W3C

HTML Accessibility Task Force Teleconference

30 Jan 2014

See also: IRC log

Attendees

Present
Janina_Sajka, Mark_Sadecki, chaals, Adrian_Roselli, John_Foliot, paulc, Cynthia_Shelly, jaymunro, Bill_Gregory, Leonie, steveF, David_MacDonald
Regrets
Chair
Janina_Sajka
Scribe
Chaals, MarkS

Contents


<trackbot> Date: 30 January 2014

<janina> Meeting: HTML-A11Y Task Force Teleconference

Identify Scribe http://www.w3.org/WAI/PF/HTML/wiki/index.php?title=Scribe_List

<scribe> scribe: Chaals

Longdesc Update

CMN: we have some light editorial comments. Most are reasonable. Minor tweaks. Others overlap with past changes, so will likely not address those

...Paul, results of CfC in WG?

<richardschwerdtfeger> I can't call in but I can monitor. I am in customer meetings.

PC: posted to list yesterday. 1 dissenter.

<paulc> See CfC decision: http://lists.w3.org/Archives/Public/public-html-admin/2014Jan/0072.html

...otherwise positive feedback

PC: Result published yesterday. One objection, just reiterating old arguments, including "there will be a replacement real soon now" - and we have heard the "real soon" for five years or so...

... so it's on the Recommendation Track.

...from the HTML WG's point of view, it is on the REC track

JS: PF got one objection reviewing old arguments, one abstention. Voted a resolution on the teleconference. PF has approved having longdesc on the Rec track..

<paulc> PF WG decision: http://lists.w3.org/Archives/Public/public-html-admin/2014Jan/0077.html

CMN: So bottom line is Mark and I don't have more excuses of waiting for ... and have to get the thing out :)

Canvas 2D Followup

MS: Had subgroup meeting on Monday.

... We got some feedback yesterday from Mozilla who have raised some concrete concerns. Expect to talk about them Monday and see if we need to do more before we get consensus.

... Also looking at testing now.

... Looks like we have pretty good coverage for that.

JS: Getting implementer feedback that shows there is more discussion needed before we get consensus to go Last Call - but no problems for heartbeats.

PC: Sam and I understand the subgroup is trying to dot the i's and cross the t's, but that's what LC / CR are for.

... I could understand if we expected to go LC to PR. I can live with what you're doing now, but we're only going to provide so much rope. You're trying to avoid going back to Last Call again, and we are not convinced that it is necessary to worry about that.

MS: OK, but want to make sure major implementers are at least basically satisfied before we get too far ahead.

PC: OK. But let's try to get the right people into the conversation - alternatively Sam and I think we can do that in Last Call.

MS: OK. Note that we said we expected to be here in Q1, and we think we're doing OK on that.

PC: You also said end of Jan, and that is where we are...

JS: We sometimes get Mozilla in and sometimes they disappear. Is there anyone major missing?

MS: Haven't heard from Apple at all - will check with hober...

<paulc> hober is on IRC

MS: There are a couple of people RichS suggested. We'd like to hear from Apple too.

PC: Note that part of our concern is that we expect to have another 60-day exclusion period stretching Last Call, so we expect that to be the long pole...

Alt Guidance & Next Steps

JS: Think the guidance is pretty good at this point.

... I could live with what is there right now. The discussion about informing users of classes of images should maybe get picked up somewhere. Maybe add another example...

... and a counter-example - don't say 'alt="logo"' without something else.

DM: We might want to file a bug to allow ARIA-* to replace alt...

JS: Perhaps...

...don't think we want to try and push that back to 5.0 though

CS: Yes, I think opening that discussion here would be counterproductive

CMN: Right.

SF: Think I can wrap this within the next week, and get the bug resolved.

PC: It seems Janina's question is "what is needed to move this into 5.0"?

SF: As far as I know we mark the change as editorial, and so it gets moved into the next Working Draft for HTML 5.0

... if there is something more substantive we have to go through the Working Group.

... for formal approval.

PC: What's the 5.1 bug number, and how much of the alt guidance issue does this cover?

SF: This covers a single example.

<janina> It's bug 23207

PC: So is this editorial or not?

<aardrian> https://www.w3.org/Bugs/Public/show_bug.cgi?id=23207

... trying to understand the gameplan for the alt guidance material in general.

... If there are a raft of changes, are they still really editorial as a set?

... Trying to get a sense of the set of changes/bugs that the TF would like backported to 5.0 - I don't know that and don't know the schedule for it.

[Paul and Steve trying to get on the same page]

SF: The Lat guidance doc hasn't been updated for a year or so. Once we have it reflecting the guidance in the spec, it can become a Note or something. I have been concentrating on resolving the problem in the spec itself.

PC: So you're making changes to 5.1 to replace the alt guidance document?

SF: Yep.

PC: What are the set of bugs that mandate those changes?

SF: Do you want a list of those bugs?

PC: I believe the answer to the question "is the set of changes made editorial overall" depends on the collection of changes made.

... we need to be transparent to say "with all these changes, the alt guidance document can go away". Is this bug part of that set of changes?

... If so, when we notify them that we make a change we should notify that the particular change is part of a broader set.

JS: Think when Steve feels we are done, and the TF feels we are done, with getting 5.1 to have good guidance, it is appropriate to tell the WG. There are significant changes involved overall.

... And I am not sure if everything was covered by a single bug.

... Instead of Ian's old language focused on literary consistency, it is important to have the alt provide functionality. This has resulted in a bunch of changes and a fundamental shift.

<JF> +1 to what Janina is saying

... But do agree that when we go to backport the changes to 5.0 we should be clear that there is a major rewrite of the section.

SF: 95% of the work has been done. The issue would have come up regardless of whether we were making the overall changes or not.

... stuff is progressively being ported back to 5.0 as we do this.

... I state what has been moved, but haven't been making the actual shifts.

... I'll find the emails that explained what I was going to do, before I started.

... More generally, I haven't seen how merges into the CR have been communicated.

CMN: Sounds like discussion has centered around the need to communicate to WG that there are significant changes. I think we should just agree that we will need to communicate the totality of the changes

JF: Do we have a summary of the changes yet?

... The /concern/ is whether we are making a thousand little changes and somehow pretending that we're not doing anything significant

CS: We can do a diff and get the list of changes easily enough.

<paulc> I suggest Steve simply add a unique keyword to the alt related bugs so that a Bugzilla search can be used to list the bugs related to fixing the alt guidance.

<JF> +1 to Paul's suggestion

JS: We decided to fix the spec first, then clarify what is happening.

<paulc> In addition we could then use that set of changes/bugs to justify publishing the original WD of alt guidance as a Note.

PC: Just tag all the bugs with a keyword, and then get the list of bugs. Easier than creating a prose summary. When the TF is happy, we use those bugs as a rationale for publishing the alt guidance as a note.

... It would be very useful to communicate the larger set of changes we're working on, as we make the various incremental changes.

<scribe> ACTION: SteveF to produce a list of the bugs relating to the alt guidance changes [recorded in http://www.w3.org/2014/01/30-html-a11y-minutes.html#action01]

<trackbot> Created ACTION-228 - Produce a list of the bugs relating to the alt guidance changes [on Steve Faulkner - due 2014-02-06].

PC: Possible that about March HTML chairs will start soliciting input on at-risk features to be removed. So you want to get this finished and in quickly.

LJW: We have used a11y_text-alt as a tag, so that saves inventing a new one.

SF: Lots have been tagged with that

LJW: Yes, but there are bugs missing the tag.

Heartbeat Publications Prep

JS: Need a few more days to get PF agreement on Heartbeats - will probably take until Wednesday next week. Apologies for the delay.

PC: Which document?
... 5.0, 5.1, polyglot and DOM LCs?

JS: We don't have a horse in that race, so no worries.

[Turns out there was nothing to discuss here]

<paulc> http://tinyurl.com/m3pvkqd for alt related bugs

PC: Hope I did the search correct for alt bugs. Problem is that it generates 121 bugs. Doubt that covers just the alt guidance material.

Any Remaining A11yTF Bugs?

JS: Know we plan a revisit of bugs...

LJW: @@

<MarkS> https://www.w3.org/Bugs/Public/buglist.cgi?bug_status=RESOLVED&keywords=a11ytf&keywords_type=allwords&list_id=31694&product=HTML%20WG&resolution=NEEDSINFO&resolution=INVALID&resolution=WONTFIX

MS: These are bugs that we want to look through and check.

<MarkS> https://www.w3.org/Bugs/Public/show_bug.cgi?id=23284

[assigned to chaals]

<MarkS> https://www.w3.org/Bugs/Public/show_bug.cgi?id=20224

JS: Think this is a good idea...

SF: Something I have to look at.

... should be moved to 5.1

... won't be implemented in 5.0

RESOLUTION: Reopen bug 20224 on HTML 5.1

<MarkS> https://www.w3.org/Bugs/Public/show_bug.cgi?id=13728

SF: Looks like this has been fixed.

RESOLUTION: Close 13278

Rationale: It's been fixed/overtaken

<MarkS> https://www.w3.org/Bugs/Public/show_bug.cgi?id=13666

[reading through the bug]

JS: ARIA will have sub-teams, one of which (EPUB) is coming rapidly. Think this is of high interest to them

<JF> +1 to aria epub's interest of this

SF: I closed this because I defined what should be done to reopen, and nobody did anything.

JS: Think the work might get done by the EPUB people who are coming into W3C to define this kind of work.

<JF> suggest reopening and adding a new tag - epub perhaps?

... DAISY would like HTML to have something defined for this.

... they are the people who want this fixed.

LJW: From HTML we should close this. If ARIA team take it, then it isn't an HTML bug...

Summary of Action Items

[NEW] ACTION: SteveF to produce a list of the bugs relating to the alt guidance changes [recorded in http://www.w3.org/2014/01/30-html-a11y-minutes.html#action01]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.138 (CVS log)
$Date: 2014-01-31 15:12:33 $