See also: IRC log
<trackbot> Date: 24 April 2014
<janina> Meeting: HTML-A11Y Task Force Teleconference
<jgunder_> http://fae20-dev.cita.illinois.edu/all-pages/1458e3b98b6fd0a8/page/1/rc/1/
<scribe> scribe: aardrian
JS: I think we are closing on remaining tweaks to SF's guidance doc before asking HTMLWG to move it forward as a Note.
<MarkS> New Draft of Standalone ALT guidance
<MarkS> Feedback from David MacDonald
JS: SF announced changes on list,
got feedback from two people not on call today.
... We believe language is good as is, dno't believe anything
left to do in spec guidance itself...
... But there are some tweaks in the Using Alt document.
... Please look at it so we can close on it ASAP.
LW: Tangentially related, DM has identified bugs against alt text doc and HTML5 spec.
JS: SF asked DM to take comments into Bugzilla, was not aware also filed against spec.
<LJWatson> https://www.w3.org/Bugs/Public/show_bug.cgi?id=25404
<MarkS> Results of Longdesc Exit Criteria CfC (passes)
MS: That is a link to results of CR Exit Criteria.
JS: Chaals suggesting both parent
working groups sign off on this, I will take up in PF.
... Don't expect any surprises.
MS: Our Exit Criteria mirror HTMLWG Exit Criteria, don't expect any problems, will start necessary documentation.
JS: One clarification I made to
Paul, so no one is surprised...
... When looking at HTML criteria for sufficiency of
implementation, most support is in AT, not necessarily in
browsers.
... Success of good service delivery relies on AT
implementation, not just UA support.
DM: No longdesc in Alt Guidance,
so issue of discoverability.
... There should be some reference in the alt text for where to
find the longdesc when not programmatically associated.
SF: I'll review it; sounds reasonable.
<MarkS> Bugs from David
JS: Just want to note that references to geography on the screen may not be a good fit.
SF: May suggest a method to identify a relationship, such as a grouping container.
AR:I know ARIA is a different spec, but should we reference ARIA in the ALT guidance doc?
AR:in the ALT document, referencing a long description. Should there be a link out to ARIA to how to programatically associate it
SF: The standalone doc is now in parity with the HTML5 ALT guidance. In the HTML spec, it does talk about using ARIA roles,so that might work as well.
JS: Because this is a recent
addition to the spec languge, we requested a landing page
pointer from the spec.
... The point of the landing page is because we are expecting
all sorts of additional related documentation.
<SteveF> http://www.w3.org/html/wg/drafts/html/master/embedded-content.html#alt
<SteveF> updated alt guidance http://rawgit.com/w3c/alt-techniques/master/index.html
SF: Also note that the same text
is in the updated Alt document.
... At this point, broken links aren't a worry, but instead
substantive changes.
DM: I agree on concern over
screen position, but think screen readers are savvy enough to
handle above/below at least?
... We've been toying with requiring a label on items with
roles.
... My other bug items were only in the Alt Text Guidance, not
the HTML doc.
SF: This is about how much text should go in an alt attribute.
DM: Versus when you transition to a longdesc.
JS: When we put up a specific number, we have issues from members.
SF: Structured text, foreign
words, etc. probably aren't a fit for alt.
... The alt becomes the accessible name that gets
announced.
DM: Agree.
LW: Guidance around the number of words in a sentence, around 25 words, might be good.
SF: If you have information to provide back-up for those notes, let me know.
DM: The example with the lamp is 107 words.
JS: I think we objected to that as too long.
SF: I think that's not in there anymore.
DM: I would stand behind a couple sentences as a limit.
JS: I think we're ending up with
a good spec.
... Reminder, HTML is looking to move forward to CR, so we need
to wrap this up.
MS: A note to be added to address
hit targeting versus WG spec.
... I drafted that and added it.
... Informed HTML editors the spec is ready to go to LC.
... Announced to list that we are available for any
questions.
... Told them we'd also be available on the call, received no
question on-list or on the call.
... Hoping we are getting a stable LC version of Canvas
ready.
RS: We may want to point out that
even though it deviates from WHATWG spec, is designed to allow
us to migrate to that spec...
... in regards to hit testing, to prevent things from
breaking.
MS: For mouse events or all work?
RS: Mouse events.
MS: Is it possible they may think our approach is sound/better and may change the WHATWG spec.
r/It is/It is
RS: I am suggesting a note to clarify this.
MS: Our note has considered Canvas Level 2, stating that we've made it to be forward-compatible should be sufficient.
RS: Ok.
MS: Do you think this would be better in the document status or right in the Hit Regions section?
SR: A note is fine, but maybe don't say "compatible with WHATWG."
SF: Suggest a note pointing to appropriate spec.
MS: Have a note pointing to WHATWG spec, may want to point to our own Level 2 spec.
LW: We had a brief call. Looking
at moving the meeting time to avoid conflicts.
... Pretty quiet at the moment. Have processed all outstanding
5.0 bugs. Several assigned to different TFs, though.
... ARIA, Media sub-teams, some individuals.
<MarkS> Current a11y Bugs
MS: I think last week we talked about pinging SF on 24679.
LW: Did that, re-opened against 5.1.
[discussion of Media Sub-Team meeting times]
<JF> Josh Meille and Owen Evans of Smith-Kettelwell (sp)
<JF> both have expressed a willingness to work with mEDIA SUB-TEAM
JS: [Media Sub-Team] looking to
create content with all different media types, add-ons.
... [Media Sub-Team] Captions, sign language, etc. so someone
has examples for testing in UAs and AT.
JS: This has been popping up since CSUN. Any updates?
MS: I think next step is to draft
guidance for creating accessible table descriptions.
... I know DM has run tests on current support, looking for
anyone else to help.
DM: I think we have to lobby
Freedom Scientific to allow aria-describedby on
<table>.
... This would apply to the overall table, not just within a
specific cell.
JS: Hearing vendor discussion, but what about changing in the spec? Timing is an issue.
JF: My recollection is that
letting summary go, LW and I were supposed to draft techniques
for WCAG Techniques.
... There a couple ways we can provided a summary outside of
ARIA. Sporadic support in the HTML5 spec.
MS: That was the gist, yes. My
proposal to DM is that we draft the example text for
HTML5.
... That section will have to be changed as details/summary are
unlikely to make it into HTML5.
... During HTMLWG F2F consensus is it won't make it.
JS: What about 5.1?
MS: It's possible.
... But we need some examples in 5.0 to describe a table.
... John, David, Leonie and I should get together.
JS: I caution the group about
time-sensitive edits are on the HTML spec, the other docs are
on their own schedules.
... So changes to HTML5 should be a primary focus given the
closing window of time.
... Soon. Very soon.
rrsagent make minutes
<MarkS> Describing a Table
MS: This is what we need to replace.
JS: Worth pointing to this on the HTMLWG list as a heads-up.
<MarkS> Wishlist for 5.1 Wiki
JS: Not much time today, but we
need to get focusing on 5.1.
... LC for 5.1 isn't that far in the future.
... Look at the wiki, see where you want to do some work.
MS: Added one more thing to
wishlist after PF editors call.
... Frustration with linking terms to definitions.