See also: IRC log
<trackbot> Date: 09 January 2014
<MarkS> Meeting: HTML-A11Y Task Force Teleconference
<MarkS> Chair: Janina_Sajka
<MarkS> agenda
<aardrian> scribe: aardrian
JS: Expecting Chaals to post
results of last survey soon, expired on Dec-31-2013.
... Group mostly in favor of publishing stand-alone
specification.
<JF> +Q
JS: Need to put together report
to show testing results, etc.
... Still targeting to have it done for CSUN.
MS: Updates from Last Call
comment period are added to draft. Working on putting together
exit criteria.
... Hoping to put together calendar for dates for moving
longdesc forward.
JF: Offering help with any of test reporting, etc.
<paulc> 1. When will the LC bugs actually be closed: http://tinyurl.com/kaqkj4h ?
PC: There appears to be ~5 LC bugs still open, when can they be closed?
<paulc> 2. What role will the HTML WG have in moving forward?
MS: Probably can be closed now, will verify.
PC: What role does HTMLWG have in working spec forward?
JS: The answer may be that HTMLWG is group that publishes it, since TF cannot publish.
PC: Any schedule should take that into consideration.
<MarkS> Minutes from last canvas sub group meeting
MS: Had second meeting on Monday,
verified completion of actions, came to better understanding of
features.
... [Reference to notes coming from Mark, minutes already
linked]
<MarkS> Summary of Canvas sub-group Meeting
<paulc> Do we have tests that can be added to the Canvas CR test suite? Note that at the F2F we discussed Canvas testing and concluded we largely had enough results to exit CR EXCEPT for the focusring work.
PC: Are there tests for the Canvas CR test suite?
RS: We have not looked at them, we should.
JS: On procedural question, should we ask for help from Testing TF on writing tests?
MS: I can take a shot, but not efficient. Will check with Canvas sub-group to see if anyone can help.
RS: If we do this, no need to go back through CR?
PC: Reason we'd go back to LC is
to remove/add features from/to CR.
... Working group in Shenzhen felt we had enough proof of
interoperability to exit now.
... Two choices: Go back to LC as soon as sub-group has
flattened bugs; delay going back to LC if we have evidence to
avoid CR.
RS: Don't know how we avoid CR -- building implementations in parallel.
PC: Without
tests/implementations, then right path might be to go back to
LC with notes...
... on establishing proof of interoperability and then can come
out of LC.
... Would ask sub-group for its thoughts on timeframe.
... Maybe comes out of LC early Feb, could come out of CR end
of March. Speculating.
... It might help if I attend next week's meeting to answer
questions in person.
RS: With ARIA CR, were done with implementations, had to wait on IP period. May not be an issue here. That's why I asked.
PC: Might be referring to
disclosure requirement, will need to check.
... Since spec has been in LC once, does this apply? Will
check.
JM: Could ask implementers if they want to help, work out test cases.
<JatinderMann> http://greweb.me/glsl-transition/example/
JM: Have found 2 sites of real-world examples of controls in Canvas.
<JatinderMann> https://personal.teamlab.com/
JM: One example uses Canvas as a control for the bezier easing function, the other is an example of real world text editor in Canvas.
JS: Bug squashing happened last week. Three bugs remaining.
JS: Hoping to do something on remaining 3 in call on Monday.
<paulc> There are only 11 HTML5 CR bugs remaining: http://tinyurl.com/mzf5l5u
SF: One member does not agree with one of the bugs. Would be good to have his buy-in.
RS: Don't need unaninimity.
<paulc> https://www.w3.org/Bugs/Public/show_bug.cgi?id=19277 is marked RESOLVES WONTFIX. If someone disagrees they should re-open the bug.
PC: All other browser makes support decision, don't agree with Alexander.
SF: Wants him to understand risks, particularly around hidden/unhidden elements in tree.
RS: Do we need to hold up
standardization process for one person who doesn't get
it?
... Wants Firefox (Alexander) to get it, but it's difficult
enough to get everyone in the room to discuss.
<SteveF> @hidden https://www.w3.org/Bugs/Public/show_bug.cgi?id=23371
SF: Argument from Alex: "We don't override disable, why should we override hidden?"
PC: Wrong link to bug in minutes,
different from link provided by SF.
... Argument about the bug is erroneous because it's the wrong
bug.
<paulc> There are only 11 HTML5 CR bugs remaining: http://tinyurl.com/mzf5l5u
PC: Earlier in call, dropped in
link to remaining bugs. Providing again.
... Vast majority of them will be accessibility related.
... When chairs met on Monday was look at HTML5 CR bugs, need
to give selves deadline just as done for Canvas bugs.
... Expecting at least 3-4 of non-a11y bugs to be related later
this week, earlier next week, leaving a11y outstanding.
JS: May need to take a call for consensus on 23371.
PC: 19277, there has been a lot
of discussion since marked as resolved.
... If anyone disagrees, then they should re-open 19277.
RS: Looking at list, which should be taken up in ARIA call on Monday?
PC: Why not at least 3 in the agenda for this meeting?
RS: Because there are lots of bugs in here.
PC: Went through CR bugs one at a
time in Shenzhen. Other bugs nominally have someone on
point.
... Suggestion is to concentrate on the three noted, then we
can look at what's left.
RS: Sees Janinna is not on point for any of them.
PC: The three in the agenda are correct. Some details were hammered out in Shenzhen.
RS: Will put those three on Monday's call, will join Monday's call.
JS: Let's see if we can get Alexander on call as well.
<SteveF> https://www.w3.org/Bugs/Public/show_bug.cgi?id=23371#c5
SF: Would be useful for Rich to
look at Alex's last comment on 23371 and respond.
... Just closed a bug that wasn't necessary to block CR. Not
one of the three on Janina's list.
<SteveF> closed this bug https://www.w3.org/Bugs/Public/show_bug.cgi?id=20730
SF: Bug #20730
JS: Are we done with image map discussion?
<MarkS> Discussion thread start
SF: Boils down to some saying no alt text needed for image in the map.
<MarkS> spec reference
SF: Updated example in spec in response to review, mainly Jukka and Leif.
<SteveF> http://www.w3.org/html/wg/drafts/html/master/embedded-content-0.html#image-maps-0
<paulc> Per previous discussion, Steve's "closed a bug" is documented in http://lists.w3.org/Archives/Public/public-html-bugzilla/2014Jan/0091.html and is bug https://www.w3.org/Bugs/Public/show_bug.cgi?id=20730
SF: From list: Shouldn't be
identifying image as image with an alt since image map role is
to provide navigation.
... Would like to get other opinions.
CS: Some image maps are navigational, some presentation.
SF: In the example, two links on a geographical map.
LW: Agree with Cynthia. Sometimes
image is important, sometimes not. Hard to qualify that
subtlety.
... Would rather know that image is there so, as user, she can
decide.
JF: Thinks image always requires
some kind of alt.
... Feels example is correct because map is more than two areas
in links.
SF: Can you please respond on list?
<SteveF> Discussion thread start
JS: Also the question of have we sufficiently tweaked the alt guidance so we can sign off?
SF: Couple bugs around relatively
minor issues outstanding.
... Feedback should be coming from WAI.
JS: Duly reminded I need to
respond, and will do that.
... Do we need to have a consensus call?
SF: Nothing controversial in additions, just examples and text.
<SteveF> last 2 examples : http://www.w3.org/html/wg/drafts/html/master/embedded-content-0.html#images-of-text
SF: Need to get people to look it over, get consensus.
<LJWatson> http://tinyurl.com/ckoq2cd
LW: Still a handful of bugs assigned to text alternative team, marked resolved, need to be reviewed by sub-teams.
JS: Goal for closing by CSUN.
Will probably run a call for consensus using snapshot.
... Mark, let's make discussion point for our call. Review and
make sure it's closed.
SF: Started to add to the head section of elements, info/pointers to appropriate ARIA references.
JS: Noted on recent PF call, people were excited about it and HTML 5.1.
SF: Trying to get feedback from the list.