Chatlog 2011-03-17

From RDFa Working Group Wiki
Jump to: navigation, search

See CommonScribe Control Panel, original RRSAgent log and preview nicely formatted version.

13:45:22 <RRSAgent> RRSAgent has joined #rdfa
13:45:22 <RRSAgent> logging to http://www.w3.org/2011/03/17-rdfa-irc
13:45:24 <trackbot> RRSAgent, make logs world
13:45:24 <Zakim> Zakim has joined #rdfa
13:45:26 <trackbot> Zakim, this will be 7332
13:45:26 <Zakim> ok, trackbot; I see SW_RDFa()10:00AM scheduled to start in 15 minutes
13:45:27 <trackbot> Meeting: RDFa Working Group Teleconference
13:45:27 <trackbot> Date: 17 March 2011
13:45:33 <manu> Agenda: http://lists.w3.org/Archives/Public/public-rdfa-wg/2011Mar/0076.html
13:45:34 <manu> Chair: Manu
13:45:37 <manu> Present: Benjamin, Ivan, Manu, Steven, Nathan, Knud, Toby
13:45:37 <manu> Regrets: Shane
13:56:34 <Benjamin> Benjamin has joined #rdfa
14:00:16 <Zakim> SW_RDFa()10:00AM has now started
14:00:23 <Zakim> +??P9
14:00:26 <manu> zakim, I am ??P9
14:00:26 <Zakim> +manu; got it
14:01:10 <manu> zakim, mute me
14:01:11 <Zakim> sorry, manu, muting is not permitted when only one person is present
14:01:36 <Zakim> +??P16
14:01:39 <Zakim> + +49.631.205.75.aaaa
14:01:45 <ivan> zakim, aaaa is me
14:01:45 <Zakim> +ivan; got it
14:01:53 <Benjamin> zakim, I am aaaa
14:01:54 <Zakim> sorry, Benjamin, I do not see a party named 'aaaa'
14:02:13 <Benjamin> zakim, I am +49.631.205.75.aaaa
14:02:13 <Zakim> sorry, Benjamin, I do not see a party named '+49.631.205.75.aaaa'
14:02:45 <Benjamin> zakim, I am ??P16
14:02:45 <Zakim> +Benjamin; got it
14:06:01 <Steven> Steven has joined #rdfa
14:06:14 <Steven> zakim, phone steven-617
14:06:14 <Zakim> ok, Steven; the call is being made
14:06:15 <Zakim> +Steven
14:06:41 <Knud> Knud has joined #rdfa
14:08:00 <Zakim> + +3539149aabb
14:08:51 <Zakim> sorry, Knud, I do not know which phone connection belongs to you
14:08:59 <ivan> scribenick: ivan
14:09:01 <Knud> zakim, I am aabb
14:09:01 <Zakim> +Knud; got it
14:09:05 <Knud> zakim , mute me
14:09:09 <Steven> zakim, aabb is Knud
14:09:09 <Zakim> sorry, Steven, I do not recognize a party named 'aabb'
14:09:17 <ivan> zakim, mute Knud
14:09:17 <Zakim> Knud should now be muted
14:09:35 <manu> Topic: ISSUE-120 HTML5 and RDFa and CURIEs
14:09:37 <manu> http://www.w3.org/2002/09/wbs/40318/issue- 120-objection-poll/results
14:09:57 <ivan> manu: if anyone here is also in the HTML5 WG...
14:09:58 <Knud> no
14:10:07 <ivan> ... put in your comment
14:10:21 <Knud> I don't think anyone in DERI is in it
14:10:37 <ivan> ... anything else to discuss? We have new comments from Henri Sivonen and James Graham that we should review. Toby should respond to HTML WG, I'll remember to send a tweet out about responding.
14:11:55 <manu> Topic: Overlap with RDF Working Group
14:12:13 <ivan> manu: currently we are having an in-depth discussion about json and rdf in json
14:12:17 <ivan> ... what it means, etc
14:12:33 <ivan> ... question came up on the alignment on the rdf(a) API
14:12:46 <ivan> ... there is a call there to parse
14:12:54 <ivan> ... we would like to have the same
14:13:07 <ivan> ... it would be good to have the two groups to get together and talk about it
14:13:18 <ivan> ... maybe this group should also discuss this
14:13:33 <ivan> ... there are a number of people in the RDF wg who are interested in the api
14:13:43 <ivan> ... we may want to lean on them to join this group
14:14:04 <manu> Ivan: I would think that we should wait a bit, there will be a F2F in mid-April in A'dam
14:14:26 <manu> Ivan: I think that one of the goals will be to clarify the directions for both groups - the discussion in RDF WG is to understand the direction.
14:15:05 <manu> Ivan: All of the discussions are around that. I hope that the RDF WG will have a clearer view after mid-April - hopefully after that the synergies between the groups will be known a bit better
14:15:55 <manu> Ivan: We'll have a clearer view of the intersection between the two groups. There are voices in the RDF WG for whom a JSON serialization is more N-Triples based in JSON. If that viewpoint prevails, there will be little overlap between that work and the work we're doing here.
14:16:06 <manu> Ivan: The F2F is on the 13th and 14th of April
14:16:52 <ivan> manu: do you know if the group went the more 'object based json round', then the right people to do that work is not around in the WG
14:17:02 <ivan> ... is that true?
14:18:42 <manu> Ivan: It is true that we have you, Nathan, Thomas Steiner and possibly Toby as people that are involved in JSON and JavaScript development.
14:19:48 <manu> Ivan: Other than that, we may not have the proper people to complete this work. Finding out the direction and target community seems to be the most difficult thing for RDF in JSON.
14:20:20 <manu> Ivan: Once that is decided, I have the impression that the JSON serialization is not rocket science...
14:20:45 <manu> Ivan: If we go JSON-LD - there are disagreements on the syntax, but nothing huge.
14:21:08 <manu> Ivan: If we go the JTriples route, that's the Talis direction and disagreements on syntax are not that huge either
14:21:21 <ivan> Are there any other comments on how best to proceed with this work?
14:21:23 <Benjamin> no
14:21:30 <manu> Topic: ISSUE-85: Projections and Property Groups
14:21:36 <trackbot> ISSUE-85 -- Determine whether both Projections and PropertyGroups are necessary -- open
14:21:36 <trackbot> http://www.w3.org/2010/02/rdfa/track/issues/85
14:21:42 <manu> http://www.w3.org/2010/02/rdfa/track/issues/85
14:23:02 <ivan> manu: based on our work we have experienced that having two different ways to describe the same confuses people and it is more complex than necessary
14:23:18 <ivan> ... I believe we can do a unified version which is fairly clean
14:23:31 <ivan> ... one concept that desribe RDF data in Javascript
14:23:54 <ivan> ... to access things via API or JSON should be the same
14:23:55 <ivan> q+
14:24:31 <manu> Ivan: I'm concerned, how does JSON come into this picture.
14:24:56 <Zakim> +[IPcaller]
14:25:08 <webr3> Zakim, i am IPcaller
14:25:08 <Zakim> ok, webr3, I now associate you with [IPcaller]
14:26:13 <manu> Ivan: We have a RDFa API, that has a parser... I have a parser API, why do we have a RDF in JSON API as well?
14:26:47 <ivan> manu: the general idea that people would not have to have an heavyweight api for json
14:26:51 <manu> var obj = rdfInJson.parse(jsonText);
14:27:39 <manu> { "name": "Ivan" }
14:27:43 <manu> obj.name
14:27:49 <manu> obj.name == "Ivan"
14:28:32 <manu> var obj = rdfa.getItemsBySubject("http://example.org/people#ivan", { ... MAPPING ...});
14:28:48 <manu> obj.name
14:29:21 <manu> var obj = rdf.buildObject("http://example.org/people#ivan", { ... MAPPING ...});
14:29:23 <manu> obj.name
14:30:28 <manu> Ivan: Ok, I understand - if the RDF WG decides that the only thing that they standardize is roughly the Talis JSON/RDF or JTriples
14:30:44 <manu> Ivan: rdfInJson.parse() should still be possible
14:31:20 <manu> Ivan: If I get my text as TURTLE, I should be able to get back the same type of "obj.name" stuff.
14:32:01 <manu> Manu: Yes, but what do these objects look like?
14:32:17 <manu> Ivan: JSON is a serialization, it's no different from TURTLE or RDF/XML
14:33:30 <manu> obj.name
14:33:40 <manu> obj.get("http://example.com/vocab#foo");
14:33:45 <manu> obj.get("ex:foo");
14:34:19 <webr3> Nathan: PropertyGroup and Projection should be the same interface, with no .properties attribute, get() should be a setter, and we need .getSubject on there - to all extents a simple object with getSubject, nothing more
14:34:31 <manu> Manu: the focus is on the object that is returned - what is it? Is it a Projection, is it a Property Group, is it a Graph?
14:34:35 <webr3> s/setter/getter
14:35:40 <ivan> manu: nathan, is the way that you say this will happen, if a person gives a mapping mechanism then you can create names on the object
14:35:52 <ivan> ... and otherwise you can get them with a get mechanism
14:37:38 <ivan> nathan: all the properties with the object should be the shortest name that the rdf mechanism and the curie mechanism gives you
14:37:56 <ivan> manu: that was the same kind of thinking
14:38:04 <ivan> ... what happens if there is no environment
14:38:34 <ivan> ... would they be able to pass a mapping (eg, foaf name should be mapped to name)
14:38:43 <ivan> ... people could also use the full URI
14:39:02 <manu> jsonText = { "name" : "Nathan" }
14:39:22 <manu> MAPPING = { "http://xmlns.com/0.1/foaf/name" : "name" }
14:39:38 <manu> var obj = api.toObject(jsonText, MAPPING);
14:39:45 <manu> Then you should be able to do:
14:40:02 <manu> obj.name and obj["http://xmlns.com/0.1/foaf/name"]
14:40:13 <manu> assume that you have an RDF environment
14:40:36 <manu> that environment specifies => { "foaf": "http://xmlns.com/0.1/foaf/" }
14:40:44 <manu> then you should be able to do this:
14:40:58 <manu> obj.name and obj["foaf:name"] and obj["http://xmlns.com/0.1/foaf/name"]
14:41:28 <ivan> manu: nathan, is that what you were thinking about?
14:41:34 <ivan> nathan: something like that
14:41:51 <ivan> ... I was thinking of some sort of a profile object
14:42:02 <webr3> graph.toObject or api.toObject(graph) - with an optional attribute which you can pass in a profile
14:42:10 <ivan> ... and that would replace the mapping that would be optional
14:44:33 <manu> Ivan: I'd like to see this stuff in writing, to understand it better
14:45:53 <manu> http://www.w3.org/2010/02/rdfa/sources/rdfa-api/Overview-src.html
14:46:38 <manu> Nathan: We should clean up the editors drafts and start conversations from there.
14:46:44 <manu> Manu: I think that's a good idea.
14:48:42 <manu> Topic: ISSUE-86: Prefixes and Terms
14:48:50 <manu> http://www.w3.org/2010/02/rdfa/track/issues/86
14:50:36 <manu> Manu: Do we want to simplify this - instead of prefix and term, we use "mapping"?
14:51:06 <Benjamin> I calle dit IriMap in my mail
14:52:16 <Benjamin> s/calle dit/called it
14:52:37 <ivan> q+
14:52:40 <manu> Ivan: We don't have the concept of relative URIs... we shouldn't have them.
14:52:44 <manu> ack ivan
14:53:26 <manu> Ivan: One question/comment - I don't know how the other RDF environment handles this issue - it's the same, you have namespaces, but you can use it by itself - you can use it as a term or a prefix
14:53:47 <manu> Ivan: I don't know what Redland and the other systems do - we may want to follow those.
14:54:41 <ivan> manu: the only thing that may block us is if nathan hit an issue while implementing
14:55:55 <ivan> ... and checking other systems
14:56:02 <manu> ACTION: Nathan to contact RDF library authors to see how they handle term/prefix mappings.
14:56:02 <trackbot> Created ACTION-70 - Contact RDF library authors to see how they handle term/prefix mappings. [on Nathan Rixham - due 2011-03-24].
14:59:58 <manu> tinkster, please note your position/objection here: http://www.w3.org/2002/09/wbs/40318/issue-120-objection-poll/results
15:00:28 <tinkster> Yes, I will do.
15:00:28 <Zakim> -ivan
15:01:27 <Zakim> -Steven
15:01:31 <Zakim> -Knud
15:01:33 <Zakim> -[IPcaller]
# SPECIAL MARKER FOR CHATSYNC.  DO NOT EDIT THIS LINE OR BELOW.  SRCLINESUSED=00000163