Chatlog 2012-10-04

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

13:51:56 <RRSAgent> RRSAgent has joined #rdfa
13:51:56 <RRSAgent> logging to http://www.w3.org/2012/10/04-rdfa-irc
13:51:58 <trackbot> RRSAgent, make logs world
13:51:58 <Zakim> Zakim has joined #rdfa
13:52:00 <trackbot> Zakim, this will be 7332
13:52:00 <Zakim> ok, trackbot; I see SW_RDFa()10:00AM scheduled to start in 8 minutes
13:52:01 <trackbot> Meeting: RDFa Working Group Teleconference
13:52:01 <trackbot> Date: 04 October 2012
13:57:52 <Steven> Steven has joined #rdfa
13:59:34 <Zakim> SW_RDFa()10:00AM has now started
13:59:43 <Zakim> +??P10
13:59:47 <manu1> zakim, I am ??P10
13:59:47 <Zakim> +manu1; got it
14:00:12 <Zakim> +ivan
14:00:28 <niklasl> niklasl has joined #rdfa
14:00:31 <danbri> danbri has joined #rdfa
14:02:04 <Steven> Steven has joined #rdfa
14:02:32 <Steven> zakim, who is on the call?
14:02:45 <Zakim> On the phone I see manu1, ivan
14:03:24 <Zakim> +??P38
14:03:26 <gkellogg> zakim, I am ??P38
14:03:28 <niklasl> zakim, I am ??P38
14:03:45 <Zakim> +??P41
14:03:53 <Steven> zakim, who is on the phone?
14:04:00 <Zakim> +gkellogg; got it
14:04:05 <Zakim> +Steven
14:04:08 <Zakim> sorry, niklasl, I do not see a party named '??P38'
14:04:25 <Zakim> On the phone I see manu1, ivan, gkellogg, ??P41, Steven
14:04:34 <niklasl> zakim, I am ??P41
14:04:59 <Zakim> +niklasl; got it
14:05:25 <manu1> Agenda: http://lists.w3.org/Archives/Public/public-rdfa-wg/2012Oct/0000.html
14:05:57 <manu1> scribenick: niklasl
14:06:20 <niklasl> manu: we need to add the rdf:HTML topic to the agenda
14:06:25 <manu1> Topic: ISSUE-126: Can xmlns: be reported as a warning?
14:06:31 <manu1> http://www.w3.org/2010/02/rdfa/track/issues/126
14:07:22 <niklasl> manu: Mike Smith has informed us that the validator w3c uses cannot even detect use of xmlns declarations
14:07:31 <Steven> Don't design the spec around bugs in software
14:08:23 <niklasl> … so the question is if conformance validators can report use of xmlns in HTML5 as an error
14:09:03 <niklasl> … what kinds of attribute use are illegal and "dropped" in html5?
14:09:25 <niklasl> manu: I don't see a big issue in doing that
14:10:01 <niklasl> ivan: I don't mind that the validator raise that. My question is whether RDFa processors should raise an error in general for this (in html5)?
14:10:36 <gkellogg> q+
14:10:41 <manu1> ack gkellogg
14:10:46 <niklasl> manu: we have a warning about that, it should be clearly noted in the spec.. But a processor should be able to use it if it can
14:11:19 <niklasl> gregg: isn't the difference between warning an error in practice just different types of warnings?
14:11:29 <niklasl> s/warnings/messages/
14:11:58 <niklasl> steven: this isn't about processors, just about conformance checkers
14:12:14 <niklasl> manu: so can we have errors that doesn't stop processors?
14:13:01 <niklasl> gregg: in general, I consider errors to mean that if the processors doesn't stop, it indicates that something strange may result
14:13:53 <niklasl> ivan: in this case, the logical case is to issue a warning in a processor, but use the value (according to core)
14:15:06 <niklasl> gregg: a processor using a conforming html5 processor cannot see the erroneous xmlns usage at all, so it cannot report anything
14:16:20 <niklasl> steven: I think it would be a bad idea to issue an error in a conformance checker for something that's not an error
14:16:45 <niklasl> manu: I think the requirement is to say something stronger than a warning
14:16:52 <ShaneM> ShaneM has joined #rdfa
14:16:53 <niklasl> ivan: validators MAY issue an error
14:18:26 <Zakim> + +1.612.217.aaaa
14:18:34 <ShaneM> zakim, I am aaaa
14:18:34 <Zakim> +ShaneM; got it
14:22:00 <gkellogg> zakim, who's making noise?
14:22:10 <Zakim> gkellogg, listening for 10 seconds I heard sound from the following: manu1 (9%), ivan (4%), Steven (65%)
14:23:12 <niklasl> steven: can we say this in a way to make it clear that conformance checker may report it as an error, but it's not *actually* an error...
14:23:19 <niklasl> ivan: in html5, it is an error
14:23:48 <niklasl> … xmlns is not an unknown thing in html5, it's a special, not allowed thing
14:26:55 <manu1> PROPOSAL: When an RDFa validator is processing an HTML5 document, it MAY report the use of xmlns: as an error. When an RDFa processor is processing an HTML5 document it MAY report the use of xmlns: as a warning.
14:27:02 <manu1> HTML5 spec: If the XML API doesn't support attributes in no namespace that are named "xmlns", attributes whose names start with "xmlns:", or attributes in the XMLNS namespace, then the tool may drop such attributes.
14:30:31 <manu1> In the HTML syntax, namespace prefixes and namespace declarations do not have the same effect as in XML. For instance, the colon has no special meaning in HTML element names.
14:34:28 <gkellogg> q+
14:34:41 <manu1> Topic: ISSUE-139: XHTML5 processing specifically excludes the use of xml:base
14:34:47 <manu1> http://www.w3.org/2010/02/rdfa/track/issues/139
14:34:49 <manu1> ack gkellogg
14:35:10 <niklasl> gregg: we have conformance tests for this in the test suite
14:36:03 <niklasl> … HTML IDL interfaces use this
14:36:21 <Steven> +1
14:36:25 <ShaneM> +1
14:36:33 <ivan> +1
14:36:39 <manu1> PROPOSAL: XHTML5+RDFa 1.1 MUST honor the use of xml:base to set the base URL of the document.
14:36:42 <manu1> +1
14:36:43 <gkellogg> +1
14:36:44 <niklasl> niklasl: +1
14:36:58 <Steven> +1
14:37:04 <ivan> +1
14:37:08 <manu1> RESOLVED: XHTML5+RDFa 1.1 MUST honor the use of xml:base to set the base URL of the document.
14:37:25 <Zakim> -Steven
14:37:26 <manu1> Topic: ISSUE-135: RDFa Lite and non-RDFa @rel values
14:37:33 <manu1> ISSUE-135 - https://www.w3.org/2010/02/rdfa/track/issues/135
14:38:14 <niklasl> .. http://lists.w3.org/Archives/Public/public-rdfa-wg/2012Sep/0008.html
14:38:26 <niklasl> … http://lists.w3.org/Archives/Public/public-rdfa-wg/2012Sep/0009.html, http://lists.w3.org/Archives/Public/public-rdfa-wg/2012Sep/0010.html
14:40:46 <niklasl> http://www.w3.org/2010/02/rdfa/meetings/2012-05-10#resolution_1
14:41:01 <niklasl> "If @property and @rel/@rev are on the same elements, the non-CURIE and non-URI @rel/@rev values are ignored. If, after this, the value of @rel/@rev becomes empty, then the then the processor must act as if the attribute is not present."
14:42:20 <manu1> PROPOSAL: If @property and @rel/@rev are on the same elements, the non-CURIE and non-URI @rel/@rev values are ignored. If, after this, the value of @rel/@rev becomes empty, then the then the processor must act as if the attribute is not present.
14:42:32 <ivan> +1
14:42:33 <manu1> +0.5
14:42:34 <niklasl> niklasl: +1
14:42:38 <gkellogg> +0.5
14:43:07 <manu1> RESOLVED: If @property and @rel/@rev are on the same elements, the non-CURIE and non-URI @rel/@rev values are ignored. If, after this, the value of @rel/@rev becomes empty, then the then the processor must act as if the attribute is not present.
14:44:37 <manu1> niklasl: When the tokens in @rel only contain non-CURIE or non-URI values (there are no terms in HTML5+RDFa), @property overrides @rel.
14:45:57 <ShaneM> the URI for the 'term' production in RDFa Core is http://www.w3.org/TR/rdfa-syntax/#P_term
14:46:11 <manu1> niklasl: When @rel and @property was used together, they used CURIEs, so we're okay there. This is to handle the general case of when @vocab comes in conflict with @rel/@property. There is no way to make everybody happy, this is the closest we could get.
14:46:55 <niklasl> gregg: the RDF 1.1 working group has added the datatype rdf:HTML.
14:46:57 <manu1> Topic: Addition of rdf:HTML datatype to RDFa
14:47:07 <ivan> q+
14:47:14 <niklasl> … it's very much like rdf:XMLLiteral, without the exclusive XML canonicalization
14:47:48 <manu1> ack ivan
14:47:48 <niklasl> … we should support this. If we don't, we'd diverge from the RDF 1.1 concepts, for a feature very much intended for (good for) RDFa
14:48:33 <gkellogg> http://www.w3.org/TR/rdf11-concepts/#section-html
14:48:34 <niklasl> ivan: two things of interest in new RDF concepts. On the XMLLiteral side, there is now much clearer language on that.
14:49:11 <niklasl> … and indeed, the rdf:HTML literal type.
14:49:28 <niklasl> … the literal is required to be valid HTML, which is much more liberal
14:50:05 <niklasl> … but we have a process issue. It's possible that RDFa in HTML5 would become a rec *before* RDF 1.1
14:50:22 <niklasl> … so we may not be able to have a formal reference in the spec
14:50:49 <niklasl> .. But we should add an informal section encouraging RDFa processors to implement handling of rdf:HTML literals
14:51:15 <niklasl> .. I (and Gregg?) have already implemented this
14:51:42 <niklasl> gregg: I've implemented this. There are no public test cases yet.
14:52:53 <manu1> PROPOSAL: Support the the rdf:HTML datatype in HTML+RDFa 1.1 (non-normatively for the purposes of ensuring that HTML+RDFa 1.1 is not blocked from REC by RDF Concepts).
14:52:59 <gkellogg> +1
14:52:59 <ivan> +1
14:53:00 <manu1> +1
14:53:00 <niklasl> niklasl: +1
14:53:03 <ShaneM> +1
14:53:13 <manu1> RESOLVED: Support the the rdf:HTML datatype in HTML+RDFa 1.1 (non-normatively for the purposes of ensuring that HTML+RDFa 1.1 is not blocked from REC by RDF Concepts).
14:55:42 <manu1> Topic: HTML+RDFa 1.1 spec
14:56:01 <Zakim> -ivan
14:56:46 <manu1> Manu: We're in good shape, as far as the spec is concerned, we'll get verification from Mike Smith, I'll update the spec and push out a new working draft (with the approval of the group)
# SPECIAL MARKER FOR CHATSYNC.  DO NOT EDIT THIS LINE OR BELOW.  SRCLINESUSED=00000124