Weekly XHTML2 WG Teleconference -- 21 Nov 2007

Reviews

Roland: ITS Best Practices

<Roland> ITS Best practices [Note new version]

Shane: I reviewed it, and my general feeling is "it's fine" and thanks to them for doing what we asked (making a module)

Steven: You had some remarks about ruby?

Shane: The ruby spec doesn't mention a namespace
... they see, to have added it to their namespace as well (as ours)

Steven: So there are two rubys

Shane: But that is in their spec, not this document

Steven: So you don't think there are any comments to make?

Shane: No

Roland: CSS Mobile
... who was reviewer?

Steven: Yam
... I have a vague feeling that I shoould have been given an action last week to forward Yam's review
... I will

<scribe> ACTION: Steven to forward Yam's CSS review [recorded in http://www.w3.org/2007/11/21-xhtml-minutes.html#action01]

Call times

<Roland> This questionnaire is open from 2007-11-14 to 2007-11-21.

Roland: the deadline is today

Steven: So we can make a decision now
... looking at the results, the only time that has no "No" is Wednesday 14:00UTC

Roland: Will we review this?
... in the spring?

Steven: No need, because it only makes it easier for Japanese (it is pinned to Boston time, so it becomes 13:00UTC

Roland: Propose moving call time one hour earlier.
... Objections?

<Rich> no objection

<oedipus> +1 from GJR

RESOLUTION: Call will be at 14:00 UTC forthwith

<scribe> ACTION: Steven to change teleconference slot [recorded in http://www.w3.org/2007/11/21-xhtml-minutes.html#action02]

Marking up assertions

http://lists.w3.org/Archives/Public/public-xhtml2/2007Nov/0024.html

Roland: We need to do something about the presentation of assertions in current documents

Steven: I think we should make them prettier and document them at the beginning of each document

Gregory: They are useful. I am currently investigating the best way to use CSS in these cases (to support assistive software). I will report later.
... Many screen readers can't cope with things that are only different by color

Steven: And we don't use del and ins?

Shane: No, because of validation problems

Roland: This is a semantics issue, not just styling

Rich: I agree strongly

Gregory: Separating style and content doesn't guarantee accessibility

<Roland> http://new.eic-community.org/index.php?option=com_wrapper&Itemid=441

Roland: There are other groups doing similar things
... that even include stronger constraints than schema, that can be automatically extracted

Shane: How about role?

Steven: Or RDFa?

Roland: Good idea

Shane: Is there a subject and object?

Roland: I think it can be done

Rich: In any case we need to map it to a framework for accessible software

<ShaneM> @role could easily be used to annotate the assertions instead of @class, which is what I am doing today.

Gregory: Let's continue on list

Future FtFs

Roland: So we just need a host for the second FtF. The rest is clear

1. 18-20 Feb Venice Italy (Host Alessio)

2. 16-18 June NE USA (such as NY or Boston, looking for a host)

3. xx-yy October, France (TPAC 2008)

Testing XHTML Basic 1.1

Roland: No Yam, so we will postpone

<scribe> ACTION: Steven to tell Yam about test reports [recorded in http://www.w3.org/2007/11/21-xhtml-minutes.html#action03]

Title issue (RDFa vs role)

http://htmlwg.mn.aptest.com/cgi-bin/xhtml2-issues/RoleAttrib?id=8023

Gregory: I took it to PF
... and what came back was that PF is very interested in the DC usage
... and RDFa syntax uses colonised attribute values
... which is a risk wrt HTML5

Shane: I appreciate the basic issue
... but what is the problem?

Steven: They asked for a value of @role for title
... and we replied that RDFa supports that usage
... But role has colonised values too!

<oedipus> [http://lists.w3.org/Archives/Member/w3c-wai-pf/2007OctDec/0194.html]

Steven: And I thought that the problem with HTML5 was colonised attributes not the values.

Rich: title isn't a section of a document surely?

Steven: Quite, it is a property of some content
... which is why property="dc:title" is better than role="title"

<oedipus> RDFa Sufficient Post: http://lists.w3.org/Archives/Member/w3c-wai-pf/2007OctDec/0191.html

Shane:I don't think property="dc:title" covers it

Steven:Well, if you use an @about if it is not the title of the document, but a section

Shane:Oh, ok

Gregory: I have asked them (PF) to explain more, and we'll see

Rich: Let's hold off until PF meets (later today)

Roland: So this is in progress, and we don't have to do anything more yet

TAG role comment (continued from last week)

http://lists.w3.org/Archives/Public/public-xhtml2/2007Nov/0013

<oedipus> GJR notes that "title" and RDFa is on the agenda for the 21 november 2007 PF telecon

Shane: They replied to my informal reply http://lists.w3.org/Archives/Public/www-html-editor/2007OctDec/0025.html
... the comment is not actually on the role spec, so I don't think we need to treat it as a LC comment
... it is about CURIES in the context where a URI ref might be used

<oedipus> TAG tracker issue cited in TAG comments: http://www.w3.org/2001/tag/group/track/issues/56

Shane: my epiphany yesterday is that the URI aspect of CURIEs is not interesting to me. I dont care that I can express a compact URI (e.g. in @resource or @about)

<oedipus> larger TAG issue: "Do the expected benefits of CURIEs outweigh the potential costs in introducing a third syntax for identifiers into the languages of the Web?"

Mark: Do we need to make a formal response?

Shane: Yes, it is framed as a LC comment

Mark: But there is not a lot of substance to their comment

<oedipus> TAG issue is named: "Abbreviating URIs in Web Languages"

Shane: in the role document CURIEs are not used in URI contexts, so their comment doesn't count

Steven: [argument about how IRIs do this already]

Shane: A fine argument, but we don't need to use it
... we don't use CURIEs in such positions anyway

<ShaneM> they did request: The TAG requests language to the following effect be included in the normative specification of CURIEs.

<ShaneM> "CURIEs, including safe_curies, MUST NOT be used in attribute or element

<ShaneM> content where URI content is specified in the relevant language

<ShaneM> specification."

Mark: I don't think we need to consider their argument at all, since it is just not clear enough

Shane: I think we should react to that for the role spec
... but put off the wider issue until the CURIE spec

Steven: Can't we just say "Thanks, this is what we do in role already"

Shane: We don't want to piss off the TAG

Roland: So how do we reply?

RESOLUTION: Reply to TAG that we don't use CURIEs in URI context in the role module

<scribe> ACTION: Reply to TAG about CURIEs in role [recorded in http://www.w3.org/2007/11/21-xhtml-minutes.html#action04]

Roland: They can respond on CURIEs when that spec is published

<ShaneM> publication of the updated CURIE spec is imminent

[ADJOURNED]

- DRAFT -

Weekly XHTML2 WG Teleconference

21 Nov 2007

Attendees

Contents