Warning:
This wiki has been archived and is now read-only.
Chatlog 2012-05-22
From SPARQL Working Group
See original RRSAgent log and preview nicely formatted version.
Please justify/explain all edits to this page, in your "edit summary" text.
14:00:13 <LeeF> Present: SteveH, LeeF, kasei, Olivier, MattPerry, bglimm, Chimezie, AndyS, Sandro, Arthur
14:00:18 <trackbot> Meeting: SPARQL Working Group Teleconference
14:00:18 <trackbot> Date: 22 May 2012
14:00:27 <Zakim> ok, LeeF, I see SW_(SPARQL)10:00AM already started
14:00:28 <SteveH> Zakim, who's on the phone
14:00:28 <Zakim> I don't understand 'who's on the phone', SteveH
14:00:33 <SteveH> Zakim, who's on the phone?
14:00:33 <Zakim> On the phone I see ??P4
14:00:38 <SteveH> Zakim, ??P4 is me
14:00:38 <Zakim> +SteveH; got it
14:00:48 <MattPerry> MattPerry has joined #sparql
14:00:48 <Zakim> +LeeF
14:00:52 <Zakim> +kasei
14:01:02 <LeeF> Chair: LeeF
14:01:05 <Zakim> +Olivier
14:01:12 <LeeF> Regrets: pgearon, axel
14:01:28 <Zakim> +MattPerry
14:01:40 <Zakim> +??P16
14:01:46 <LeeF> Agenda: http://lists.w3.org/Archives/Public/public-rdf-dawg/2012AprJun/0142.html
14:01:53 <bglimm> Zakim, ??P16 is me
14:01:53 <Zakim> +bglimm; got it
14:02:11 <LeeF> scribenick: bglimm
14:02:15 <LeeF> zakim, who's here?
14:02:15 <Zakim> On the phone I see SteveH, LeeF, kasei, Olivier, MattPerry, bglimm
14:02:16 <Zakim> On IRC I see MattPerry, SteveH, Zakim, RRSAgent, Olivier, LeeF, AndyS, bglimm, pgearon, trackbot, iv_an_ru__, ya, NickH, kasei, ericP, sandro
14:02:55 <chimezie> chimezie has joined #sparql
14:03:10 <Zakim> +Yigal
14:03:55 <Zakim> +Chimezie
14:04:07 <LeeF> zakim, Yigal is AndyS
14:04:07 <Zakim> +AndyS; got it
14:04:15 <LeeF> zakim, who's on the phone?
14:04:15 <Zakim> On the phone I see SteveH, LeeF, kasei, Olivier, MattPerry, bglimm, AndyS, Chimezie
14:04:16 <AndyS> zakim, Yigal is temporarily me
14:04:17 <Zakim> sorry, AndyS, I do not recognize a party named 'Yigal'
14:04:28 <chimezie> Zakim, mute me
14:04:31 <Zakim> Chimezie should now be muted
14:04:35 <Arthur> Arthur has joined #sparql
14:04:47 <Zakim> +Sandro
14:05:15 <LeeF> topic: Admin
14:05:32 <LeeF> PROPOSED: Approve minutes from http://www.w3.org/2009/sparql/meeting/2012-05-15
14:06:02 <bglimm> Main topic last week: updated property path test cases
14:06:23 <Zakim> + +1.512.651.aaaa
14:06:25 <LeeF> close ACTION-619
14:06:25 <trackbot> ACTION-619 Approve PP test cases as per resolution closed
14:07:07 <AndyS> Austin, Texas
14:07:12 <Zakim> - +1.512.651.aaaa
14:07:47 <sandro> sandro has changed the topic to: agenda http://lists.w3.org/Archives/Public/public-rdf-dawg/2012AprJun/0142.html
14:07:51 <Zakim> + +1.512.651.aabb
14:08:08 <LeeF> RESOLVED: Approve minutes from http://www.w3.org/2009/sparql/meeting/2012-05-15
14:08:11 <LeeF> zakim, aabb is Arthur
14:08:11 <Zakim> +Arthur; got it
14:08:15 <bglimm> Last week's minutes look ok?
14:08:25 <bglimm> (silence)
14:09:15 <bglimm> LeeF: Next meeting 29th, usual time and place
14:09:19 <Olivier> regrets for next week
14:09:29 <bglimm> Axel sends regrets (ESWC)
14:09:39 <bglimm> LeeF: The week after is SemTec
14:09:56 <bglimm> ... Steve, Author, Lee and Matt will be there
14:10:09 <LeeF> s/Author/Arthur
14:10:12 <bglimm> s/Authur/Arthur/
14:10:29 <Zakim> +??P21
14:10:46 <bglimm> We can decide whether we have a teleconf on the 5th of June next week
14:11:19 <bglimm> LeeF: We want to see where we are for going to cand. REC or LC
14:11:36 <LeeF> zakim, who's on the phone?
14:11:36 <Zakim> On the phone I see SteveH, LeeF, kasei, Olivier, MattPerry, bglimm, AndyS, Chimezie (muted), Sandro, Arthur, ??P21
14:12:01 <Zakim> -??P21
14:12:19 <LeeF> topic: Document Status
14:12:20 <LeeF> http://www.w3.org/2009/sparql/wiki/To_PR
14:12:32 <cbuilara_> cbuilara_ has joined #sparql
14:12:39 <LeeF> subtopic: Query
14:12:41 <bglimm> ... let's start with Query
14:13:05 <bglimm> ... Andy worked in the new VALUES (?) in
14:13:30 <bglimm> ... Steve asks whether we want to use UNBOUND instead of UNDEF
14:13:36 <bglimm> ... as a keyword
14:13:49 <bglimm> AndyS: I have a slight preference for UNDEF
14:14:14 <bglimm> LeeF: Steve, you have a strong preference here?
14:14:22 <bglimm> SteveH: I have a mild preference
14:14:30 <SteveH> very mild :)
14:14:35 <kasei> I share Steve's mild preference.
14:14:52 <bglimm> LeeF: Let's just keep it as it is then
14:15:16 <bglimm> ... Andy & Steve can you summarize the TODOs
14:15:26 <bglimm> AndyS: We need another review for the PP stuff
14:16:01 <bglimm> ... There are some changes for string functions (?)
14:16:17 <bglimm> ... there are 4 open comments (mostly for the bindings stuff)
14:16:37 <bglimm> ... some for syntax for prefixes (no change planned)
14:17:21 <bglimm> LeeF: no open issues for Query
14:17:23 <kasei> q+
14:17:33 <LeeF> ack kasei
14:17:48 <bglimm> kasei: We skipped over the test suite
14:17:51 <AndyS> pgearon is down to review PP.
14:18:13 <bglimm> LeeF: We'll talk about this
14:19:07 <bglimm> LeeF: After Paul's review, should we do review of the whole doc?
14:19:19 <bglimm> AndyS: If somebody has the time it would be great
14:19:48 <AndyS> Also -- action number 620 Contact carlos to implement change from BINDINGS to VALUES in Fed-query for PR
14:19:48 <bglimm> LeeF: The substantive changes are PP, VALUES, little things like functions
14:20:00 <AndyS> Also -- Action number 619 Propose fix in definition of "in-scope" for BIND, cf. http://lists.w3.org/Archives/Public/public-rdf-dawg/2012AprJun/0120.html
14:21:05 <AndyS> s/619/618/
14:21:42 <AndyS> (editorial - fix references - check for unused refs)
14:21:54 <bglimm> LeeF: We need one review for the changes in the functions stuff
14:22:02 <bglimm> ... I suggest Greg or Matt
14:22:06 <kasei> sure
14:22:07 <MattPerry> Sure I can do it
14:22:29 <LeeF> zakim, choose Greg or Matt
14:22:29 <Zakim> I don't understand 'choose Greg or Matt', LeeF
14:22:34 <LeeF> zakim, random number?
14:22:34 <Zakim> I don't understand your question, LeeF.
14:22:44 <AndyS> 4027659
14:23:12 <bglimm> LeeF: I threw a coin and it's Matt
14:23:22 <LeeF> ACTION: Matthew to review UUID, STRBEFORE, STRAFTER changes in query
14:23:22 <trackbot> Created ACTION-623 - Review UUID, STRBEFORE, STRAFTER changes in query [on Matthew Perry - due 2012-05-29].
14:24:00 <bglimm> LeeF: With those two reviews, we are set
14:24:08 <Zakim> +??P29
14:25:00 <Zakim> -??P29
14:25:31 <bglimm> AndyS: There's an at risk box in the grammar section
14:25:38 <bglimm> I think we wanted to remove it
14:25:56 <LeeF> http://www.w3.org/2009/sparql/docs/query-1.1/rq25.xml#grammar
14:25:59 <bglimm> It covers the change for decimals and dates to be like in Turtle
14:26:19 <bglimm> LeeF: We only have to remove it for PR
14:26:33 <bglimm> AndyS: It doesn't have a need any more
14:26:52 <bglimm> LeeF: Do we have a syntax test for that?
14:27:12 <bglimm> ... It would make sense to do that, so that implementors update their parsers
14:27:30 <LeeF> ACTION: Lee to draft syntax tests for at-risk grammar features of query
14:27:30 <trackbot> Created ACTION-624 - Draft syntax tests for at-risk grammar features of query [on Lee Feigenbaum - due 2012-05-29].
14:27:58 <bglimm> LeeF: I'll create the tests, we approve them (next week) and then remove the at risk note
14:28:56 <bglimm> LeeF: Only the UUID test is not yet approve and the STRBEFORE/STRAFTER tests might need adptations
14:29:08 <bglimm> ... If there are no such tests we would need some
14:29:50 <bglimm> ... Greg, will you look at the STRBEFORE/STRAFTER tests?
14:29:53 <LeeF> ACTION: Greg to look at changes to or adding new STRBEFORE/STRAFTER tests to reflect updated behavior of error case (empty string, no language tag)
14:29:53 <trackbot> Created ACTION-625 - Look at changes to or adding new STRBEFORE/STRAFTER tests to reflect updated behavior of error case (empty string, no language tag) [on Gregory Williams - due 2012-05-29].
14:29:54 <bglimm> kasei: Yes
14:30:41 <bglimm> LeeF: We we have the tests, we can check the implementations
14:31:20 <bglimm> ... we do not have any of the update editors here
14:31:32 <bglimm> ... Does anybody here know the status?
14:31:42 <bglimm> ... (silence)
14:32:00 <bglimm> ... I'll draft an email to gather this information
14:33:06 <bglimm> .... I see one open Update comment about RDF merge and bnodes
14:33:12 <bglimm> .. from Kjetl
14:33:14 <LeeF> http://lists.w3.org/Archives/Public/public-rdf-dawg-comments/2012May/0023.html
14:34:29 <bglimm> LeeF: Greg (others) any idea about the tests?
14:34:46 <bglimm> kasei: the copy and move tests have not yet been approved
14:34:56 <bglimm> ... other than that we have three implementations
14:35:05 <bglimm> LeeF: Can we approve the tests?
14:35:11 <kasei> http://www.w3.org/2009/sparql/implementations/
14:35:17 <bglimm> kasei: Yes, they can be approved
14:36:22 <LeeF> PROPOSED: Approve the 6 COPY tests and 6 MOVE tests that are currently Not Classified
14:36:29 <kasei> +1
14:36:38 <Olivier> +1
14:36:46 <bglimm> I abstain as I haven't looked at them
14:37:03 <chimezie> same for me (for same reason)
14:37:26 <bglimm> AndyS: I have to run them
14:37:41 <bglimm> ... I pass all of them
14:37:47 <LeeF> RESOLVED: Approve the 6 COPY tests and 6 MOVE tests that are currently Not Classified, bglimm and chimezie abstaining
14:38:02 <bglimm> LeeF: Greg, can you mark them as approved
14:38:07 <bglimm> kasei: sure
14:38:16 <LeeF> ACTION: Greg to mark MOVE/COPY tests approved
14:38:17 <trackbot> Created ACTION-626 - Mark MOVE/COPY tests approved [on Gregory Williams - due 2012-05-29].
14:39:17 <chimezie> Zakim, unmute me
14:39:17 <Zakim> Chimezie should no longer be muted
14:39:20 <LeeF> http://lists.w3.org/Archives/Public/public-rdf-dawg-comments/2012May/0009.html
14:39:35 <bglimm> LeeF: I want to discuss the James Leigh (?) comment
14:39:44 <bglimm> s/(/)//
14:40:33 <bglimm> LeeF: There was some discussion around this
14:41:04 <bglimm> kasei: The SD doc was not designed to work with the graph store protocol and we didn't realise that this could be critical
14:41:18 <bglimm> ... at the moment they don't work well together
14:41:49 <bglimm> ... the graph store doc is REST based and we cannot discover anything well then
14:42:47 <bglimm> (scribe is lost now...)
14:43:00 <AndyS> Valid concern - framing seems unhelpful - many assumptions about "REST style"
14:43:11 <chimezie> There is a procedural issue with James comment: service discovery is out of scope for GSP
14:43:41 <chimezie> There is just other questions about whether this is really a hardcoding issue since it applies to other specified HTTP APIs
14:44:05 <AndyS> The templating and {} substitution seems to be the tip of an iceberg.
14:44:09 <bglimm> LeeF: I don't know whether there is anything we have to do right now
14:44:22 <bglimm> Sandro: I don't think we should change anything at this point
14:44:53 <bglimm> .... I think it is useful for what we need now with SPARQL
14:45:29 <bglimm> ... there is just no standard defined way (or only a complicated way) for getting from GSP to SD
14:46:15 <bglimm> Arthur?: We could just say that it is out of scope for this round of specification
14:46:26 <chimezie> s/Aurthur/chimezie
14:46:35 <bglimm> AndyS: We could put it on the future work list
14:46:51 <bglimm> ... we think there is nothing that blocks a future extension
14:46:57 <chimezie> s/Arthur?/chimezie
14:47:40 <LeeF> http://www.w3.org/2009/sparql/wiki/Future_Work_Items
14:47:49 <bglimm> LeeF: I'll add it to the future work items list, Chime, you can add this to your reply if you want
14:48:46 <chimezie> Comments requiring minor changes (IMO): MSO-1 (reasonable status codes to support)
14:49:06 <LeeF> http://lists.w3.org/Archives/Public/public-rdf-dawg-comments/2012May/0010.html
14:49:09 <kasei> q+
14:49:28 <bglimm> LeeF: It's not clear whether they are appropriate
14:49:28 <LeeF> ack kasei
14:49:42 <bglimm> kasei: The codes he suggests are not used for what he wants them to be used for
14:50:25 <bglimm> .... for example, the request is too large is different from the rest operation is being to large (query result)
14:51:13 <bglimm> LeeF: My reading of GSP is that HTTP status codes can be used as long as they are in line with the HTTP spec. We should clarify this
14:51:41 <chimezie> "implementations MUST include a status code [RFC2616] appropriate for the operation indicated and the result from invoking the operation"
14:51:48 <bglimm> chimezie: I agree with Greg
14:52:42 <chimezie> in 5.1 Status Codes
14:52:49 <bglimm> ... I wonder if our phrasing regarding the status codes is clear enough
14:53:05 <LeeF> """
14:53:07 <LeeF> The SPARQL Protocol uses the response status codes defined in HTTP to
14:53:07 <LeeF> indicate the success or failure of an operation. Consult the HTTP
14:53:07 <LeeF> specification [RFC2616] for detailed definitions of each status code.
14:53:07 <LeeF> While a protocol service should use a 2XX HTTP
14:53:07 <LeeF> response code for a successful query, it may
14:53:08 <LeeF> choose instead to use a 3XX response code as per HTTP.
14:53:10 <LeeF> """
14:53:58 <bglimm> LeeF: I think the SPARQL protocol spec is a bit more specific, but there's nothing wrong with this text IMO
14:53:59 <LeeF> """
14:54:01 <LeeF> The HTTP response codes applicable to an unsuccessful query operation include:400 if the SPARQL query supplied in the request is not a legal sequence of characters in the language defined by the SPARQL grammar; or,500 if the service fails to execute the query. SPARQL
14:54:01 <LeeF> Protocol services may also return a 500 response code if they
14:54:01 <LeeF> refuse to execute a query. This
14:54:01 <LeeF> response does not indicate whether the server may or may not
14:54:02 <LeeF> process a subsequent, identical request or requests. A protocol service may use other 4XX or 5XX HTTP response codes for other failure conditions, as per HTTP.
14:54:05 <LeeF> """
14:54:47 <bglimm> Chimezie: I draft a response saying that the first and the last use is not correct
14:56:00 <LeeF> Is there anything to review after the last pub?
14:56:27 <chimezie> no
14:56:32 <LeeF> Only MSO comment open?
14:56:40 <chimezie> Yes
14:56:50 <LeeF> We have to check the test suite status next week (running out of time)
14:57:02 <LeeF> protocol will be a longer discussion
14:57:12 <LeeF> any other business for the last minutes?
14:57:23 <bglimm> (silence)
14:57:38 <LeeF> Thanks to all
14:57:50 <LeeF> Adjourned
# SPECIAL MARKER FOR CHATSYNC. DO NOT EDIT THIS LINE OR BELOW. SRCLINESUSED=00000217