19 Nov 2013


See also: IRC log


David_MacDonald, Joshue, Sailesh_Panchang, Kiran_Kaja, Katie_Haritos-Shea, Adam, Michael_Cooper, Kathy_Wahlbin, kerstin, Marc_Johlic
Bruce_Bailey, Kathleen_Anderson, Alan_Smith, Peter_Thiessen


<Joshue> https://www.w3.org/2002/09/wbs/35422/Nov5_2013/

<Joshue> https://www.w3.org/2002/09/wbs/35422/InstM12112013/

<Joshue> We would like the group to think of how we can improve WCAG - discussion

<adam_solomon> im the ipcaller

<Joshue> Scribenick:Katie

<Ryladog> scribe: Ryladog

Continuing survey for 12th of November

<MichaelC> https://www.w3.org/2005/06/tracker/users/my shows your action items across all groups

Joshue: 1.1.1 test procedure concerns

* /invite rrsagent #wai-wcag

<Joshue> http://www.w3.org/WAI/GL/wiki/Aria-Edit:_F65:_Failure_of_Success_Criterion_1.1.1_due_to_omitting_the_alt_attribute_on_img_elements,_area_elements,_and_input_elements_of_type_image

<Joshue> https://www.w3.org/2002/09/wbs/35422/Nov5_2013/results

David: We are changing the failure - which opens the door to not failing the technique
... the community thinks I need to stick by my stance

Joshua: James seemed to think that this is OK

<Kathy> +1 for David's comment

David: I am not sure he felt that way

Joshua: James did have some reticence

<kerstin> zakim ipcaller is me

Joshua: I can see sometimes where alt is un-necessary also - this is to get the others up to speed

David: Sometimes folks do not understnad

Joshua: This is a deep subject

AK: Is there agreement that we need to update F65?

Joshua: I am not sure

AK: That seems to be the important question. That is a very strong failure statement. Are we going to say that the hierarchy is important? If so we have to get rid of F65.

<Joshue> +q

Joshua: the easiest way is to remove the F65

David: Andrew, could you please elaborate?

AK: In HTML5 we are discussing this, basically the documents are saying it is OK to rely on aria-labelled (?)

<Joshue> -q

SP: I support David, and not using - tools and testers check for alt
... valididty needs to be kept in mind

<Joshue> KHS: I agree with David and Sailesh

SP: Where you cannot use HTML you need to use aria labelled-by for the image - AT uses that and alt is not. Better to do use aria described-by reather than aria labelled-by

<adam_solomon> mic [roblems

<Zakim> AWK, you wanted to say that validity is not an issue with WCAG and WCAG doesn't need to worry about training costs as much as whether it works.

Katie: I agree with David and Salesh

<Joshue> +q

AK: Validity is not a requirement for WCAG compliance - so that I do not agree with.

<adam_solomon> just wantedto say that we shouldnt keep failures because of automatic checkers, and also if we keep the alt compatibilitywe need to be compatible for all of html (fieldset legend) and all the aria techs

AK: We shouldnt say aria is good and not provide aria with a clear and consistent message for developers NOR solely based on how it has always been done
... This is a topic that gets people excited
... HTML Accessibility Task Force is our role to rehash that or refer to the good work that they did

David: Andrew it looks like you are saying that the hierarchical order that ARIA provides is ??
... All are competing for the alt description
... the word of Alt is not label - it was intended for dynamic forms

<adam_solomon> imless concerned with the original intention as with current support in at

David: I do not think that HTML A11Y TF were looking for a replacement for alt. We should speak with them. Steve Faulkner was extremely adament that this was a bad idea
... Andrew I agree with you Andrew about validation

Joshua: I do not see why there is such a flurry of arguments about this. If it turns out that there is a lot of confusion. I think this failure is just some examples of what not to do

<MichaelC> chair: Joshue_O_Connor

Joshua: this is an emotive issue

David: I do not want to be a fundamentalist about this - and I do not want to be shrill. I want to be clear that the emotive quality I do not want to put that label on myself. I want us to understand clearly what we are doing here
... 'Sleeping next to an elephant' it is not a small thing if we make this change

Joshua: We will not arbitrarily make a decisiion

Adam: Can we say BTW given the ARIA spec and available APIs here are the acceptable usage - and to explain the options

Joshua: yes something like a NOte to explain this is a good idea
... We will talk about this. Andrew what do you think is missing in this Failure techniques? I am not a massive fan of thi failure technique

AK: I agree with what David said about getting clarity from the HTML A11Y TF and the figcaption relevance is
... We need to think of ways to address the concerns in the failure
... Failure techniques are particularly powerful
... Getting clarity is necessary

Joshua: OK lets do that then
... we do not need to recreate the wheel

David: I can take an action item to

<Joshue> ACTION: Joshue to contact the HTML5 a11y TF regarding issue F65: Failure of Success Criterion 1.1.1 to harmonise our advice [recorded in http://www.w3.org/2013/11/19-wai-wcag-minutes.html#action01]

<trackbot> Created ACTION-231 - Contact the html5 a11y tf regarding issue f65: failure of success criterion 1.1.1 to harmonise our advice [on Joshue O Connor - due 2013-11-26].

Joshua: thank you all

Institutional Memory Exercise

Updated technique Using the group role to identify related form controls

<David> ACTION: David to discuss with HTML working group the WCAG WG group discussion of removing the failure of missing alt if an aria label, labelled by or title is present. [recorded in http://www.w3.org/2013/11/19-wai-wcag-minutes.html#action02]

<trackbot> 'David' is an ambiguous username. Please try a different identifier, such as family name or username (e.g., dmacdona, dtodd2).

David: Salesh sent me some code to test and I am still working on this

Joshua: Sailesh did update the example. Please let us know

Salesh: i tested the fildset legend it worked even with out role equals group. If you have role equals checkbox it works, even custom controls

David: What may not be true?

Salech: You can use fieldset legend

Institutional Memory Exercise

Joshua: I left the survey open, ther should be a few comments

<Joshue> https://www.w3.org/2002/09/wbs/35422/InstM12112013/

MC: I agree with you for the most part, I think having the wiki is avialable to add content. Some of the exersize we need to do is the push folks memoroeis

AK: I cannot think od anything beyond what i wrote in the survey
... Some of the discussion pre-dated my joining

<David> link to institutional memory?

AK: Logical is subjective

David: 1.3.2 the whole thingabout tables goes back to WCAG 1 if you has a newspaper it would read a line from the first column - line by line - when things strted to get moed around with CSS that changed things

Joshua: That is good David that should be added

David: I rarely see these failures anymore

<Joshue> https://www.w3.org/2002/09/wbs/35422/InstM12112013/

<David> is there a wiki?

James: Mobile first UI is the field just go one on top of the order - but on a larger PC - it depsned on the sizing for the order

Joshua: these discussion are useful. Let please capture hem in the wiki
... Meaningful Sequence - might not seem as relevant as it was

WCAG and the Future

Joshua: We want everyones input. What are hot buttons?
... We do not folks to want to get the wrong message. Micheal can you frame this?

MC: No
... Lets see. I am nit sure what you want to cover Josh

Joshua: OK lets talk about WCAG s it is, based on the fact that technoogies are changing

Marc: I snet a note out about 'Language of Parts" - the difference between a site or content owner. User generated content issue is importnat for today where

Joshua: You are saying who is really responsible for user generated content

David: Institutional memory
... That was discussed alot in WCAG with Greg and Loretta
... especially with images


David: the commubity is expecting us to review it and to plan
... we are still 5 years out with whatever we do expeciall wth dynamic changing of content
... How much do we need to notify AT as to what has changed
... We need a lot more work on forms stuff
... there are artifacts from Magnification to 200% we knew hat was an issue at the time

David maybe a re-arrangement

David: Use of text and images of text
... A flow chart it is going to be very difficlut

James we will be using SVG for that

David: We may require SVG when it catches up

<Joshue> KHS: Regarding UGC its about policy that organisations that allow something to go up are doing it accessibly via SLAs etc.

Katie: Please join abd encourage others to join the new W3C SVG Accessibility Community Group

Joshua: any other suggestions fo rthe future? Strengths and weaknesses

James: First of all the conformance requirments
... the ALL or NOTHING is unrealistic is out of date in todays wolrd
... Software alwats has bugs

Joshua: That only works for simple websites
... Yes it does get compliacted
... The resort to that would be we have to set a benchmark

James: Where WCAG has driven policy maker is ALL or NOTHING is a problem because we governements are haing to say your softawre must be perfect
... User generated conetxt - we have to address that - we cannot force a website to be responsiblefor what users tye in
... If the user types in a different language is problematic in that regard

David: There could be a mechanism to say I am inserting this info in French

Joshua: we should provide a structure for users to input correct content

<Joshue> http://www.w3.org/community/svga11y/

James: Where is the SVG A11Y

<Joshue> http://www.w3.org/community/svga11y/participants

MC: He felt like he was not getting enough people
... He being Doug Sheppers

Katie: It had Doug and Chaals

MC: If there is an interest to finalize this a a serious group and woukd like to particptae being difficult for industery to participate in

Joshua: We woud like to get folks in there

James: I am getting question is my work how do I use SVG or should I use canvas?

MC: There are many SVG editors

Katie: there are several SVG editing tools

Kathy: I agree with James about conformance. And campatability testing with AT - how we shoudl apppraoch that under WCAG and to put out some sort of guidance around that

Joshua: Yes that would be helpful

<kerstin> +1 for more guidance concerning compability testing with AT

Kathy: many of the QA teams they are not testing around WCAG they are just using JAWS to test - not really looking at WCAG conformance. I was involved in WCAG Eval TF
... there are issues where some techniques that dont work on Mac and VoiceOver. We should preovide some guidance

David: Institutional memory on Accessibility Support we did not want to take on as we were expecting that say in a University to dedicate a few people to be devoted to creating a Accessibility support database
... Outside of W3C technoogies is where we got into Accessibility Support

Joshua: Any other comments?

News item - upcoming WCAG EM Note from the Eval TF.

Joshua: We are going to have to review that

Summary of Action Items

[NEW] ACTION: David to discuss with HTML working group the WCAG WG group discussion of removing the failure of missing alt if an aria label, labelled by or title is present. [recorded in http://www.w3.org/2013/11/19-wai-wcag-minutes.html#action02]
[NEW] ACTION: Joshue to contact the HTML5 a11y TF regarding issue F65: Failure of Success Criterion 1.1.1 to harmonise our advice [recorded in http://www.w3.org/2013/11/19-wai-wcag-minutes.html#action01]
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.138 (CVS log)
$Date: 2013-11-19 18:42:45 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
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/Salesh/Sailesh/
Succeeded: s/workd/works/
Succeeded: s/simole/simple/
Succeeded: s/retort/resort/
Succeeded: s/wsa/was/
Succeeded: s/exoecting/expecting/
Found ScribeNick: Katie
Found Scribe: Ryladog
Inferring ScribeNick: Ryladog
WARNING: No scribe lines found matching previous ScribeNick pattern: <Katie> ...
ScribeNicks: Katie, Ryladog
Default Present: David_MacDonald, Joshue, Sailesh_Panchang, Kiran_Kaja, Katie_Haritos-Shea, Adam, Michael_Cooper, Kathy_Wahlbin, kerstin, Marc_Johlic
Present: David_MacDonald Joshue Sailesh_Panchang Kiran_Kaja Katie_Haritos-Shea Adam Michael_Cooper Kathy_Wahlbin kerstin Marc_Johlic
Regrets: Bruce_Bailey Kathleen_Anderson Alan_Smith Peter_Thiessen
Agenda: http://lists.w3.org/Archives/Public/w3c-wai-gl/2013OctDec/0106.html
Got date from IRC log name: 19 Nov 2013
Guessing minutes URL: http://www.w3.org/2013/11/19-wai-wcag-minutes.html
People with action items: david joshue

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.

[End of scribe.perl diagnostic output]