W3C

- DRAFT -

RDF Working Group Teleconference

26 Jun 2013

See also: IRC log

Attendees

Present
Guus_Schreiber, +1.408.992.aaaa, +1.415.686.aabb, EricP, TallTed, +1.707.318.aacc, cgreer, AndyS, pchampin, yvesr, Sandro, +1.415.686.aadd, GavinC, PatH, Arnaud, markus, Souri, ScottB
Regrets
Chair
Guus
Scribe
markus, yvesr

Contents


<trackbot> Date: 26 June 2013

<AndyS> zakim problems? code not recognized

<TallTed> trackbot, start meeting

<trackbot> Meeting: RDF Working Group Teleconference

<trackbot> Date: 26 June 2013

<markus> scribe: markus

guus: comments on minutes of last meeting?

<sandro> trackbot, start meeting

<trackbot> Meeting: RDF Working Group Teleconference

<trackbot> Date: 26 June 2013

RESOLUTION: accept the minutes of the 19 June telecon

pfps: there's no note on skolemization etc. it's not done

guus: ok, I'll change the state to "open"

<pfps> Concepts doesn't have anything on compatability or skolemization as per the action item.

publication of test suites

guus: the official title is Turtle test suite but that will have to change

<ericP> http://www.w3.org/2013/TurtleTests/

<ericP> Tests for Turtle - Terse RDF Triple Language

ericP: I believe the directory is also specific so we'll use other directories for other test suites

guus: I was talking about the wiki page

<ericP> http://www.w3.org/2011/rdf-wg/wiki/Turtle_Test_Suite#Turtle_Tests

ericp: we can certainly change that. I'll fix it
... there were 3-4 actions on me on this: license, zip, respond to some commenters
... we declined one request *sorry missed it, something with ASCII chars*
... have replies for other commenters ready
... everything is reflected in wiki page

guus: there's also a link from comments from andy

<gavinc> and someone needs to write back on PrEfIx/BaSe

ericp: there's zip file (maybe not up to date).. I'll update it

<AndyS> EricP: rolleg NT results to be classic NT i.e. ASCII only, not UTF-8

guus: license?

ericp: it's fixed

guus: so ready to be published?

ericp: once we agree to accept those changes

andys: where did you make the changes?

ericp: perhaps I didn't push it yet
... everything under cvs now, not under mercurial anymore

gavinc: (?) do I have to get the zip and diff to get the changes?

ericp: I can also give you the changes

gavinc: log from cvs is fine too

<ericP> http://www.w3.org/2013/TurtleTests/manifest.ttl

ericp: manifest has the changes.. look for "jeremy"
... I'll give you the log for the ascii in n-triples changes

<AndyS> AndyS: 291 tests, all passed, latest zip file.

<gavinc> yeah, but http://www.w3.org/2013/TurtleTests/comment_following_localName.ttl

<gavinc> So no, I can't see changes ;)

gkellogg: will this wiki page be updated with info for nquads and trig?
... I think there's no test suite for n-quads yet
... I'm happy to create a first version of the n-quads test suite

guus: I would just call it RDF Test Suites

sandro: maybe just RDF Tests?

<ericP> RDF Test Sweets?

<gavinc> Clearly

ACTION on gkellog to create first version of n-quads test suite

<trackbot> Error finding 'on'. You can review and register nicknames at <http://www.w3.org/2011/rdf-wg/track/users>.

ACTION for gkellog to create first version of n-quads test suite

<trackbot> Error finding 'for'. You can review and register nicknames at <http://www.w3.org/2011/rdf-wg/track/users>.

<ericP> RDF Test Suites

<scribe> ACTION: gkellog to create first version of n-quads test suite [recorded in http://www.w3.org/2013/06/26-rdf-wg-minutes.html#action01]

<trackbot> Error finding 'gkellogg'. You can review and register nicknames at <http://www.w3.org/2011/rdf-wg/track/users>.

<tlr> heads-up: I expect a *very* highly attended TPWG call at the top of the hour.

<tlr> Please clear the bridge around 11:58 or so

<tlr> thank you

<scribe> ACTION: gkellogg to create first version of n-quads test suite [recorded in http://www.w3.org/2013/06/26-rdf-wg-minutes.html#action02]

<trackbot> Created ACTION-275 - Create first version of n-quads test suite [on Gregg Kellogg - due 2013-07-03].

<AndyS> gavinc: TriG ready LC draft next week

gavinc: n-triples and n-quads originally planned to publish as one doc

<AndyS> ... NQ, NT: easier as two docs , ready for review next week

gavinc: we currently have two docs. think it's easier to keep it that way

<pfps> so we are delaying action on Concepts even further??

sandro: I have some issues with Trig/SPARQL alignment

gavinc: we already have resolutions on that

<ericP> +1 to could and should address now

sandro: I understand that but I think we could resolve these problems now pretty easily

gavinc: there are no open issues at the moment

JSON-LD

<yvesr> scribe: yvesr

gkellogg: We had a productive call with pfps yesterday
... We ended up after an hour of discussion on not refering to the 'JSON-LD data model'
... But just on 'data model'
... Which is based on the RDF data model
... So that we don't perpetuate the idea that there is a distinct data model for JSON-LD
... There are still issues around native datatypes
... and about blank nodes as predicates
... Some important distinctions in the handling of lists and sets in JSON-LD
... my personal opinion is that JSON-LD lists are exactly the same as RDF lists

<AndyS> where is the list of changes from last LC?

pfps: My first comment is the schism between the two different worlds

<markus> andys, there's a changelog at the beginning of the doc

<gavinc> well... hey the JSON-LD website changed a lot...

pfps: If the formal basis of JSON-LD ends up being RDF graphs or an extended version of that I would be happy

<AndyS> markus - which is the editors doc currently?

Guus: Can you leave with the current changes?

pfps: I haven't seen them yet, but yes

<markus> andys, it's always on json-ld.org.. we sync to w3.org manually from the time to time

markus: This will require a little bit of back and forth

gkellogg: We think these changes won't require another LC

Guus: Can we leave at that at the moment, pending edits?

sandro: We need to make a decision on whether we need that extension to RDF graphs
... and whether it should leave in RDF concepts

pfps: It's also possible for the JSON-LD document to define that extension of RDF graphs
... that said it would be better in RDF concepts
... If it ends up being duplicated in the two documents, we can still rip that out later from one of those documents
... It is purely editorial

<PatH> please dont call it a linked data graph.

sandro: If we extended RDF graphs as a, said, 'Linked Data graph', how would it be different from the current definition of RDF graphs

<gavinc> No.

pfps: There are no commonalities between the two atm - everything is being redefined from the ground up
... the only common ground is IRIs
... It is reasonable to say that Linked Data is what you think of when you're programming against it

<gkellogg> Its hard to see how adding bnode predicates would be described in a Linked Data graph.

<gavinc> Yeah, gkellogg, that.

pfps: RDF provides the details

<markus> markus: I'm afraid that if we call it Linked Data graph we end up having another Linked Data discussion

sandro: Or 'Generalized RDF graph'?

<markus> markus: I would prefer JSON-LD graph

PatH: We shouldn't call anything 'Linked Data graph'

Guus: Thanks for having this discussion which helped resolve this

<PatH> JSON-RDF-graph?

<markus> scribe: markus

Concepts and Semantics

<pfps> consensus has been reached!

<pfps> there is one editorial issue remaining

guus: antoine can you confirm we reached consensus?
... ok, not on the call. is my observation correct?

<PatH> Antoine said he woul dbe late on the call but would be here.

path: his latest email suggested a niggle about one phrase
... maybe an editorial tweak will address that
... rest should be ok

guus: what about datasets and default graphs?

path: I'm waiting to know what to put in the document
... did we resolve that already

sandro: we did

path: ok, if we are clear on that I'll update the document

guus: three messages from pfps

pfps: the first one is out of scope (bnodes as predicates)
... the rest is explaining the current state, all editorial
... the third one is a proposal to define generalized graphs and datasets in Concepts
... this is the result of the JSON-LD discussion

guus: I guess "extended" could be "generalized"

pfps: yes

<pfps> that's 'extended' could be 'generalized', not extended could be generalized (further)

sandro: would like to get an idea where people stand on this

<pfps> the proposal is to define generalized RDF graphs in Concepts, for use in Semantics

path: it has come up over and over again in the last decade.. would like to solve it once for all

<pfps> The section could certainly be labelled non-normative

<gavinc> No.

<gavinc> You can't use Turtle.

sandro: we need to be careful how we phrase it (non-normative etc.) so that people who don't want it don't have to do anything

<gavinc> Correct pfps!

<pfps> one could certainly build generalized turtle, generalized RDF/XML (well, probably not!), ...

<Guus> http://lists.w3.org/Archives/Public/public-rdf-wg/2013Jun/0204.html

<Guus> strawpoll: what's your opinion of adding this to Concepts?

<PatH> +1

+1

<pfps> s/extended/generalized/ and make non-normative

<Arnaud> +1

sandro: +1

<pfps> +1

<pchampin> +1

<cgreer> +a

<AndyS> OK : but not "define" - discussion text, not definitional

<ericP> +1

<yvesr> +1

<gkellogg> +1

<PatH> Sandro +1

<cgreer> +1

<TallTed> +1

<gavinc> 0

<PatH> a <0.0005?

<Souri> 0

<PatH> Yes I think so yvesr

<cgreer> a == 1

<PatH> Does it allow literals in property position?

<PatH> OK

<AndyS> and datatypes?

<gavinc> "An extended RDF triple is an RDF triple, but with subjects, predicates, and objects all allowed to be IRIs, blank nodes, or literals."

guus: ok, I think that covered all the discussions we had

<gavinc> Doesn't talk about datatyes

guus: where do we stand with publishing Concepts and Semantics?
... we should try to vote on it next week

pfps: Semantics is more or less done
... only thing left is to label some sections as non-normative
... Semantics already uses extended RDF graphs

path: but it doesn't say so

pfps: yeah
... and the semantics of the default graph are not mentioned yet

guus: have all reviews been addressed?

pfps: yes

guus: markus, you have been asked by david to help as editor for Concepts

markus: right.. waiting for a reply to my questions before making a decision whether I accept

pfps: I reviewed the comments

guus: pfps, can you inform the WG when Semantics is ready?

pfps: yes

guus: I will ask David to do the same for Concepts
... we really need to get it out now

<AndyS> Move on to RDF/JSON?

<PatH> OK peter, you work on it and email me when done, then I can take it from there.

<ericP> gavinc, you should be see the diffs in your inbox

<pfps> action on Concepts editors to add bit on generalized RDF graphs?

<trackbot> Error finding 'on'. You can review and register nicknames at <http://www.w3.org/2011/rdf-wg/track/users>.

action guus to ask david on the LC draft of Concepts

<trackbot> Created ACTION-276 - Ask david on the LC draft of Concepts [on Guus Schreiber - due 2013-07-03].

<ericP> i didn't compile diffs for the changes moving all U+FFFF to U+FFFD ('cause U+00..FFFE and U+00..FFFF are non characters)

<PatH> I could help with Concepts editing also. Whatever.

<pfps> I can make changes to Concepts if needed

sandro: I may suggest peter to step in and just make the changes

guus: ok

<ericP> gavinc, i take it back. i tried to send it through gmail which is resisting

RDF/JSON as Note

<gavinc> heh, okay, yeah didn't see it

<pfps> I have no problems with RDF/JSON as a note

arnaud: I already updated the draft to make the status of the draft clearer
... I'm aware that the WG is concerned about publishing two RDF serializations in JSON

<ericP> gavinc, gzip'ing worked. should be there now

arnaud: all I'm asking to revisit this decision
... I'm happy to make all the necessary changes

<AndyS> update to datasets?

arnaud: just so that we have a stable reference
... I updated the draft to clearly state that the WG did chose JSON-LD and that people should use that

<AndyS> I can review

guus: I would like to have 1-2 WG members to have a look at the current doc

sandro: will do

cgreer: me too

guus: so there won't be anything to vote on today

<ericP> gavinc, try now?

gavinc: no, I'll send a mail to the list

<tlr> zakim congestion happening now. Please end this meeting.

<ericP> (i could have sworn i did that already)

ADJOURNED

<tlr> thanks!

bye

<Guus> trackbot, end meeting

Summary of Action Items

[NEW] ACTION: gkellog to create first version of n-quads test suite [recorded in http://www.w3.org/2013/06/26-rdf-wg-minutes.html#action01]
[NEW] ACTION: gkellogg to create first version of n-quads test suite [recorded in http://www.w3.org/2013/06/26-rdf-wg-minutes.html#action02]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.138 (CVS log)
$Date: 2013-06-26 16:01:58 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.138  of Date: 2013-04-25 13:59:11  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/pushed/push it/
Succeeded: s/trig yet/n-quads yet/
Succeeded: s/gkellog/gkellogg/
Succeeded: s/andys: there's/andys, there's/
Succeeded: s/models/worlds/
WARNING: Bad s/// command: s/extended/generalized/ and make non-normative
Found Scribe: markus
Inferring ScribeNick: markus
Found Scribe: yvesr
Inferring ScribeNick: yvesr
Found Scribe: markus
Inferring ScribeNick: markus
Scribes: markus, yvesr
ScribeNicks: markus, yvesr
Default Present: Guus_Schreiber, +1.408.992.aaaa, +1.415.686.aabb, EricP, TallTed, +1.707.318.aacc, cgreer, AndyS, pchampin, yvesr, Sandro, +1.415.686.aadd, GavinC, PatH, Arnaud, markus, Souri, ScottB
Present: Guus_Schreiber +1.408.992.aaaa +1.415.686.aabb EricP TallTed +1.707.318.aacc cgreer AndyS pchampin yvesr Sandro +1.415.686.aadd GavinC PatH Arnaud markus Souri ScottB
Found Date: 26 Jun 2013
Guessing minutes URL: http://www.w3.org/2013/06/26-rdf-wg-minutes.html
People with action items: gkellog gkellogg

[End of scribe.perl diagnostic output]