See also: IRC log
<trackbot> Date: 21 November 2013
<janina> Meeting: HTML-A11Y Task Force Teleconference
<scribe> scribe: MarkS
JS: everyone needs to agree to the new charter and deliverables.
MS: We will be following up to update TF members in the coming weeks.
JS: Not a lot to report due to
other items of focus at TPAC. Expect chaals to send feedback on
comments. Will begin preparing for exiting LC. Will be
measuring consensus from TF regarding publication path for
longdesc. Most likely early in December. By mid-december we
should have consensus from group and ducks in a row for moving
into CR.
... comment responses and testing/implementation report
... my goal to have this completed by the CSUN conference
mid-March.
JS: had a CfC with two proposals
for Canvas RE at risk features and moving them to next
revision
... there was concern from MS (from cynthia)
... wanted to continue that discussion here.
... consensus has not been reached
... have not declared a result on that just yet
<JatinderMann> http://www.w3.org/2013/11/14-html-wg-minutes.html#item12
JM: talked about this at HTML WG
F2F. Also talked to Rik from Adobe and Dominic and folks from
Apple
... like the direction this is going, tieing shadow dom to
focus, etc.
... had some concerns about how to implement
... generating momentum
... raising true concerns that we want resolved.
... think we shouldn't rush them through with Canvas
Level1
... like Rik said if we design this right, we may not even need
hit regions
... Rik has some good question
... would like to see these resolved in the Level2 spec
... end user will enjoy solid implementations that they can
actually use
... do not want to rush it out with Level1 spec since other
parts of canvas have been stable for a long time
RS: I agree that CustomFocusRing
could have better specificity.
... high contrast media queries, can help the author in drawing
rings
... the issues we are having are around hit testing, which has
been put off to Level2
... hixie did not want to document how this would work in the
browser
... the current focus ring is not going to be an issue for
Level1
... MS wrote the initial spec with us but have since been
unresponsive and coming in now and saying they are not ready to
implement this.
... my company needs this now.
... none of the questions being raised are not covered in the
spec. hixie didn't want it.
... i need to have MS at the table
... not confident that this is going to change in Level2
JM: I actually worked on the
initial canvas implementation. agree that we are a little late
to this. I'm actively interested in implementing these
features
... we are not concerned about level1 v level2, we will ship
when its ready.
... i think Rik should be involved in this since he raised
concerns as well.
... we're allowing a dev to override user settings right
now
... i think the scrolling issue is a big one and should be
worked out
... you should not be concerned that we are not ready to
implement. we just want to work out the details and get it
right. don't care whether its level1 or level2, we just want to
get it right.
JB: you mentioned that hit
testing might not be needed. that is not what I have been
hearing, which is that some form of hit testing will be needed
for this. perhaps there is information missing here.
... also sounds as though you thought it wouldn't matter that
much (which level). If a spec comes out of W3C that is not
accessible, it will matter greatly to many users.
... welcome your involvement in this, but have those two
concerns.
ach richardschwerdtfeger
ach richardschwerdtfeger
JM: I think Rik raised the hit
testing concern
... i think he would be best to comment on it.
... we have two early implementations, one from Rik and one
from Dominic and they both raised the same concerns.
... I don't think its MS as a 3rd party observer, we are
getting feedback from implementers.
...RE: whether L1 or L2. Not sure I understand this so much,
cynthia might. From my perspective, the goal is a solid
implementation for users and a clear api for developers.
... feel like we need a little more work to clean that up
... not sure why the Level matters
... developers look for interoperable implementations so they
know they can use a technology
RS: I was doing the testing with
Rik and Dominc. I understand the issue re: hit testing and
custom focus ring and can deal with that in Level2, same goes
for scrolling.
... we haven't talked about this yet
... we need something in the spec that people can start
implementing today
... pushing it off to L2 leaves low vision user with no options
for using canvas
... the API itself is not going to change for drawing a
systemFocusRing. its pretty vanilla.
... discussions with Rik re: how it gets developed is that its
not clearly defined
JS: is it good enough to provide
meaningful a11y for users who need it now.
... we need this now for mag users. imperfection for AT users
is nothing new, they will appreciate any sort of support
... looking at timeframe, we have discussed this with HTML WG,
one of chairs said we have until the end of Q1 2014 to have
implementations and think that is enough time to implement a
solid version of this.
TO: I think janina summarized it well, the disagreement is around whether the current spec meets a min bar.
<plh> hober, you're breaking up
TO: i think what became clear at
F2F was that there are two engineers that have implemented,
neither of whom thinks its ready and that every browser
implementer in the room said they would like to develop this in
L2
... something else we talked about in Shenzhen was that L2
doesn't mean 5 years from now. NO reason we can't have a quick
update to canvas that adds this feature, very soon after L1
spec
... we need to get it right
... which may mean not shipping in L1
... don't want to ship a mistake
JM: to rich's point, if we change the spec between versions, there could be compatibility issues moving forward.
JB: for the specific aspects of
the min bar of interop for getting this into L1, do we have
specific statements to what the question are?
... have we done an assessment to see what resources are needed
to get this implemented?
... if we don't focus on getting those addressed in L1, there
is not a great prognosis for getting it done in L2
... would like to see continued focus on this for L1
... may not be typical scenario for a developer, but the means
to have access is better than no access at all.
... there are already applications that are waiting for this
now, so its not all devs that will wait for total interop
RS: I'm hearing from the two co's
who have not been involved and are now asking to push it
out.
... we agreed that path could be pushed off so that we can
focus on what low vision users needed to use canvas
... Pearson is trying to get rid of flash to move to canvas and
are waiting for this functionality
... there is a pattern of continued delay when moving parts to
next version
... people need to have access today
... shouldn't be expanding the issue to include
customFocusRing
TO: interop is something that
happens after implementation. the only two engineers who have
implemented are not comfortable shipping there
implementations.
...RE: apps waiting for this, we would do a disservice to them
if we put out an imperfect product
... several issues raised on several threads
RS: has apple tried to implement this in webkit?
<JatinderMann> http://lists.w3.org/Archives/Public/public-html-a11y/2013Nov/0019.html
<JatinderMann> http://lists.w3.org/Archives/Public/public-html-a11y/2013Nov/0021.html
<JatinderMann> http://lists.w3.org/Archives/Public/public-html-a11y/2013Nov/0004.html
TO: riks code has been rejected by webkit
JB: this has been cycling for some time. if there is not progress, we need to figure out why otherwise same thing will happen for L2
TO: I think we agreed at the F2F that this needed to move to L2
JM: Rik said it, we are working on this together and getting lots of feedback, there is momentum and we should use it to our advantage. I think we should entertain the idea of a 1.1 concept.
<hober> webkit-dev thread: https://lists.webkit.org/pipermail/webkit-dev/2013-October/thread.html#25670
JM: some of the concerns are
being worked out. RE: even the naming of the methods are
confusing.
... doesn't need to be a ring, its any path that can be drawn.
anything that is confusing to developers, will hurt
adoption.
... some of our concerns are quite basic.
RS: MS helped define that API
JM: it seems as though there was
some native windows API that was using that convention, but we
should question that moving forward instead of continue an
illogical naming convention
... some of the feedback we received was that a lot of this API
is confusing and unclear. A web developer is even more likely
to be confused.
JB: lets talk timeline
... think this could be worked out in some conference
call
... timing wise, if we are looking at issues as simple as
naming, we could work these out in the number of months
remainging
JS: I have not heard anything
that indicates we cannot clear all of this up by L1
... I would like to keep this moving.
<Zakim> Judy, you wanted to inquire about timing of finding solutions, if the type of specifics are simply on the level of naming and to make a suggestion on a way to focus the
JB: its very helpful to hear at
least one of the specifics. Ted, thanks for offering to
aggregate discussion to date.
... lets look at issues and start working through them.
... if there are thornier implementation issues, lets lay them
out clearly
... I want to emphasize that I like the new energy around
this
... see if we can really get these issues nailed down and
worked out. doesn't sound like we need a year to do this.
PLH: we talked about issues, usually we have open issues or bugs somewhere. Can we formally open bugs in bugzilla?
JS: we had only one objection via email.
JB: seems like the people who discovered the issues should report them
PLH: jatinder, would you like to work with me to create this issues?
JM: yes
JS: the TF list is a good place to have this discussion
JM: I think it makes sense to have the two implementers on the call and if they believe they are willing to ship those implementations.
<plh> ACTION: plh to work with Jatinder to open issues on Canvas API [recorded in http://www.w3.org/2013/11/21-html-a11y-minutes.html#action01]
<trackbot> Created ACTION-215 - Work with jatinder to open issues on canvas api [on Philippe Le Hégaret - due 2013-11-28].
RS: we cannot tie this to a major L2 release. There are big things left to do in L2. We need to have this min bar met
TO: then lets do it in a 1.1
JB: not trivial to produce an interim spec
TO: Paul suggested we do this as an extension spec.
JS: we have not said why we can't do it in 1
<JatinderMann> I said the second sentence
JM: as an editor of L2, I don't have a problem with a 1.1.
JS: very critical that we get this in L1
RS: canvas will not work in iOS without basic support for this
JM: Would love to see this worked out so we can implement it one day
TO: agree with Jatinder
JS: this is good news
... will not get to ARIA bugs, looks like there is some
discussion on list.
TO: I can help out with that
JS: no meeting next week, back on 12/5. The list never closes.
This is scribe.perl Revision: 1.138 of Date: 2013-04-25 13:59:11 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/extension spec. as an editor of L2, I don't have a problem with a 1.1./extension spec./ Succeeded: s/agreed/agree with Jatinder/ Found Scribe: MarkS Inferring ScribeNick: MarkS Default Present: Janina_Sajka, Rich_Schwerdtfeger, Mark_Sadecki, David_MacDonald, hober, Jatinder, John_Foliot, Judy, Plh, [IPcaller], leonie Present: Janina_Sajka Rich_Schwerdtfeger Mark_Sadecki David_MacDonald hober Jatinder John_Foliot Judy Plh [IPcaller] leonie Found Date: 21 Nov 2013 Guessing minutes URL: http://www.w3.org/2013/11/21-html-a11y-minutes.html People with action items: plh[End of scribe.perl diagnostic output]