like the example - wonder if the filter/scope example is worth putting in the doc.
20:04:29 <AndyS> I think the important point is that it reflects different ways of thinking about negation - not one can do more or less than the other
20:25:22 <SteveH> +1
20:28:09 <AndyS> @@/not critical for LC (actually a tad difficult to avoid for LC :-(
20:49:07 <iv_an_ru> iv_an_ru has joined #sparql