W3C


RDF Working Group Teleconference

18 Dec 2013

See also: IRC log

Attendees

Present
Andy Seaborne, Guus Schreiber, Gavin Carothers, Ted Thibodeau, Sandro Hawke, Gregg Kellogg, Ivan Herman, Peter Patel-Schneider, Arnaud l Hors, Yves Raimond, Richard Cyganiak, Antoine Zimmerman, David Wood, Pat Hayes, Eric Prud'hommeaux, Zhe Wu, Markus Landhaller, Manu Sporny, Souripriya Das
Regrets
Chair
Guus
Scribe
Ted Thibadeau, Gaving Carothers

Contents


<trackbot> Date: 18 December 2013

<gavinc> Hey AndyS, https://dvcs.w3.org/hg/rdf/raw-file/e054e768ca9f/drafts/trig/Overview.html should be everything for TriG

<AndyS> gavinc - phew!

<gavinc> Only question editorial I had was do we want to just use PREFIX style examples in TriG?

<AndyS> gavinc - don't mind - don't consider to necessary as I expect @prefix to be around for a while. No big reason to change. Old data, old examples will be around until ...?

<TallTed> scribenick: TallTed

PROPOSED: accept the minutes of the 11 Dec telecon, https://www.w3.org/2013/meeting/rdf-wg/2013-12-11

<pfps> minutes look fine

<AndyS> +1

RESOLUTION: accepted the minutes of the 11 Dec telecon

action items

<AZ> I can't load the minutes

<pfps> last week's minutes are at http://www.w3.org/2011/rdf-wg/wiki/Meetings:Telecon2013.12.18

<pfps> Oops, not there!

http://www.w3.org/2011/rdf-wg/wiki/Meetings:Telecon2013.12.11

<pfps> I just was able to look at the minutes, but they are not loading right now - I claim that they looked OK

WG activity plan

Guus: plans in Agenda seem feasible, with 3 month WG extension

<pfps> +1 to request extension

<davidwood> +1 to extend

<yvesr> +1

PROPOSED: to Request W3M to extend for 2/3 months, with voluntary biweekly meetings in that period

<gkellogg> +1

sandro: +1

+1

<pchampin> +1

<lanthaler> +1

<Arnaud> +1

<cygri> +1

<AZ> +1

<ivan> +1

RESOLUTION: Request W3M to extend for 2/3 months, with voluntary biweekly meetings in that period

<PatH> +1

PROPOSED: the RDF WG petition the Director to take JSON-LD to Recommendation immediately after the publication of the Proposed Recommendations for the rest of the RDF 1.1 work.

<davidwood> +1

<gkellogg> +1

<PatH> +1

<lanthaler> +1

<AndyS> 0

<gavinc> 0

<pfps> -1

<cygri> 0

<yvesr> +0

<manu> +1

<Zakim> davidwood, you wanted to discuss messaging

<AZ> 0

<davidwood> I prefer to allow JSON-LD to go to REC early. It is messaging to a different community. We wanted to tie JSON-LD to RDF, which which have done. We do not also need to mess up their messaging.

<davidwood> PR-REC is a staff decision.

RESOLUTION: the RDF WG petition the Director to take JSON-LD to Recommendation immediately after the publication of the Proposed Recommendations for the rest of the RDF 1.1 work. over objection from pfps

<pfps> my concern is that JSON-LD should progress in step with the rest of RDF, as it should closely depend on the core aspects of RDF

CR implementations

PROPOSED: the WG decides it has met the CR Exit Criteria for RDF 1.1 Turtle, TriG, NiTriples, and N-Quads (see statistics in PR Request http://www.w3.org/2011/rdf-wg/wiki/RDF11-PR-Request#CR_Exit_Criteria_Turtle.2C_Trig.2C_N-Triples_and_N-Quads )

<pfps> +1

<gkellogg> +1

<ivan> +1

<gavinc> +1

<PatH> +1

<ericP> +1

<yvesr> +1

<davidwood> +1

<AndyS> +1

<zwu2> +1

<cygri> Implementation reports: http://www.w3.org/2011/rdf-wg/wiki/#Implementation_reports

<gavinc> https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/reports/index.html

<gavinc> https://dvcs.w3.org/hg/rdf/raw-file/default/trig/reports/index.html

<gavinc> https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/reports-nt/index.html

+1

<gavinc> https://dvcs.w3.org/hg/rdf/raw-file/default/nquads/reports/index.html

RESOLUTION: the WG decides it has met the CR Exit Criteria for RDF 1.1 Turtle, TriG, NiTriples, and N-Quads (see statistics in PR Request

<AZ> +1

PROPOSED: the WG decides to rescinded the two tests below as they are not helpful to interoperability, for the reasons outlined in http://lists.w3.org/Archives/Public/public-rdf-wg/2013Dec/0305.html -- https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-mt/reports/index.html#test_datatypes-intensional-xsd-integer-string-incompatible ; https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-mt/reports/index.html#test_xmlsch-02-whitespace-face

t-3

<gkellogg> +1

<PatH> +q

PatH: if unimportant, why were they go here at all?

<gavinc> http://www.w3.org/TR/2005/WD-swbp-xsch-datatypes-20050427/#sec-values-differ ?

Guus: first one is axiomatic to basic functionality; if you don't have it, you can't operate at all...

<PatH> Yes, exactly.

<ericP> another take on sandro's point is whether *customers* would care about an implementation which passes those tests

<ivan> +1

<markus> +1

<cygri> +0.5

<gavinc> +meh

<gavinc> +0

<ericP> +1

<AZ> +1

<zwu2> +1

<davidwood> +1

<pfps> +0.5 with the same concern that Pat had

+0.5

<Guus> +1

<PatH> +1 though I feel rushed here.

RESOLUTION: the WG decides to rescinded the two tests below as they are not helpful to interoperability, for the reasons outlined in http://lists.w3.org/Archives/Public/public-rdf-wg/2013Dec/0305.html -- https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-mt/reports/index.html#test_datatypes-intensional-xsd-integer-string-incompatible ; https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-mt/reports/index.html#test_xmlsch-02-whitespace-face

t-3

<PatH> Andy, the sematnics actually makes some infrences valid that are so obscure that nobody cares about them. Who knew?

<scribe> ACTION: gkellogg to update manifest and regenerate implementation report [recorded in http://www.w3.org/2013/12/18-rdf-wg-minutes.html#action01]

<trackbot> Created ACTION-335 - Update manifest and regenerate implementation report [on Gregg Kellogg - due 2013-12-25].

PROPOSED: the WG decides it it has met the CR Exit Criteria for RDF 1.1 Semantics (see implementation report, https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-mt/reports/index.html )

<PatH> gavin, it predicts the opposite, ie the subclass is D-inconsistent.

<cygri> +1

<markus> +1

+1

<ivan> +1

<pfps> +1

<gavinc> +1

<AZ> +1

<gkellogg> +1

<yvesr> +1

<Guus> +1

<zwu2> +1

<davidwood> +1

<Souri> +1

<AndyS> +0

<PatH> +1

RESOLUTION: the WG decides it it has met the CR Exit Criteria for RDF 1.1 Semantics (see implementation report, https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-mt/reports/index.html )

<gavinc> path, yeah that's what I meant

<PatH> Sorry, see that now.

Guus: question for the WG. markus and gkellogg produced RDF 1.1 Test Cases, https://dvcs.w3.org/hg/rdf/raw-file/default/rdf11-testcases/index.html
... do we want to publish this as a NOTE, so we can clearly reference it through other docs?

<davidwood> I don't believe the W3C mercurial repository is stable - the sysreq team is already talking about migrating it.

<PatH> +1 cygri

cygri: corresponding doc from 2004 was Rec. downgrade to Note seems reasonable; having it in TR space seems a good thing.

Guus: we may try to do this in January.

Status of Syntax documents

gavinc: all the Agenda-listed things have been done

<gavinc> ISSUE-119?

<trackbot> ISSUE-119 -- Spec should reference the test suite -- open

<trackbot> http://www.w3.org/2011/rdf-wg/track/issues/119

<markus> have the syntax documents been updated to mention rdf:langString.. or is that considered to be unnecessary?

<gavinc> PROPOSE: Resolve ISSUE-119, Turtle has a link to test suite https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/index.html

<ericP> PROPOSED: close issue-119 given that spec now references test suite in multiple ways

<gkellogg> =1

<gkellogg> +1

<gavinc> +1

+1

<markus> +1

<davidwood> +1

<ericP> +1

<ivan> +1

<AndyS> +1

<Guus> +1

<Arnaud> +1

<Souri> +1

<pfps> +1

<zwu2> +1

RESOLUTION: close issue-119, Turtle has a link to test suite https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/index.html

<AZ> +1

<yvesr> +1

<AndyS> markus - they have taken the line of being "syntax" so no. Arguable either way.

<PatH> Dancing lemmings?

<gavinc> The literal has a lexical form of the first rule argument, String. If the '^^' iri rule matched, the datatype is iri and the literal has no language tag. If the LANGTAG rule matched, the datatype is rdf:langString and the language tag is LANGTAG. If neither matched, the datatype is xsd:string and the literal has no language tag.

markus: questions address of rdf:langstring in all syntax docs

<ericP> https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/index.html#handle-RDFLiteral

Concepts & Semantics

<ericP> If the LANGTAG rule matched, the datatype is rdf:langString and the language tag is LANGTAG.

issue-148?

<trackbot> issue-148 -- CR comment: IRIs do *not* always denote the same resource -- open

<trackbot> http://www.w3.org/2011/rdf-wg/track/issues/148

<davidwood> I am already on record as saying that I consider the existing wording to be adequate, but would accept some minor adjectival adjustments.

<ericP> -1 to "are expected ot"

<PatH> where are the alternatives?

<PatH> OK

<cygri> http://www.w3.org/2011/rdf-wg/wiki/Meetings:Telecon2013.12.18#ISSUE-148:_IRIs_do_.2Anot.2A_always_denote_the_same_resource

<markus> IRIs have global scope by definition. Thus, two different appearances of an IRI denote the same resource. RDF is based on this principle and violations of it might lead to inconsistencies or interoperability problems.

option 1 - "IRIs have global scope by definition. Thus, two different appearances of an IRI denote the same resource. RDF is based on this principle and violations of it might lead to inconsistencies or interoperability problems."

<markus> vs.

<markus> By design, IRIs have global scope. Thus, two different appearances of an IRI denote the same resource. Violating this principle constitutes an IRI collision [WEBARCH].

option 2 - "By design, IRIs have global scope. Thus, two different appearances of an IRI denote the same resource. Violating this principle constitutes an IRI collision [WEBARCH]."

<ericP> pref 1 'cause it's stronger

<pfps> I don't care - either are fine by me

<cygri> [WEBARCH] would be: http://www.w3.org/TR/webarch/#URI-collision

<ericP> 1

<cygri> 2

<davidwood> People lie. People make mistakes. This is RDF 1999.

<markus> 1

<PatH> 2

<pfps> 1.5

<yvesr> 2

<gavinc> don't care.

<AZ> 2

<gkellogg> 1=2

<zwu2> 1

<Arnaud> 2

1

<davidwood> 2

<ivan> 2

<AndyS> 2 - preder ref to WEBARCH

<ericP> 2 has it

<Souri> 2

<davidwood> 2

<davidwood> :)

<PatH> No need to solicit an objection. They will come.

PROPOSED: Resolve ISSUE-148 by changing the "IRIs have global scope" bullet point in section 1.3 in Concepts to "By design, IRIs have global scope. Thus, two different appearances of an IRI denote the same resource. Violating this principle constitutes an IRI collision [WEBARCH]." http://www.w3.org/TR/webarch/#URI-collision

<ivan> +1

<PatH> +1 to that remark

<pfps> +1 and allow normal wordsmithing edits

<davidwood> +1

<cygri> +1

<Guus> +1

<gavinc> +0

<Souri> +1

<markus> +0

<zwu2> +0

<AndyS> +0

<PatH> +1 to the proposal as well.

<Arnaud> +1

<ericP> A+1

<AZ> +1

+0.1

<gkellogg> _+1

RESOLUTION: Resolve ISSUE-148 by changing the "IRIs have global scope" bullet point in section 1.3 in Concepts to "By design, IRIs have global scope. Thus, two different appearances of an IRI denote the same resource. Violating this principle constitutes an IRI collision [WEBARCH]." http://www.w3.org/TR/webarch/#URI-collision

<PatH> +1 ??

<scribe> ACTION: Guus to inform commenter of resolution of issue-148 [recorded in http://www.w3.org/2013/12/18-rdf-wg-minutes.html#action02]

<trackbot> Created ACTION-336 - Inform commenter of resolution of issue-148 [on Guus Schreiber - due 2013-12-25].

issue;165?

issue-165?

<trackbot> issue-165 -- CR Comment: datatype map -- open

<trackbot> http://www.w3.org/2011/rdf-wg/track/issues/165

PROPOSED: the WG resolves to close this issue, with the rationale stated in the last response to the commenter (http://lists.w3.org/Archives/Public/public-rdf-comments/2013Dec/0098.html), noting the objection from the commenter in the Transition Request.

<pfps> +1

cygri: what are consequences of formal objection at this stage?

ivan: director must be convinced to over-ride objection

<PatH> I think "formal objection" =/= "objection" (does that make sense)?

<pfps> I had thought that WG members could change their mind if new information comes to light, and it seems to me that the information from Michael could be considered to be new information of a sort.

<gavinc> There is of yet NOT a formal objection

<gavinc> there is a statement that says there will be one

<pfps> +1 (again)

<markus> +1

<AndyS> +1

<PatH> +1

+1

<gkellogg> +1

<ivan> +1

<AZ> -1

<gavinc> +1

<ericP> +1+1

<sandro> +1

<cygri> +1

<zwu2> +1

<davidwood> +1

<Guus> +1

<yvesr> +1

<AZ> (agree to move forward and close the issue, but disagree with the rationale)

<Souri> +1

RESOLUTION: the WG resolves to close this issue, with the rationale stated in the last response to the commenter (http://lists.w3.org/Archives/Public/public-rdf-comments/2013Dec/0098.html), noting the objection from the commenter in the Transition Request, over objection of AZ

Marking rdf:HTML and rdf:XMLLiteral as non-normative?!

PROPOSAL: to accept the resolution text in http://lists.w3.org/Archives/Public/public-rdf-wg/2013Dec/0283.html

<pfps> I agree with Guus's proposal as being the best way forward giving the current circumstances.

sandro: we should be very clear about the decision space we're working in...

<cygri> http://lists.w3.org/Archives/Public/public-rdf-wg/2013Dec/0267.html

<scribe> scribenick: gavinc

cygri: I very much want to see the URIs mentioned in this section one way or another. Trying to keep the datatypes normative by peppering notes in the relevant sections.

markus: The whole thing is about removing a MUST statement?

cygri: I'd like to see the notes mentioning WHY they have to be normative. Alert the readers to the existence of the datatypes.

markus: Doesn't have to be the MUST statement in 5.4

cygri: See email http://lists.w3.org/Archives/Public/public-rdf-wg/2013Dec/0267.html showed my preferred way of dealing with it but other ways could work.

<PatH> Seems to me that this owuld be us saying, whatever the other WG does, we want RDF to treat it as normative.

sandro: Try to explain what we actually mean, and leave to the Directors meeting figuring out how to do that

<PatH> What is the future tense of "normative"?

+q

ericP: There are no tests, how would that work?

markus: the features are optional anyway

<Zakim> ericP, you wanted to ask whether "normative" would require tests

<ericP> gavinc: how does everything else that refs DOM4 deal with this?

<ericP> ... we can't be the first to have this problem

<PatH> So, this is what a tar-pit feels like.

Guus: want to have a resolution on record
... can we add to the resolution something about agreeing during transition meeting?

<AZ> isn't it similar to having OWL 2 become a rec before XSD 1.1 became rec?

<Guus> PROPOSED: to accept the resolution text in http://lists.w3.org/Archives/Public/public-rdf-wg/2013Dec/0283.html.

<PatH> AZ, what did they do?

<ericP> +1 to delegating to the chairs and editors the resolution of refs to DOM4

AZ, no this is the left over debris from XHTML2 :P

<cygri> +1

<sandro> PROPOSED: to accept the resolution text in http://lists.w3.org/Archives/Public/public-rdf-wg/2013Dec/0283.html but delegate to the chairs and editors the authority to change on this matter

<sandro> PROPOSED: to accept the resolution text in http://lists.w3.org/Archives/Public/public-rdf-wg/2013Dec/0283.html but delegate to the chairs and editors (of RDF Concepts) the authority to change on this matter

<cygri> +1

<PatH> +1

+1

<sandro> +1

<yvesr> +1

<davidwood> +1

<AZ> +1

<ivan> +1

<gkellogg> +1

<TallTed> +1

<zwu2> +1

<PatH> AZ, tnx.

<sandro> AZ, we do NOT want to do that again. It was a nightmare.

<markus> -0.5 (I'm not happy with the "implementation-dependent" text)

<Guus> +1

<davidwood> It is good to have cyri back just in time for this issue.

<davidwood> oops :)

<PatH> FOFL

<sandro> RESOLUTION: to accept the resolution text in http://lists.w3.org/Archives/Public/public-rdf-wg/2013Dec/0283.html but delegate to the chairs and editors (of RDF Concepts) the authority to change on this matter

<cygri> .

Guus: Did we agree on References between Concepts and Semantics?

markus: Well, no one is complaining loudly anymore

PROPOSED: Make reference from Concepts to Semantics non-normative

<cygri> +1

<markus> +1

<AZ> +1

<davidwood> I have no objection

<davidwood> +1

<gkellogg> +1

<yvesr> +1

+1 Concepts does not require passing the semantics test cases

<AndyS> 0 (AKA it seems strange to me)

<ivan> 0

<Arnaud> 0

<Guus> )

<PatH> 0

<zwu2> 1

<Guus> 0

<TallTed> +1

<ericP> +1

RESOLUTION: Make reference from Concepts to Semantics non-normative

<Guus> PROPOSED: to accept the resolution text in http://lists.w3.org/Archives/Public/public-rdf-wg/2013Dec/0283.html.

PROPOSED: to request the Director to advance Concepts and Semantics to Proposed Recommendation.

<ericP> +1

<davidwood> +1

<markus> +1

<ivan> =1

<AndyS> +1

<ivan> +1

<Souri> +1

+1

<gkellogg> +1

<TallTed> +1

<Arnaud> +1

<cygri> +1.1

<PatH> +1

<zwu2> +1

<sandro> +1

<yvesr> +1

RESOLUTION: to request the Director to advance Concepts and Semantics to Proposed Recommendation

<PatH> Was that two negative votes form Ivan or just a shift wrror?

<PatH> :)

RDF Schema

Guus: Already voted on it last week
... only thing to do is to remove "This section is non-normative." from rdf:HTML and rdf:XMLLiteral

RDF/XML

Guus: There is an editors draft
... it has half of the changes from Ivan's review

most notable change is to XMLLiteral

scribe: gavinc should check if I did it the right way

https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-xml/index.html#parseTypeLiteralPropertyElt

ACTION gkellogg find some test cases for https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-xml/index.html#parseTypeLiteralPropertyElt

<trackbot> Created ACTION-337 - Find some test cases for https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-xml/index.html#parsetypeliteralpropertyelt [on Gregg Kellogg - due 2013-12-25].

PROPOSED: to request the Director to advance RDF 1.1 XML Syntax to Proposed Edited Recommendation.

<gkellogg> +1

<PatH> +1

+1

<TallTed> +1

<AndyS> +0.9

<yvesr> +1

<Arnaud> +1

<AZ> +1

<ivan> +1

<zwu2> +1

<Souri> +1

<Guus> +1

<davidwood> +1

<ericP> it's ok. i was just wondering if readers of the serialize() spec actually understood the XQuery or XSLT semantics which back it up

<cygri> +1

<ericP> +1

<markus> +0.9 (let it die :-)

RESOLVED: the WG requests the Director to advance RDF 1.1 XML Syntax to Proposed Edited Recommendation.

ericP, more likely then understanding infoset node sets and c14n

<Guus> PROPOSED: the WG decides to keep the Turtle features at risk: The addition of sparqlPrefix and sparqlBase which allow for using SPARQL style BASE and PREFIX directives in a Turtle document.

<ericP> http://www.w3.org/2013/meeting/rdf-wg/2013-05-29#Turtle

<ericP> [[

<ericP> RESOLUTION: Allow PREFIX and BASE in Turtle, and thus keep the feature in the document. This is based on the poll results at https://www.w3.org/2002/09/wbs/46168/at-sign/results

<PatH> w3c.org is feeling tired.

<ericP> ]]

<PatH> Seems to have come back up now.

Open issues (other than CR comments)

Guus: We did ISSUE-78 in primer. Making clear limitations, etc
... marking resolved.
... ISSUE-102 on primer now

cygri: too late to put in Concepts or RDF Schema, putting it in the primmer has limited benefit. Defining the term well formed list would have been good, but we didn't do it.
... just close with a No.

<cygri> PROPOSAL: Close ISSUE-102 doing nothing as the WG is out of time.

<pfps> +1

+1

<PatH> +1

<davidwood> +1

<Arnaud> +1

<ericP> +0 (have done no research)

<sandro> +0 agreed we're out of time; sad we did nothing

<yvesr> +0

<TallTed> +0

<zwu2> +1

<cygri> +0

<AZ> +1

<ivan> 0

RESOLUTION: Close ISSUE-102 doing nothing as the WG is out of time

<gkellogg> +1

ericP: if we say TriG is the One

Guus: Leaving open issue-138

<PatH> There is a widespread assumption that RDF/XML is priviledged

<pfps> ... and cursed

AOB

15 January?

<pfps> 15 Jan is fine by me

<davidwood> ok by me

<AZ> ok

<PatH> Fine

okay

<yvesr> ok

scribe: every other week

<Arnaud> fine with me

<davidwood> Bi-weekly schedule from 15 Jan

<Arnaud> good job!

Guus: Adjourned!

<ericP> RESOLUTION: next meeting 15 Jan, and every other week following

<PatH> Consensus (n): The state of exhaustion where one no longer cares what is decided.

<PatH> Happy holidays, everyone.

<AndyS> Bye all!

<AndyS> ADJOURNED


Summary of Action Items

[NEW] ACTION: gkellogg to update manifest and regenerate implementation report [recorded in http://www.w3.org/2013/12/18-rdf-wg-minutes.html#action01]
[NEW] ACTION: Guus to inform commenter of resolution of issue-148 [recorded in http://www.w3.org/2013/12/18-rdf-wg-minutes.html#action02]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.138 (CVS log)
$Date: 2013/12/19 11:55:32 $