Chatlog 2013-01-16

See panel, original RRSAgent log or preview nicely formatted version.

Please justify/explain non-obvious edits to this page, in your "edit summary" text.

15:54:25 <RRSAgent> RRSAgent has joined #rdf-wg
15:54:25 <RRSAgent> logging to http://www.w3.org/2013/01/16-rdf-wg-irc
15:54:32 <davidwood> Zakim, this is rdf
15:54:33 <Zakim> davidwood, I see SW_RDFWG()11:00AM in the schedule but not yet started.  Perhaps you mean "this will be rdf".
15:54:44 <davidwood> Zakim, this will be rdf
15:54:44 <Zakim> ok, davidwood; I see SW_RDFWG()11:00AM scheduled to start in 6 minutes
15:55:05 <davidwood> Chair: David Wood
15:55:23 <davidwood> I think we will need a scribe replacement...
15:57:01 <Guus> Guus has joined #rdf-wg
15:57:46 <Zakim> SW_RDFWG()11:00AM has now started
15:57:53 <Zakim> +Guus
15:58:09 <Zakim> +bhyland
15:58:26 <davidwood> Zakim, bhyland is me
15:58:27 <Zakim> +davidwood; got it
15:58:37 <AZ> AZ has joined #rdf-wg
15:59:49 <Zakim> +[GVoice]
15:59:55 <ericP> Zakim, GVoice is me
15:59:55 <Zakim> +ericP; got it
16:00:15 <markus> markus has joined #rdf-wg
16:00:31 <Zakim> +??P13
16:00:44 <Zakim> +??P14
16:00:50 <SteveH> Zakim, ??P14 is me
16:00:50 <Zakim> +SteveH; got it
16:00:53 <Arnaud> Arnaud has joined #rdf-wg
16:01:06 <AndyS> AndyS has joined #rdf-wg
16:01:11 <gavinc> gavinc has joined #rdf-wg
16:01:35 <Zakim> +Arnaud
16:01:56 <Zakim> +GavinC
16:02:04 <yvesr> Zakim, who is on the phone?
16:02:04 <Zakim> On the phone I see Guus, davidwood, ericP, ??P13, SteveH, Arnaud, GavinC
16:02:09 <yvesr> Zakim, ??P13 is me
16:02:10 <Zakim> +yvesr; got it
16:02:15 <zwu2> zwu2 has joined #rdf-wg
16:02:21 <Zakim> +Sandro
16:02:32 <Zakim> + +1.617.838.aaaa
16:02:35 <Zakim> +??P24
16:02:40 <TallTed> Zakim, aaaa is me
16:02:40 <Zakim> +TallTed; got it
16:02:46 <Zakim> +[IPcaller]
16:02:47 <AZ> Zakim, ??P24 is me
16:02:48 <Zakim> +AZ; got it
16:02:48 <Zakim> -[IPcaller]
16:02:49 <TallTed> Zakim, mute me
16:02:49 <Zakim> TallTed should now be muted
16:03:10 <Zakim> + +1.408.992.aabb
16:03:15 <Zakim> +[IPcaller]
16:03:25 <AndyS> zakim, IPCaller is me
16:03:25 <Zakim> +AndyS; got it
16:04:00 <Zakim> +cygri
16:04:03 <ivan> zakim, dial ivan-voip
16:04:03 <Zakim> ok, ivan; the call is being made
16:04:04 <Zakim> +Ivan
16:04:25 <Zakim> +??P26
16:04:35 <markus> zakim, ??P26 is me
16:04:35 <Zakim> +markus; got it
16:04:49 <davidwood> http://www.w3.org/2011/rdf-wg/meeting/2013-01-09
16:05:34 <Zakim> + +1.650.265.aacc
16:05:49 <zwu2> zakim, +1.650.265.aacc is me
16:05:49 <Zakim> +zwu2; got it
16:05:51 <davidwood> Zakim, who is here?
16:05:51 <Zakim> On the phone I see Guus, davidwood, ericP, yvesr, SteveH, Arnaud, GavinC, Sandro, TallTed (muted), AZ, +1.408.992.aabb, AndyS, cygri, Ivan, markus, zwu2
16:05:52 <zwu2> zakim, mute me
16:05:55 <Zakim> On IRC I see zwu2, gavinc, AndyS, Arnaud, markus, AZ, Guus, RRSAgent, Zakim, cygri, tbaker, FabGandon, TallTed, gkellogg, ivan, SteveH, trackbot, mischat, davidwood, manu1, manu,
16:05:55 <Zakim> ... yvesr, sandro, ericP
16:05:55 <Zakim> zwu2 should now be muted
16:06:08 <cygri> scribe: cygri
16:06:15 <Zakim> -ericP
16:06:16 <davidwood> PROPOSED to accept the minutes of the 9 January telecon: http://www.w3.org/2011/rdf-wg/meeting/2013-01-09
16:06:29 <cygri> topic: Minutes of last meeting
16:06:47 <cygri> sandro: i fixed the problem in last week's minutes
16:07:03 <cygri> RESOLVED: Accept the minutes of the 9 January telecon: http://www.w3.org/2011/rdf-wg/meeting/2013-01-09\
16:07:10 <davidwood> Review of action items
16:07:10 <davidwood> http://www.w3.org/2011/rdf-wg/track/actions/pendingreview
16:07:10 <davidwood> http://www.w3.org/2011/rdf-wg/track/actions/open
16:07:16 <cygri> s/09\/09/
16:07:20 <pfps> pfps has joined #rdf-wg
16:07:23 <cygri> topic: Review of action items
16:07:33 <gavinc> ACTION-190 was done a while ago?
16:07:33 <AZ> q+
16:07:57 <davidwood> ack AZ
16:08:15 <cygri> AZ: I had an action to complete for yesterday; it's not yet finished
16:08:24 <cygri> … drafting the document on dataset semantics
16:08:35 <cygri> … hope to have a draft by end of the week, or beginning of next
16:08:47 <cygri> … how do i publish it on the w3c server?
16:09:06 <cygri> ivan: you should be able to write to mercurial
16:09:15 <cygri> … open a new folder, edit it there, and commit it
16:09:25 <Zakim> + +33.4.92.96.aadd
16:09:33 <cygri> … once it becomes a proper W3C publication, webmaster copies it from mercurial
16:09:55 <cygri> davidwood: we will need to extend the charter soon
16:09:59 <LeeF> LeeF has joined #rdf-wg
16:10:00 <FabGandon> Zakim, +33.4.92.96.aadd is me
16:10:00 <Zakim> +FabGandon; got it
16:10:01 <cygri> … so need to get our documents in order
16:10:14 <cygri> … guus, status of the primer?
16:10:20 <cygri> Guus: not much progress since last time
16:10:39 <cygri> davidwood: we will need to get some work done in the next weeks to avoid embarrassment
16:10:43 <davidwood> Topic: RDF Concepts
16:10:45 <cygri> scribe: gavinc
16:10:46 <ivan> zakim, mute me
16:10:46 <Zakim> Ivan should now be muted
#16:10:48 <cygri> topic: Concepts
16:10:56 <cygri> davidwood: we have some open issues
16:11:18 <cygri> … i think some of them can be handled quickly, others may need more discussion
16:11:22 <davidwood> http://www.w3.org/2011/rdf-wg/track/issues/105: Graphs, datasets, authoritative representations, and content negotiation
16:11:34 <cygri> ISSUE-105?
16:11:34 <trackbot> ISSUE-105 -- Graphs, datasets, authoritative representations, and content negotiation -- open
16:11:34 <trackbot> http://www.w3.org/2011/rdf-wg/track/issues/105
16:12:35 <Zakim> +[GVoice]
16:12:45 <gavinc> cygri: This is a nebulous issue, there are multiple parts. One is the media types of graphs vs datasets
16:13:03 <gavinc> ... this came up with JSON-LD.
16:13:38 <sandro> q?
16:13:41 <gavinc> ... another is how do fragment identifiers work in datasets? If you have graph names with fragments does that mean anything?
16:14:24 <gavinc> ... I think the summary from before Christmas was that if we don't say anything that Datasets and Graphs can be used interchangeably. That's what people will do if we don't say anything anyway
16:15:00 <Zakim> +PatH
16:15:01 <gavinc> ... if we do say anything about the relationship between Datasets and Graphs we may be creeping back into defining dataset semantics which we said we wouldn't do.
16:15:19 <sandro> q+
16:15:30 <davidwood> ack sandro
16:15:51 <gavinc> davidwood: Goal is to determine if we can close this quickly
16:16:03 <gavinc> sandro: I think JSON-LD says something about this
16:16:48 <pfps> Is Sandro saying that one can always return a trig document when an rdf graph is being requested?
16:16:58 <pchampin> pchampin has joined #rdf-wg
16:17:04 <Zakim> +[IPcaller]
16:17:11 <LeeF> zakim, IPcaller is me
16:17:11 <Zakim> +LeeF; got it
16:17:18 <gavinc> davidwood: No, pfps, I don't think he said that.
16:17:19 <AndyS> Does not reflect concern on the list from Steve IIRC.
16:17:20 <LeeF> zakim, mute me please
16:17:21 <Zakim> LeeF should now be muted
16:17:29 <Zakim> +??P36
16:17:36 <sandro> PROPOSED: We advise people that when they are trying to get a graph and get a dataset instead (in Trig or JSON-LD), it's okay to just use the default graph as your graph, without issuing a warning or error.
16:17:44 <gavinc> pfps: If your asking for a graph you should never get back a dataset
16:18:04 <davidwood> ok
16:18:08 <markus> +1
16:18:27 <Zakim> + +1.415.686.aaee
16:18:33 <gkellogg> zakim, I am aaee
16:18:33 <Zakim> +gkellogg; got it
16:18:35 <SteveH> +1 to pfps
16:18:54 <gavinc> pfps: If I ask for a RDF document, and I get back something else, what should I do?
16:19:18 <gavinc> sandro: if you can't parse the TriG document you clearly you can't do anything
16:19:20 <cygri> +1
16:19:23 <yvesr> +1
16:19:27 <gavinc> ... but if you can, use the default graph
16:19:36 <davidwood> -1 from PatH (via phone)
16:19:43 <gavinc> pfps: if this is acceptable then providers will do it.
16:19:47 <davidwood> http://www.w3.org/2011/rdf-wg/track/issues/106: Relationship between RDF Concepts and RDF Semantics
16:19:51 <Arnaud> should we really specify how an error is to be handled?
16:19:52 <SteveH> you absolutely should not interpret the default graph as if it were graph content
16:19:53 <gavinc> davidwood: Moving on.
16:19:54 <gkellogg> +1
16:19:55 <pfps> -1 from me because of implementation burden
16:19:56 <cygri> ISSUE-106?
16:19:56 <trackbot> ISSUE-106 -- Relationship between RDF Concepts and RDF Semantics -- open
16:19:56 <trackbot> http://www.w3.org/2011/rdf-wg/track/issues/106
16:20:25 <gavinc> cygri: Mostly editoral. There is some content that moved between them.
16:20:47 <gavinc> ... I guess this is mostly editoral, but I would like to keep open until there is a RDF Semantics draft
16:21:10 <gavinc> davidwood: Propose to close?
16:21:42 <gavinc> cygri: I would like to keep it open, as there will be some work in the semantics draft...
16:22:01 <gavinc> PatH: I think they are all just editoral, no diffrence of opinion
16:22:45 <gavinc> cygri: I would like to have a marker in the concepts, so that I can refer to the fact that there isn't a semantics draft for RDF concepts to point to
16:22:47 <TallTed> it's useful for bookkeeping; seems reasonable to leave it and move on
16:23:11 <gavinc> cygri: if the chairs are happier with it closed and an action instead that's fine
16:23:35 <gavinc> davidwood: I prefer to close and turn into action.
16:23:44 <gavinc> cygri: I'll do that.
16:23:59 <gavinc> cygri: I raised it, I can close it?
16:24:11 <gavinc> davidwood: Prefer to close and refer to action
16:24:21 <gavinc> PatH: Ringing off
16:24:27 <Zakim> -PatH
16:24:56 <gavinc> ACTION on PatH to work with cygri to make there is no duplicated content between RDF Concepts and RDF Semantics
16:24:56 <trackbot> Error finding 'on'. You can review and register nicknames at <http://www.w3.org/2011/rdf-wg/track/users>.
16:25:03 <gavinc> ACTION PatH to work with cygri to make there is no duplicated content between RDF Concepts and RDF Semantics
16:25:03 <trackbot> Created ACTION-221 - Work with cygri to make there is no duplicated content between RDF Concepts and RDF Semantics [on Patrick Hayes - due 2013-01-23].
16:25:30 <gavinc> CLOSE ISSUE-106 Converted to ACTION-221
16:25:47 <cygri> ACTION: cygri to work with PatH to make sure there is no duplicated content between RDF Concepts and RDF Semantics
16:25:47 <trackbot> Created ACTION-222 - Work with PatH to make sure there is no duplicated content between RDF Concepts and RDF Semantics [on Richard Cyganiak - due 2013-01-23].
16:25:54 <gavinc> CLOSE ISSUE-106
16:25:54 <trackbot> Closed ISSUE-106 Relationship between RDF Concepts and RDF Semantics.
16:25:57 <cygri> ACTION-222?
16:25:57 <trackbot> ACTION-222 -- Richard Cyganiak to work with PatH to make sure there is no duplicated content between RDF Concepts and RDF Semantics -- due 2013-01-23 -- OPEN
16:25:57 <trackbot> http://www.w3.org/2011/rdf-wg/track/actions/222
16:26:30 <davidwood> http://www.w3.org/2011/rdf-wg/track/issues/107: Revised definition of blank nodes
16:26:31 <gavinc> ISSUE-107
16:26:31 <trackbot> ISSUE-107 -- Revised definition of blank nodes -- open
16:26:31 <trackbot> http://www.w3.org/2011/rdf-wg/track/issues/107
16:26:34 <AZ> q+
16:26:34 <ivan> trackbot, associate ACTION-222 with ISSUE-106
16:26:35 <trackbot> ACTION-222 (Work with PatH to make sure there is no duplicated content between RDF Concepts and RDF Semantics) associated with ISSUE-106.
16:27:00 <gavinc> cygri: We didn't resolve this in the last WD
16:27:26 <TallTed> trackbot, associate ACTION-221 with ISSUE-106
16:27:26 <trackbot> ACTION-221 (Work with cygri to make there is no duplicated content between RDF Concepts and RDF Semantics) associated with ISSUE-106.
16:27:30 <gavinc> ... next step to work with AZ to take one of the proposals and make both of us happy with it.
16:27:47 <gavinc> ack az
16:28:00 <gavinc> AZ: drafted a mail but didn't send it
16:28:12 <gavinc> ... I just have to find it and complete it in the next few days
16:28:18 <davidwood> http://www.w3.org/2011/rdf-wg/track/issues/109: What's the consequence of a literal being ill-typed?
16:28:21 <ericP> i'm getting pressure to go to another call and don't think i'm critical here
16:28:25 <gavinc> ISSUE-109
16:28:25 <trackbot> ISSUE-109 -- What's the consequence of a literal being ill-typed? -- open
16:28:25 <trackbot> http://www.w3.org/2011/rdf-wg/track/issues/109
16:28:49 <gavinc> davidwood: Lots of disciusion, no resolution
16:29:12 <gavinc> cygri: One way of doing this is making it part of the semantics. PatH said there was a way of doing that that makes sense to him.
16:29:24 <gavinc> ... there would also need to be a small edit to RDF Concepts to make it happen.
16:29:26 <gkellogg> s/disciusion/discussion/
16:29:35 <ericP> -> https://dvcs.w3.org/hg/rdf/file/tip/rdf-turtle/coverage/tests/ atomic tests for Turtle
16:29:42 <ericP> (committed after much pain)
16:29:46 <gavinc> ... it might be controversial
16:29:52 <Zakim> -ericP
16:30:12 <gavinc> ... some clarification about what applications are supposed to do with malformed typed literals? is it an error or not?
16:30:20 <davidwood> q?
16:30:24 <gavinc> pfps: If it's an error then where is the error message?
16:30:30 <gavinc> ... it's NOT an error!
16:30:43 <gavinc> ... It's perfectly fine as far as RDF is concerned
16:30:46 <gavinc> ... what's the issue?
16:30:58 <gavinc> cygri: The issue is that the datatype can't assign a value.
16:31:23 <gavinc> ... it clearly is an error, someone did something wrong publishing that data.
16:31:37 <gavinc> ... You shouldn't be publishing that kind of data.
16:31:51 <gavinc> ... is this an error or not? The current spec doesn't say one way or another
16:31:59 <gavinc> pfps: I think everything is perfectly clear.
16:32:07 <gavinc> pfps: It isn't an error.
16:32:13 <gavinc> pfps: How can it be an error?
16:32:31 <yvesr> +1 with cygri - we should be clear on that, and it should be flagged by applications
16:32:35 <gavinc> davidwood: any time the spec talks about an inconsistent graph...
16:32:48 <pchampin> q+
16:32:49 <gavinc> sandro: An error is when someone does something they aren't supposed to do.
16:32:59 <gavinc> pfps: an error is when the system barfs
16:33:02 <LeeF> If we don't agree on what an error is then it doesn't seem so productive to discuss whether this condition is an error or not :)
16:33:06 <cygri> RDF 2004 says: "Such a case, while in error, is not syntactically ill-formed."
16:33:14 <TallTed> a mistyped literal, like "abc" typed as a datetime.  that seems like an error...
16:33:27 <AndyS> Quote text.
16:33:33 <pchampin> q-
16:33:40 <davidwood> q?
16:33:58 <gavinc> sandro: Lexical space provided by a regular expression...
16:34:09 <gavinc> pfps: if you go outside, stuff happens, but it's not an error
16:34:22 <gavinc> ... then you need to have an error condition and error return
16:34:38 <gavinc> cygri: RDF Concepts says there is an error
16:34:43 <gavinc> pfps: Then RDF Concepts is wrong.
16:34:51 <gavinc> ... there should be some handling method for that error
16:35:02 <davidwood> s/RDF Concepts says there is an error/2004 RDF Concepts says there is an error/
16:35:20 <gavinc> cygri: No, it doesn't say what happens when you use a language tag that isn't a language, it doesn't say what to do if you use an IRI that isn't an IRI
16:35:34 <gavinc> pchampin: A data model can have an illtyped literal
16:36:06 <gavinc> cygri: RDF 2004 says that in the case of an ill-typed it is an error but not a syntax error
16:36:20 <gavinc> s/pchampin/pfps
16:36:30 <gavinc> pfps: It's a stupid thing to do, but not an "error"
16:36:39 <gavinc> cygri: I would like it to be an error
16:36:47 <yvesr> should we do a strawpoll on whether we think it should be an error or not?
16:36:53 <gavinc> pfps: That would be a change to the RDF Semantics
16:36:57 <pchampin> I would like it to be an inconsistency as well
16:37:11 <davidwood> yvesr, we know now that we will not agree...
16:37:20 <gavinc> sandro: Do I have to accept it? Do I have to ignore it? What do I do?
16:37:25 <gavinc> pfps: I'd like to keep doing what I do today, I accept it.
16:38:11 <davidwood> AZ, can you please put that in the minutes?
16:38:40 <sandro> so it's "in error" but not "an error" ?
16:38:40 <AZ> I think that RDF 2004 saying "... while in error, is not syntactically ill formed" is actally meaning "... while stupid as hell, is not formally an error"
16:38:57 <sandro> +1 cygri it should be an inconsistency
16:39:00 <gavinc> cygri: Semantics is strange in this regard. It would be simple to say that the presence of an ill typed literal creates an inconsistency
16:39:01 <AndyS> May not know the datatype map.  Hence need to deal with it in some sense.  Inconsistency is unknown.
16:39:06 <gavinc> pfps: That requires a change to the RDF Semantics
16:39:27 <gavinc> davidwood: I think he did propose that.
16:39:41 <gavinc> pfps: Right now we have two documents that say error where there is no error.
16:40:10 <gavinc> cygri: We need to address this, we should change this from a Concepts into a Semantics issue.
16:40:46 <gavinc> davidwood: is there a pointer to the resolution being talked about in IRC?
16:41:11 <AndyS> s/resolution/proposal/
16:41:24 <AndyS> http://lists.w3.org/Archives/Public/public-rdf-wg/2012Nov/0268.html
16:41:38 <gavinc> davidwood: PatH says he's fine with that. pfps you disagree?
16:42:02 <gavinc> pfps: If the change is going to be made, it needs to be made correctly.
16:42:27 <gavinc> pfps: That proposal doesn't make sense.
16:43:15 <gavinc> cygri: I don't know how to change that proposal into the semantics.
16:43:40 <gavinc> ... the mecanics would need to be worked out.
16:44:09 <gavinc> pfps: The proposal is kind of weird. In any D entailment... ... ... ...
16:44:50 <ivan> q+
16:44:56 <davidwood> ack ivan
16:44:57 <ivan> zakim, unmute me
16:44:58 <Zakim> Ivan was not muted, ivan
16:45:02 <gavinc> davidwood: hoping to move on...
16:45:13 <Zakim> +PatH
16:45:26 <gavinc> ivan: What pfps described for D entailment, don't we already have that for XMLLiterals?
16:45:55 <gavinc> cygri: The pure presense of an ill-typed literal doesn't cause an inconsistency. Only if there is a range statement.
16:46:28 <gavinc> ... same for all datatypes. You have to have a range statement AND an ill-typed literal to get an inconsistency.
16:47:21 <gavinc> PatH: SOmething could be consistant in RDF, but not with a specific datatype map.
16:47:23 <ivan> zakim, mute me
16:47:23 <Zakim> Ivan should now be muted
16:47:31 <davidwood> http://www.w3.org/2011/rdf-wg/track/issues/111: Should RDF Concepts define any operations on RDF datasets?
16:47:34 <gavinc> davidwood: lets move on
16:47:36 <gavinc> ISSUE-111
16:47:36 <trackbot> ISSUE-111 -- Should RDF Concepts define any operations on RDF datasets? -- open
16:47:36 <trackbot> http://www.w3.org/2011/rdf-wg/track/issues/111
16:47:56 <gavinc> davidwood: If we find a resolution this will effect ISSUE-105
16:48:03 <cygri> http://www.w3.org/2011/rdf-wg/wiki/Dataset_Operations
16:48:36 <Zakim> -GavinC
16:48:59 <path> path has joined #rdf-wg
16:49:20 <path> paste that again?
16:49:42 <SteveH> path, http://www.w3.org/2011/rdf-wg/wiki/Dataset_Operations
16:49:54 <path> ta
16:50:17 <Zakim> +GavinC
16:50:38 <gavinc_> gavinc_ has joined #rdf-wg
16:50:51 <davidwood> AZ:Tentative definition:
16:50:51 <davidwood> """
16:50:51 <davidwood> Two RDF datasets (DG1, NG1) and (DG2, NG2) are dataset-isomorphic iff:
16:50:51 <davidwood>   - DG1 and DG2 are graph-isomorphic;
16:50:51 <davidwood>   - For each (n1,g1) in NG1, there exists (n2,g2) in NG2 such that n1=n2
16:50:51 <davidwood> and g1 and g2 are graph-isomorphic;
16:50:51 <davidwood>   - For each (n2,g2) in NG2, there exists (n1,g1) in NG1 such that n1=n2
16:50:52 <davidwood> and g1 and g2 are graph-isomorphic.
16:50:52 <davidwood> """
16:51:09 <cygri> sandro: i was assuming we won't define any of this
16:51:17 <cygri> … given that we don't do semantics for datasets
16:51:18 <AZ> q+
16:51:35 <cygri> … i'm not opposed to this but seems unnecessary
16:51:37 <SteveH> +1 to sandro
16:51:48 <davidwood> q?
16:52:16 <davidwood> ack AZ
16:53:26 <cygri> sandro: i guess we should do for datasets whatever we do for graphs
16:53:27 <cygri> scribe: pchampin
16:53:28 <pchampin> I can scribe
16:53:41 <Zakim> -zwu2
16:53:57 <AZ> AZ: isomorphism is not related to semantics, it's purely about structure
16:54:03 <zwu2> Sorry, have to go to another meeting.
16:54:03 <pfps> Hmm, things are fine in Sunnyvale.
16:54:06 <davidwood> q?
16:54:12 <AZ> ...(not to be confused with equivalence)
16:54:50 <pchampin> cygri: we still need some discussion; is there anything that makes sense and would not create too much contention
16:55:24 <pchampin> path: what we could say: if you replace in a dataset a graph by an isomorphic graph, the two datasets are isomorphic
16:55:52 <pchampin> ... That extends simply the notion of graph-isomotphism to datasets.
16:56:05 <pchampin> AZ: this basically is what I proposed.
16:56:58 <pchampin> davidwood: should RDF concepts define only isomorphism on RDF datasets?
16:57:11 <pchampin> cygri: I'm not sure other operations are not useful
16:57:26 <pchampin> ... e.g. union
16:57:31 <sandro> q+
16:57:40 <SteveH> it's not uneccessary, but there's not enough experience yet!
16:57:48 <pchampin> ... but may be I can be convinced it is not necessary
16:58:02 <sandro> q-
16:58:06 <pchampin> path: my problem is that there are 3 different ways to define union,
16:58:15 <sandro> yeah -- we don't have consensus on which kind of union to use
16:58:16 <pchampin> ... so that would take a lot of time to reach a consensus
16:58:50 <yvesr> i wonder if there's room to define all three of them, and give them different operatiosn name :)
16:59:00 <sandro> q+
16:59:11 <yvesr> UNION-MERGE, UNION-REPLACE, etc.
16:59:23 <davidwood> ack sandro
17:00:00 <pchampin> sandro: defining different unions with different names would help people to have a simpler debate in the future
17:00:10 <AndyS> Blog about it.
17:00:13 <SteveH> I don't see the point of going to the effort of defining them all right now
17:00:18 <SteveH> +1 to AndyS
17:00:24 <TallTed> +1 sandro -- name and define different (non-exhaustive!) possibilities, and then leave for future debate
17:00:35 <davidwood> Review/accept Pat's solution to http://www.w3.org/2011/rdf-wg/track/actions/205: Add note to explanation of skolemization
17:00:56 <sandro> davidwood: Sandro, Yves, etc, go ahead and do that if you want to.
17:01:43 <Zakim> +[GVoice]
17:01:46 <AndyS> Remove the :
17:01:47 <davidwood> http://www.w3.org/2011/rdf-wg/track/actions/205
17:02:01 <cygri> http://www.w3.org/TR/rdf11-concepts/#section-skolemization
17:02:46 <pchampin> cygri: it has been added, then changed after comments from Peter; I'd like Peter to check the current version (URL above)
17:02:57 <pchampin> s/Peter/Pat/
17:03:59 <davidwood> close ACTION-205
17:03:59 <trackbot> Closed ACTION-205 Add note to explanation of skolemization.
17:04:19 <davidwood> http://www.w3.org/2011/rdf-wg/track/issues/23: Does going from single-graph to multi-graph require new format and new media types?
17:04:30 <cygri> topic: ISSUE-23
17:04:49 <pchampin> davidwood: isn't issue 23 a duplicate?
17:05:24 <pchampin> cygri: I think this was already discusses
17:05:41 <pchampin> davidwood: suggest to close issue 23, the answer being yes
17:06:06 <path> gavin, pedal harder.
17:06:29 <pchampin> markus: does this mean JSON-LD would require different media-types for single-graph / multi-graph?
17:07:05 <pchampin> davidwood: no
17:07:08 <TallTed> it's the shift from "one giant graph" of RDF2004 to "there are multiple graphs" of RDFnow ...
17:07:16 <AZ> +1 to close
17:07:21 <cygri> PROPOSAL: Close ISSUE-23; we are indeed defining new formats and new media types for multi-graph.
17:07:23 <SteveH> +1 to close
17:07:23 <pchampin> gkellog: will this not raise problem if RDFa decides to support multiple graphs in the future?
17:07:24 <path> +1
17:07:25 <TallTed> answer is obviously "yes" and we've been doing it
17:07:32 <davidwood> +1
17:07:34 <AZ> +1
17:07:38 <yvesr> +1
17:07:40 <gkellogg> s/gkellog/gkellogg/
17:07:40 <cygri> +1
17:07:42 <TallTed> +1
17:07:42 <sandro> understood that this is about Trig and doesn't affect RDFa or JSON-LD.
17:07:45 <sandro> +1
17:07:49 <gavinc_> +1
17:07:50 <gkellogg> +0
17:07:52 <AndyS> +1
17:07:53 <markus> +0
17:08:23 <pchampin> davidwood: we can't speculate on what RDFa will do in the future
17:08:35 <cygri> RESOLVED: Close ISSUE-23; we are indeed defining new formats and new media types for multi-graph.
17:08:37 <davidwood> TOPIC: JSON-LD
17:08:47 <davidwood> http://www.w3.org/2011/rdf-wg/track/issues/16: What is the normative serialization of the JSON grammar?
17:08:51 <path> i have to ring off.
17:09:00 <Zakim> -PatH
17:09:15 <gavinc_> err... yeah, that's about What is JSON ;)
17:09:29 <gavinc_> JSON, not RDF
17:09:31 <cygri> markus: i think this is about JSON itself
17:09:56 <cygri> … JSON-LD uses the JSON RFC as the basis
17:09:58 <AndyS> RFC4627
17:10:09 <gavinc_> +1 for RDF4627
17:10:10 <pchampin> markus: JSON-LD uses RFC 4627
17:10:36 <cygri> davidwood: propose to resolve this by saying that this WG uses JSON-LD to address the charter, which internally uses RFC4627
17:10:54 <cygri> PROPOSAL: Resolve ISSUE-16 by saying that this WG uses JSON-LD to address the charter, which internally uses RFC4627
17:11:01 <markus> +1
17:11:02 <gkellogg> +1
17:11:02 <davidwood> +1
17:11:03 <cygri> +1
17:11:08 <AZ> +1
17:11:12 <TallTed> +1
17:11:14 <yvesr> +1
17:11:18 <pchampin> +1
17:11:30 <Arnaud> +1
17:11:31 <SteveH> +1
17:11:32 <sandro> +1
17:11:36 <cygri> RESOLVED: Resolve ISSUE-16 by saying that this WG uses JSON-LD to address the charter, which internally uses RFC4627
17:11:40 <davidwood> http://www.w3.org/2011/rdf-wg/track/issues/92: Mention RDF in the JSON-LD Syntax Introduction.
17:11:57 <pchampin> markus: that's done
17:12:07 <TallTed> +1
17:12:11 <ivan> +1
17:12:14 <pchampin> PROPOSED: Resolve ISSUE-92 by saysing that it has been done
17:12:19 <gavinc_> +1 it's done
17:12:21 <SteveH> +1
17:12:21 <davidwood> +1
17:12:22 <gkellogg> +1
17:12:23 <pchampin> +1
17:12:24 <markus> +1
17:12:24 <TallTed> +1 again/still
17:12:24 <yvesr> +1
17:12:28 <AndyS> +1
17:12:46 <AZ> +1
17:12:54 <cygri> RESOLVED: Resolve ISSUE-92 by saying that it has been done
17:13:01 <davidwood> Working Group status: https://www.w3.org/2011/rdf-wg/track/products
17:13:42 <davidwood> Topic: AOB
17:13:45 <markus> q+
17:13:50 <davidwood> ack markus
17:14:25 <cygri> markus: what's the group's position on IRI vs URL
17:14:36 <ivan> zakim, unmute me
17:14:36 <Zakim> Ivan should no longer be muted
17:14:44 <ivan> q+
17:14:52 <cygri> … I propose we close the issue, and re-open it only if someone complains
17:14:53 <pchampin> markus: the JSON-LD group resolved to stick to the technically correct term IRI
17:15:03 <cygri> davidwood: chair hat off, redefining terms is dangerous
17:15:18 <AndyS> Nice idea but URL-NG is not yet ready.
17:15:21 <cygri> sandro: communities use different terms
17:15:23 <davidwood> ack ivan
17:15:31 <cygri> davidwood: IRI is not a semweb-specific term
17:15:32 <pchampin> sandro: this is not only about redefining terms; different communities use different terms
17:15:47 <ericP> there are lots of protocol RFCs to rewrite if we use "URL" to mean IRI
17:16:14 <cygri> ivan: someone proposed to use IRI, but put a comment in the document that explains the issue, stating that some people use URL to mean the same thing
17:16:36 <pchampin> ivan: we can add an editorial note stressing the fact that there is some inconsistency, but that this is not the job of this group to solve this inconsistency
17:16:59 <pchampin> sandro: the document could use "identifier" in 99% of the document,
17:17:01 <cygri> sandro: editorially, you can probably say "identifier", and just define that term somewhere as meaning IRI/URL
17:17:12 <ivan> zakim, mute me
17:17:12 <Zakim> Ivan should now be muted
17:17:19 <pchampin> ... and add at some point "identifier means IRI, but you can consider is very like URL"
17:17:28 <ivan> +1 to Gregg
17:17:51 <pchampin> gkellogg: my position is that JSON-LD should comply with the rest of the documents of this WG, using IRI
17:18:06 <cygri> sandro: stick with IRI, possibly move to URL in the future
17:18:10 <tbaker> +1 to Ivan and Gregg
17:18:16 <TallTed> we should use the correct word/acronym wherever possible, and say something like "others may use or have used URL or URI or other terms for IRI would have been correct; RDF-WG cannot unify/fix all such"
17:18:22 <cygri> s/sandro: stick/gkellog: stick/
17:18:45 <Zakim> -Guus
17:18:54 <Zakim> - +1.408.992.aabb
17:18:57 <Zakim> -gkellogg
17:18:59 <pchampin> davidwood: if you want a resolution from this WG, please ask it by mail
17:19:03 <Zakim> -SteveH
17:19:09 <Zakim> -GavinC
17:19:11 <Zakim> -Ivan
17:19:36 <Zakim> -AZ
17:19:41 <Zakim> -markus
17:19:43 <Zakim> -cygri
17:19:49 <ericP> https://dvcs.w3.org/hg/rdf/file/tip/rdf-turtle/coverage/tests/
17:19:54 <TallTed> please to post that info to the list, ericP :-)
17:20:16 <cygri> RRSAgent, make logs public
17:20:35 <Zakim> -LeeF
17:20:48 <Zakim> -Sandro
17:21:47 <Zakim> -AndyS
17:21:59 <AndyS> AndyS has left #rdf-wg
17:23:49 <pchampin> scribe: pchampin
17:23:53 <Zakim> -TallTed
17:23:56 <Zakim> -davidwood
17:23:58 <Zakim> -ericP
17:24:00 <Zakim> -Arnaud
17:24:09 <Zakim> -FabGandon
17:25:21 <Zakim> -pchampin
# SPECIAL MARKER FOR CHATSYNC.  DO NOT EDIT THIS LINE OR BELOW.  SRCLINESUSED=00000473