<MikeSmith> trackbot, start meeting
<trackbot> Date: 13 August 2009
<scribe> scribe: DanC
<pimpbot> Title: {agenda} HTML WG telcon 2009-08-13 from Sam Ruby on 2009-08-12 (public-html-wg-announce@w3.org from July to September 2009) (at lists.w3.org)
<rubys> next agedum
Sam: I'd like to get the stuff blocking Ian Hickson's progress unblocked
MC: the WAI PF WG is processing last call comments in batch, since they can interact. But since this is blocking progress, perhaps we could tentatively share our response in this case
Sam: that would be great
MC: I don't have the details to hand...
Rich: there have been several related comments [related to what? scribe could use some help]
Rich: where's the best place to send our tentative response?
Sam: on public-html is fine
MC: I can collaborate with concerned parties and get that out in a few days
Rich: current [aria] design is that with the exception of @role, host language overrides.
[discussion of details of @role and other details exceeds scribe's bandwidth]
[scribe hopes this will get replayed in email]
mjs: I see 2 separate issues
here...
... with aria state conflicting with native state...
... one is w.r.t. implementation; e.g. in the case of input
type=-radio_button role=check_box, what does the assistive
tech do?
mjs: 2nd issue is w.r.t. conformance... a host language making something [non?]conforming are [orthoganal?] to the implementation issue
MC: tentatively, yes that [help? that=?] is a way we'd be willing to go...
mjs: tentative answers are
totally OK...
... specifically on @role...
... I think Ian and some others are inclined to say some values
of @role in some cases are non-conforming
<Philip> "< Hixie> e.g. <h1 role=checkbox> shouldn't be valid either, and should act like an <h1> to ATs, not a checkbox"
<Philip> (from #whatwg last night)
<msporny_> DanC, combobox on role="checkbox" should raise a validation error.
mjs gives some details regarding strong and not so strong situations...
<dsinger> I think maciej is asking that some of these conflicts at least should be conformance errors?
Rich: that [some @role cases being non-conformting] sounds, tentatively, like something we could work together on, yes.
<msporny_> so <input type="radiobutton" role="combobox" ...> should raise a validation error.
<shepazu> [that all seems reasonable and obvious to me... why would this have been blocking?]
mjs: yes, we'll be sure to get adequate review on this... but specifically, there's a prohibition on [more details that sound familiar from public-html email... about a specific ARIA constraint about host langauges overriding]
Rich: MC, I think we can take this back to the [WAI PF] WG
Rich: anybody else in that call/collaboration?
mjs: I suggest hsivonen; much of the technical detail I'm relaying come from him
Rich: many thanks for the review comments; these are helpful for crafting the next ARIA draft
<msporny_> Yes, Maciej - appreciate your comments, they have been very helpful (even if I don't agree with all of them)
DanC: so who has the ball?
MC: so I can send tentative details. and 2nd, work with macie and hsivonen and Ian
action-114 due next week
<trackbot> ACTION-114 Report progress on ARIA TF due date now next week
<rubys> http://dev.w3.org/html5/pf-summary/spec.html
<pimpbot> Title: HTML 5 (at dev.w3.org)
<DanC>(I couldn't find the relevant part of http://dev.w3.org/html5/pf-summary/spec.html , fwiw)
action-128?
<trackbot> ACTION-128 -- Cynthia Shelly to work with PF to find an owner for drafting @summary text proposal -- due 2009-08-06 -- OPEN
<trackbot> http://www.w3.org/html/wg/tracker/actions/128
<pimpbot> Title: ACTION-128 - HTML Weekly Tracker (at www.w3.org)
Cynthia: let's continue this action for a few weeks
DanC: it looks done, to me; what's left to do?
<mjs> presumably something has to be posted to public-html for this item to be done
Cynthia: get review/consensus from various people in WAI PF and HTML WG. Have we talked about the TF proposal yet?
Sam: not yet; that's on today's agenda
action-128 due next week
<trackbot> ACTION-128 Work with PF to find an owner for drafting @summary text proposal due date now next week
action-133?
<trackbot> ACTION-133 -- Richard Schwerdtfeger to develop an accessibility API and model for canvas as well as attributes to specify alternative content -- due 2009-12-17 -- OPEN
<trackbot> http://www.w3.org/html/wg/tracker/actions/133
<pimpbot> Title: ACTION-133 - HTML Weekly Tracker (at www.w3.org)
Sam: is a december timeframe OK?
Rich: I'm working on this... looking at implementation stuff...
Cynthia: help from mozilla, opera and/or mac/apple would help
dsinger: ok... I've been looking more at accessibility of audio/video, but perhaps we should bump up the priority of canvas accessibility
mjs: what would help most in particular?
Rich: I'm interested to talk to developers of canvas applications
Doug: perhaps the bespin
developers?
... things like processing.js that are just images aren't as
relevant as something like bespin
<dsinger> There are always 'tour-de-force' demonstrations (like writing an editor in Canvas), but we should probably focus on 'reasonable' uses
Cynthia: we'd also like help from somebody that knows the apple accessibility APIs
<Philip> Perhaps canvas graphing libraries are interesting
<Philip> (Canvas games presumably aren't interesting, because they're usually inherently visual and can't be non-visually accessible)
Doug: are there limits to our expectations on canvas accessibility? e.g. a shoot-em-up-game
Cynthia: yes, a review of the use cases to consider practical limitations makes sense
<kliehm> (Cannot get into telcon, is full) At PF Task Force we agreed to examine the canvas examples on Laura's wiki page and note down use cases. Bespin probably could have a shadow fallback DOM with paragraphs, list items, code, and buttons. We need to identify common cases first, then look for a solution.
<jgraham> I thought avoiding the performance penalty of DOM was a goal of bespin
<kliehm> @jgraham, speed is an issue as canvas is faster than SVG, also convenience: canvas / JavaScript is made for human developers, SVG / XML is output from machines. But keeping objects for re-use in the DOM as a memory could be an argument for developers, enhancing accessibility at the same time.
Sam: volunteers for this task force?
<msporny_> +1 for HTML Accessibility Task Force
<msporny_> (creation of, not volunteering)
Doug: I'm interested
<kliehm> +1 (and help reviewing the wiki use cases appreciated)
Sam: I gather there's plenty of support; any against?
dsinger: I don't think actual technical discussion of accessibility is drowning out other issues in public-html...
dsinger: and which IPR realm would it work under? [not sure I scribed that right]
<msporny_> I don't support the process of breaking off into a separate group either.
dsinger: there are also governance issues... would the TF be advisory? it couldn't make binding decisions because it's not the actual WG
<masinter> This is a HTML-WG telephone call, so I would assume it would be a HTML-WG task force
<rubys> @masinter: joint, per http://www.w3.org/WAI/PF/html-task-force
<masinter> @rubys I understand now, sorry
dsinger: I'm frustrated that process keeps coming up to the exclusion of progress on the technical issues
MC: [missed; help?]
mjs: I think it's fine for people to get together and hash out proposals before bringing them to the HTML WG is fine, but HTML WG decisions need to get a healthy amount of discussion in public-html...
mjs: also, it's not good to spend _too_ much time baking proposals, because that can [raise social issues] too
<Zakim> msporny_, you wanted to discuss keeping AT TF on HTML WG mailing list.
manu: I think it's fine to use the public-html mailing list [I think I missed the gist of his point]
Cynthia: with some hesitation, I feel obliged to bring up culture differences. I've been contacted by people who have posted to public-html and the response seemed like a flame
<Zakim> MichaelC, you wanted to say I expect it would be ok to use the HTML list as the task force list
MC: xtech was our original
proposal so as not to deluge public-html, but perhaps
public-html would work... I could discuss that with concerned
parties
... there's also a question of which tracker to use
<Zakim> dsinger, you wanted to admit he's thinking of something similar...
<dsinger> I have discussed with a few people having an informal get-together to talk over issues, ideas, and experiments for audio/video accessibility. We will probably try to organize this before the TPAC, so ideas that come up can be brought back to the tech. meetings.
<Zakim> msporny_, you wanted to discuss expertise not being valued (RDFa experience)
Manu: my experience is that when we engaged the HTML WG directly, that's when the bulk of the useful feedback came. So while I'm sympathetic to the flaming concerns, I don't think that [should be the overriding factor?]
<dsinger> So, I am not opposed to smaller groups getting together to discuss ideas - I am in favor!
dsinger: I'm hesitant about formalizing a task force, though I'm fine with small groups getting together to make proposals; I'm engaged in doing that myself with video/audio accessibility
<mjs> I think on canvas accessibility and video accessibility, technical discussion of the issues will be much more productive
<mjs> and posting on public-html will be a great way to recruit technical help
<MikeSmith> action-34 is not related to any of my actions
DanC: This was assigned to Lachlan for a long time; then it got assigned to me when I wasn't here; I didn't mind too much, but it's been a while now and I haven't made any progress and I'm not sure when/if I will. I'm inclined to close/withdraw
close ACTION-34
<trackbot> ACTION-34 Prepare "Web Developer's Guide to HTML5" for publication in some way, as discussed on 2007-11-28 phone conference closed
<MikeSmith> action-106?
<trackbot> ACTION-106 -- Dan Connolly to work out with co-chair and staff how to spark test suite coordination next week -- due 2009-06-30 -- PENDINGREVIEW
<trackbot> http://www.w3.org/html/wg/tracker/actions/106
<pimpbot> Title: ACTION-106 - HTML Weekly Tracker (at www.w3.org)
DanC: I
... I'm ambivalent about keeping test suite stuff in the
tracker
<scribe> ACTION: Doug look at ways to integrate test from browsers into a WG test suite [recorded in http://www.w3.org/2009/08/13-html-wg-minutes.html#action01]
<trackbot> Created ACTION-134 - Look at ways to integrate test from browsers into a WG test suite [on Doug Schepers - due 2009-08-20].
action-134 due 15 Sep
<trackbot> ACTION-134 Look at ways to integrate test from browsers into a WG test suite due date now 15 Sep
close action-106
<trackbot> ACTION-106 Work out with co-chair and staff how to spark test suite coordination next week closed
Sam: I'm willing to make an announcement about TPAC registration. Is that the next step?
Mike: yes.
Sam: OK. will do.
<masinter> action-129?
<trackbot> ACTION-129 -- Larry Masinter to insure that the versioning discussion at least touches on XHTML 2 interactions with the /1999/xhtml namespace -- due 2009-08-13 -- CLOSED
close action-129
<trackbot> ACTION-129 insure that the versioning discussion at least touches on XHTML 2 interactions with the /1999/xhtml namespace closed
<trackbot> http://www.w3.org/html/wg/tracker/actions/129
<pimpbot> Title: ACTION-129 - HTML Weekly Tracker (at www.w3.org)
Sam: mjs suggested dispositions for a number of issues...
mjs: it was a little rushed and maybe not under the best subject heading... I could make a more clear presentation of the suggestions
mjs: I think there's cruft in the issue tracker...
<DanC+1 a summary of why to close for each issue, please
<msporny_> +1 for cruft in the issue tracker, and support Maciej's efforts to close items.
<dsinger> notes that if we make a mistake, it's not exactly hard to create issues etc.
<dsinger> so as long as we close 'without prejudice' i.e. allowing people to re-open without lots of justification, we're fine
<masinter> Is there a way of insuring that the people who raised the issue in the first place have had a chance to respond to closing it?
<dsinger> perhaps the name of the raiser could be indicated on each issue, in the email?
<mjs> I can Cc the originators
<dsinger> e.g. issue-314159 (raised by Mordred) should the HTML WG kill orcs?
[discussion of mechanics]
Sam: ok, so a separate message for each message to public-html, copied to the originator and a summary on the announce list
Sam: reminder, the poll is ongoing