Chatlog 2011-12-15

From RDFa Working Group Wiki
Jump to: navigation, search

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

14:58:22 <RRSAgent> RRSAgent has joined #rdfa
14:58:22 <RRSAgent> logging to http://www.w3.org/2011/12/15-rdfa-irc
14:58:24 <trackbot> RRSAgent, make logs world
14:58:24 <Zakim> Zakim has joined #rdfa
14:58:26 <trackbot> Zakim, this will be 7332
14:58:26 <Zakim> ok, trackbot; I see SW_RDFa()10:00AM scheduled to start in 2 minutes
14:58:27 <trackbot> Meeting: RDF Web Applications Working Group Teleconference
14:58:27 <trackbot> Date: 15 December 2011
14:58:47 <Zakim> SW_RDFa()10:00AM has now started
14:58:54 <Zakim> +??P15
14:58:57 <manu1> zakim, I am ??P15
14:58:57 <manu1> Guest: Dan (danbri) Brickley
14:58:57 <Zakim> +manu1; got it
14:59:38 <niklasl> niklasl has joined #rdfa
14:59:47 <Zakim> +??P18
14:59:53 <gkellogg> zakim, I am ??P18
14:59:53 <Zakim> +gkellogg; got it
15:00:46 <Zakim> +??P24
15:00:52 <niklasl> zakim, I am ??P24
15:00:52 <Zakim> +niklasl; got it
15:01:50 <Zakim> +scor
15:06:08 <Zakim> +??P43
15:06:16 <ShaneM> zakim, ??P43 is ShaneM
15:06:17 <Zakim> +ShaneM; got it
15:07:34 <manu1> scribenick: niklasl
15:07:44 <manu1> Agenda: http://lists.w3.org/Archives/Public/public-rdfa-wg/2011Dec/0052.html
15:08:16 <niklasl> q+
15:08:55 <niklasl> https://www.w3.org/2010/02/rdfa/track/issues/119
15:09:56 <niklasl> niklasl: we're waiting for feedback on ISSUE 119 (feedback on RDFa 1.1 Lite)
15:10:31 <niklasl> manu: technically we don't need to wait for it since it's not part of what's on track for last call
15:10:35 <manu1> niklas: Are these the only remaining issues for LC? What about ISSUE-119?
15:10:47 <manu1> manu: ISSUE-119 is not a blocker for LC for RDFa Core 1.1 and XHTML+RDFa 1.1
15:11:17 <manu1> Topic: ISSUE-123: HTMLLiterals
15:12:07 <manu1> https://www.w3.org/2010/02/rdfa/track/issues/123
15:12:15 <niklasl> gregg: issue is about problems with XHTML canonicalization and that HTML "tag soup" aren't really appropriate for that
15:12:33 <niklasl> gregg: also, the RDF WG are looking into the value/need of a new HTMLLiteral
15:12:52 <niklasl> gregg: implementable by an innerHTML parser
15:13:19 <niklasl> gregg: an HTMLLiteral would be a better match for HTML+RDFa
15:13:37 <niklasl> gregg: but we probably need to wait for the RDF WG resolution
15:14:04 <niklasl> gregg: also, there is the LC issue...
15:14:29 <niklasl> manu: we could put this into the HTML+RDFa spec in a couple of months depending on the outcome from the RDF WG
15:14:44 <niklasl> manu: any objections?
15:15:36 <manu1> PROPOSAL: Ensure that language in XHTML+RDFa 1.1 does not prevent the implementation of an HTMLLiteral datatype in the future. Ensure that HTML+RDFa 1.1 is not prevented from implementing an HTMLLiteral datatype.
15:15:53 <manu1> +1
15:15:54 <niklasl> niklas: +1
15:15:57 <gkellogg> +1
15:16:11 <ShaneM> +1
15:16:12 <scor> +1
15:16:14 <manu1> RESOLVED: Ensure that language in XHTML+RDFa 1.1 does not prevent the implementation of an HTMLLiteral datatype in the future. Ensure that HTML+RDFa 1.1 is not prevented from implementing an HTMLLiteral datatype.
15:16:38 <manu1> Topic: ISSUE-124: RDFa Lite Document Conformance
15:16:44 <manu1> https://www.w3.org/2010/02/rdfa/track/issues/124
15:17:11 <niklasl> manu: there is a weird issue coming from RDFa Lite building on top of HTML+RDFa
15:17:53 <niklasl> manu: there's been a suggestion that RDFa Lite should be lower in the stack of specs (on top of RDFa 1.1 Core)
15:18:39 <niklasl> manu: the document conformance for RDFa 1.1. Lite is exactly the same as for RDFa 1.1 Core
15:19:06 <niklasl> manu: make HTML+RDFa to depend on both Lite and Core
15:19:22 <gkellogg> q+
15:19:28 <manu1> ack niklasl
15:19:33 <ShaneM> q+ to discuss last call sequence
15:19:37 <niklasl> manu: the benefits would be to be able to take RDFa Lite to LC (before HTML+RDFa)
15:20:16 <manu1> ack gkellogg
15:20:20 <niklasl> gregg: the risk is that people might restrict to implementing only Lite…
15:20:23 <manu1> ack shanem
15:20:23 <Zakim> ShaneM, you wanted to discuss last call sequence
15:20:31 <niklasl> shane: I'm worried about timing
15:21:13 <gkellogg> HTML+Lite+RDFa?
15:21:47 <niklasl> shane: is somebody upset about there not being a specific Lite document conformance?
15:21:52 <niklasl> manu: we don't know yet
15:22:28 <niklasl> manu: with this change (X)HTML+RFa would depend on RDFa Lite and not (as now) the other way around
15:22:49 <niklasl> shane: neither makes any sense; RDFa Lite is just a "profile"
15:22:56 <niklasl> q+
15:23:42 <niklasl> manu: what normative statement should we put in Lite to make it clear that it has now own document conformance
15:23:50 <manu1> ack niklasl
15:24:18 <manu1> niklas: I was wondering if a normative statement could be made like so: RDFa Lite is a usage pattern and not speak about conformance levels and technical details...
15:24:50 <gkellogg> q+ to ask what processing rules would be different for RDFa 1.1 Lite?
15:24:55 <niklasl> manu: the closest we've come to that is what Shane just said. We need to get a clear idea of what e.g. google wants
15:25:06 <niklasl> manu: I don't think we should have a normative statement there.
15:25:09 <scor> isn't a primer non-normative?
15:25:21 <niklasl> manu: but can a doc without a normative stmt go to Rec track?
15:25:46 <niklasl> shane: how knows? It's semantically "null"…
15:25:54 <scor> so, we could consider RDFa Lite like another primer...
15:26:09 <gkellogg> q-
15:26:12 <niklasl> manu: RDFa Lite should probably be a note…
15:26:36 <niklasl> q+
15:27:46 <niklasl> manu: would we want to say that an RDFa Lite document must only use RDFa Lite attributes – but an RDFa (lite) processor must process any RDFa (attributes + features)
15:28:14 <manu1> ack niklasl
15:28:41 <manu1> niklas: RDFa Lite is somewhat similar to code conventions
15:28:55 <ShaneM> q+ to say that document conformance could just say 'host langage and only those RDFa attributes mentioned in this spec'
15:29:02 <gkellogg> or features of JavaScript to avoid using
15:29:19 <manu1> ack shanem
15:29:19 <Zakim> ShaneM, you wanted to say that document conformance could just say 'host langage and only those RDFa attributes mentioned in this spec'
15:29:25 <manu1> q+ to discuss the approach
15:29:29 <gkellogg> q+
15:29:59 <manu1> ack manu1
15:29:59 <Zakim> manu1, you wanted to discuss the approach
15:30:22 <niklasl> shane: conform ant document only use the Lite subset; they are interpreted the RDFa core processing rules
15:31:18 <manu1> ack gkellogg
15:31:43 <niklasl> gregg: what is it about RDFa 1.1 lite that allows this group to publish it, but not publish HTML+RDFa 1.1
15:32:11 <ShaneM> RDFa 1.1 Lite does NOT define a host language.  So it really has no dependency upon HTML+RDFa nor XHTML+RDFa
15:32:29 <niklasl> manu: this group is chartered to update XHTML+RDFa, by publishing any number of documents
15:33:27 <niklasl> manu: but the HTML WG is in control of documents related to that. RDFa Lite is not bound to HTML, so that's ok.
15:35:28 <Zakim> -scor
15:35:58 <manu1> PROPOSAL: RDFa Lite 1.1 will make normative statements about Document Conformance, but stay silent on RDFa Processor conformance. HTML+RDFa 1.1 will depend on RDFa Core 1.1 and RDFa Lite 1.1 and will make normative statements both on Document Conformance and RDFa Processor conformance.
15:36:06 <manu1> +1
15:36:09 <gkellogg> +1
15:36:11 <niklasl> niklas: +1
15:36:13 <ShaneM> +1
15:36:26 <scor> +1
15:36:29 <manu1> RESOLVED: RDFa Lite 1.1 will make normative statements about Document Conformance, but stay silent on RDFa Processor conformance. HTML+RDFa 1.1 will depend on RDFa Core 1.1 and RDFa Lite 1.1 and will make normative statements both on Document Conformance and RDFa Processor conformance.
15:37:28 <Zakim> +scor
15:37:41 <niklasl> manu: HTML+RDFa 1.1 will use RDFa Lite to speak of document conformance: the full core and the lite subset; processing rules are the same as Core + XHTML (process all attributes plus host language requirements)
15:37:58 <niklasl> shane: you could have the same effect without referencing Lite
15:38:42 <niklasl> manu: how can someone say that their HTML+RDF documents are conformant to Lite?
15:38:56 <niklasl> shane: by saying it conforms to RDFa Lite
15:39:55 <niklasl> q+
15:40:11 <manu1> ACTION: Manu and Shane to work on RDFa Lite Document Conformance language.
15:40:11 <trackbot> Created ACTION-105 - And Shane to work on RDFa Lite Document Conformance language. [on Manu Sporny - due 2011-12-22].
15:40:14 <manu1> ack niklasl
15:40:52 <manu1> niklas: There are some things that strike a similar perspective in the HTML5 spec - HTML5 includes every old element like <font>, but discourages their use.
15:41:14 <manu1> niklas: self-closing elements, even though it isn't an XHTML5 document. Perhaps there is something in the wording that we could borrow from the HTML5 spec.
15:41:50 <niklasl> manu: good point. We need to find the language for this.
15:41:54 <manu1> Topic: Any other issues?
15:42:06 <gkellogg> q+
15:42:17 <manu1> ack gkellogg
15:43:02 <niklasl> gregg: danbri had a comment earlier today about the sense of the group's position on how we stand regarding @resource/@about and Lite
15:43:05 <manu1> q+
15:43:09 <manu1> ack manu1
15:44:00 <niklasl> manu: considering the arguments, I'm fairly opposed to the change. It might convey an instability in the spec. It's about the teaching reasons.
15:44:22 <niklasl> manu: the name @resource isn't as good as @about – we've been teaching @about.
15:44:58 <MacTed> MacTed has joined #rdfa
15:45:01 <niklasl> manu: technically I can understand why @resource might be a bit better; but that could be lost to people learning it
15:45:37 <niklasl> gregg: we should consider two things: 1 adding @resource to Lite
15:45:50 <niklasl> gregg: 2: should we remove @about
15:46:21 <niklasl> manu: I'm opposed to both: adding @resource complicates Lite; removing @about sends a message that it's not as stable as we've said it is
15:47:21 <niklasl> gregg: if the sense of the group is that we don't want to do this, we should pull back the question on feedback
15:47:23 <niklasl> q+
15:47:29 <manu1> ack niklasl
15:48:06 <manu1> niklas: It's a tricky thing, understand your position (Manu), impression of instability is illusory to me.
15:48:48 <manu1> niklas: There might be that impression, but there shouldn't be that instability. There are two different perspectives on how to use features of RDFa - neither is going away. Thinking of promoting @resource gives it a simpler shape.
15:49:34 <niklasl> manu: technically, in a vacuum, it might be simpler; but switching the mindset is much harder; and @about has more meaning
15:49:35 <gkellogg> q+
15:49:52 <manu1> ack gkellogg
15:50:45 <niklasl> gregg: I'm not sure about the confusion; the audience of Lite is people who don't know RDFa or who've perceived full RDFa as too complex anyway
15:51:21 <niklasl> manu: people will see the schema.org examples and copy them
15:51:46 <niklasl> manu: then when doing more complex stuff, they'll see a whole bunch of RDFa using @about
15:52:02 <niklasl> manu: we're not teaching a single best practise
15:52:46 <niklasl> manu: how do I convey why using @resource is better than @about in general
15:53:13 <niklasl> manu: telling people that there are two very different ways of using RDFa
15:53:16 <niklasl> q+
15:53:30 <manu1> ack niklasl
15:54:32 <manu1> niklas: Yes, important point - alluded to this proposal as not limited to RDFa Lite... most pressing part, subset of RDFa - main question: Do people think that we should re-imagine the best practice of RDFa to use @resource vs. @about?
15:55:08 <manu1> niklas: If we were to do this, we could avoid these issues, we should make the same conceptual change in RDFa Core 1.1.
15:55:49 <niklasl> manu: I agree that if we do the change we should do it all the way; but we've been teaching @about for 4 years.
15:56:25 <niklasl> manu: we don't know if people will prefer and understand this better
15:57:10 <gkellogg> q+
15:57:12 <niklasl> manu: what happens if @resource does *not* click with people is worse than the upsides of if it clicks. 
15:57:17 <manu1> ack gkellogg
15:57:22 <niklasl> gregg: good point.
15:57:37 <niklasl> gregg: the real issue is what schema.org will use in their markup.
15:57:46 <niklasl> q+
15:58:11 <manu1> q+ to discuss what schema.org might want.
15:58:14 <manu1> ack niklasl
15:58:18 <niklasl> gregg: without feedback we should remove this issue
15:58:33 <manu1> niklas: I agree re: schema.org - what are they going to want to use?
16:00:01 <manu1> ack manu1
16:00:03 <Zakim> manu1, you wanted to discuss what schema.org might want.
16:00:05 <niklasl> manu: that might be good to add to the Primer.
16:00:17 <niklasl> manu: I don't think they're gonna use @resource.
16:00:18 <danbri> (I guess someone needs to sit down and slog through converting all the microdata examples to Lite, and see how that reflects on this issue? or maybe it was done...?)
16:00:31 <niklasl> q+
16:02:10 <manu1> ack niklasl
16:02:20 <manu1> ack niklasl
16:05:00 <Zakim> -gkellogg
16:05:24 <scor> q+
16:05:25 <niklasl> manu: I would support adding it in the primer. We've hadn't had time to test the idea. It's premature to promote this new way.
16:05:37 <manu1> ack scor
16:06:04 <niklasl> scor: what's happened on the initiative on scraping the web for usage patterns?
16:06:19 <niklasl> q+
16:06:33 <manu1> ack niklasl
16:06:52 <niklasl> manu: we're waiting on commoncrawl to give us some hello-world examples
16:09:15 <niklasl> (maybe I can use clojure then. I've already done some RDFa-processing with that: https://github.com/niklasl/clj-rdfa/blob/master/src/rdfa/core.clj) :)
16:10:13 <niklasl> manu: no telecon next week.
16:10:38 <Zakim> -ShaneM
16:10:40 <Zakim> -scor
16:10:42 <Zakim> -manu1
16:10:47 <Zakim> -niklasl
16:10:47 <Zakim> SW_RDFa()10:00AM has ended
16:10:49 <Zakim> Attendees were manu1, gkellogg, niklasl, scor, ShaneM
# SPECIAL MARKER FOR CHATSYNC.  DO NOT EDIT THIS LINE OR BELOW.  SRCLINESUSED=00000189