W3C

- DRAFT -

WCAG

23 Jul 2013

Agenda

See also: IRC log

Attendees

Present
David_MacDonald, Peter_Korn, Joshue, Michael_Cooper, Bruce_Bailey, Andrew_Kirkpatrick, James_Nurthen, hhalpin, +1.978.443.aaaa, Kathy, Ryladog, Marc_Johlic
Regrets
Kerstin_Probiesch
Chair
Andrew Kirkpatrick
Scribe
joshue, David

Contents


<AWK_> scribe: joshue

2013: https://www.w3.org/2002/09/wbs/35422/ARIATechs_June21_2013/

https://www.w3.org/2002/09/wbs/35422/20120716_im/

AWK: Lets give folks a min.

<MichaelC> scribe: Joshue

AWK: Ok, lets take up item 1

WCAG review of Techniques Task Force: ARIA techniques for 21st June

AWK: Theres the link
... Last week we did one, the first. I'd like to firstly look at the unanimous ones 5,6, 7.
... Any objection to taking them up as proposed?
... Hearing none, alright.

ACCEPTED AS PROPOSED: http://www.w3.org/WAI/GL/wiki/Using_a_WAI-ARIA_role_to_expose_the_role_of_a_user_interface_component

ACCEPTED AS PROPOSED: http://www.w3.org/WAI/GL/wiki/Using_WAI-ARIA_state_and_property_attributes_to_expose_the_state_of_a_user_interface_component

AWK: Lets discuss 7
... Lets take up number 2.

http://www.w3.org/WAI/GL/wiki/Using_aria-label

AWK: We have some questions, Lorettas not on the call. Comment number 3 and then comments on programmatic association?
... Last week did we decide that programmatic association was ok?

PK: I thought we liked the phrase but didn't want to suggest it was a quote etc.
... My fave was to make it a proper quote and use the term.

DMacD: Please explain?

PK: We could say this is programmatic association.. but not quote them out.. we may say this 'programmatic association of x with y' may mean programmatic determination so by doing this you are achieving the same objective.

<discussion on programmatic determination and association>

AWK: It looks like on last week, we just took out the quote..
... So thats it.

PK: That fine too.. but its just not, to quote Gregg, a teachable moment..

AWK: Ok
... <gets James up to speed>
... So in keep with what we did last week, we should probably remove the 'quoted' sentence.

<another small change discussed on the basis of last weeks decisions.>

AWK: Is there text on the page?

<discussion on aria-labelledby referencing a text string.>

AWK: Thats the sentence Peter..

JN: What about to provide a text label on an option such as a button?

AWK: Sure
... What do others think?

DMacD: I'm not sure we're gaining much by taking it out.

JN: I don't like this, its an accessible name not an accessible description.

DMacD: Agrees

JN: What would we call this?

DMacD: A label.

AWK: Resolving the first sentence will make the second easier.

<AWK> How about: The purpose of this technique is to provide a label for objects that can be read by assistive technology.

JN: So it's purpose is to provide a label for AT..

<discussion on verbiage>

<AWK> second sentence: The aria-label attribute provides the text label for an object, such as a button.

JN: Updates the wiki with text.

AWK: It's a quibble, I would say the label is provided 'for' the object..

JN: Ok sure.
... Sounds fine.

AWK: Can everyone see the wiki?

KW: I'm sorry, what are you guys talking about?
... Thanks

DMacD: Are the first couple of sentences redundant? Repetative?

AWK: Right

KW: In example 1, the ARIA label we should make sure that we don't have the word navigation there.

AWK: <Gives update>

DMacD: <more word smithery>

KW: <more word smithery>

DMacD: Ok

JN: Is there a gain?

+q

JOC: I agree with Kathy.

AWK: Not there yet

KW: Add the word main.

DMacD: I'm concerned about the use of main, as its a role.

KW: Primary?

JOC: I agree with DMacD

DMacD: We need to get this right. I'll test.

JN: It would be better if we were more specific.

+q

JOC: These are too generic examples for correct use of aria-label IMO

DMacD: Yes, maybe it is better to have it less generic.

JOC: A more specific example would be better.

KW: I'd also change the IDs, left nav, right nav.. etc

AWK: Or give them generic names..

<Loretta> Who is voluntering to improve the example?

JN: Lets give it a generic example..

<discussion on example>

JN: How about removing the IDs?

AWK: I was talking about the ARIA labelled one..

JN: We don't want main..

AWK: Yes

<discussion of JAW 13 and 14 and role support>

AWK: So James the IDs aren't necessary?

JN: Yes, so I'd chop them.

KW: <Discussion on primary nav etc>

JN: Putting in links makes this so much longer..

<discussion on AT support for aria-label etc>

<David> www.davidmacd.com/test/arialabel.html

DMacD: I like this.

AWK: Are we happy? Is anyone unhappy with this?
... Ok, we're done with example 1.
... Loretta did have a comment

JN: I've adressed that

KW: Should we say portlet?

JN: It is a term thats used?

DMacD: The Math ML example works great.

KW: If you have Math player installed you may get different results.

<discussion on MathML>

<and regions, roles, document, widget etc>

AWK: I have a question about the MathML role, and the use of quotes.

JN: Does it matter?

DMacD: You can in HTML5 use both.

JN: Would look better if consistent.

<jamesn> http://www.w3.org/TR/wai-aria/roles#math

AWK: Setting aside, if this is right for Maths, we may have to leave this open..

KW: This doesn't mention the aria label at all.

<discussion on use of images and Maths>

<discussion on repair techniques>

AWK: Can we resolve example 3?
... Is there an appropriate use of aria-label?

JN: It seems be appropriate based on the ARIA spec.

DMacD: Yes.

KW: But if you had Math player what would happen?

DMacD: I can test this.

JN: They need to start to cope with this.

AWK: David is testing..

JN: Is using times rather than x..

KW: It should be equivalent to what MathML would do..

JN: Each player could do this in a different way.

+q to agree that we shouldn't just focus on how one UA parses an example.

BB: It comes up in discussion a lot..

-q

<discussion on MathML support and examples>

AWK: Loretta has a question about the procedure

JN: Is it appropriate to use aria-label or a label on the page instead.

LGR: We fail someone for aria-label but they could have used some thing else?

<discussion on use of aria-label its use etc>

+q there is still confusion about when to use aria-label, vs labelledby etc

+q to say there is still confusion about when to use aria-label, vs labelledby etc

-q

<discussion on use of aria-label>

zakim unmute me

+q

LGR: We don't want to fail people.

JOC: Agreed.

AWK: Is there anything that we do want to fail use of aria-label on?

DMacD: Its quite agressive.

JN: Thats what it's supposed to do.

AWK: Are there cases?

+q to say we need to have a discussion about how and where incorrect use of ARIA would result in failures.

<Zakim> Joshue, you wanted to say we need to have a discussion about how and where incorrect use of ARIA would result in failures.

JOC: Developers aren't as familiar as the experts here on how ARIA impacts on the user experience, we need to be careful of the message we give.

<discussion on ARIA semantics trumping etc>

<jamesn> http://www.w3.org/TR/wai-aria/roles#textalternativecomputation

AWK: I'm hearing we should get rid of procedure, item number 2.. ask people to review the hierarchy in the ARIA spec so they are aware of the order that ARIA label content would be read.

JOC: +1

AWK: Is that easy

LGR: <agrees>

<Ryladog> +1

KW: I can test it.

AWK: Thats great.

LGR: The first step isn't something that can be true..

JN: OK

AWK: Last week we discussed the title..

JN: It seems similar to me..

<David> http://www.w3.org/TR/wai-aria/states_and_properties#aria-label

<David> ifthe label text is visible on screen, authors SHOULD use aria-labelledby and SHOULD NOT use aria-label.

<David> sf/if

AWK: What we want is h44?

<AWK> http://www.w3.org/TR/WCAG20-TECHS/H44.html

JN: Sounds reasonable to be

<AWK> H44: Using label elements to associate text labels with form controls

<discussion of SR support>

AWK: We should point to official documented heirarchy

JN: Yes, but it's complex
... Many won't understand and shouldn't need to read it..

DMacD: Agreed.

JN: Devs making complex widgets should read the spec.

DMacD: Agreed.

JN: Is there anything in the application developers guide on this? Michael?

MC: It could be the same thing..

<AWK> Authors should be aware that aria-label may be disregarded by assistive technologies in situations where aria-labelledby is used for the same object. For more information on the naming hierarchy please consult the ARIA specification (http://www.w3.org/TR/wai-aria/roles#textalternativecomputation).

JOC: What about ARIA in HTML5 doc? http://www.w3.org/TR/2013/WD-aria-in-html-20130214/

JN: Maybe we could ask Steve to add something.

JOC: Yes, good idea.

JN: There are subtleties.

AWK: Is my text ok?

JN: I've added it./

DMacD: It's a powerful thing.

LGR: But the label has to be the complete description.

JOC: +1 to Lorettas suggestion.

JN: Not strictly true.

<Loretta> Is aria-labelledby the only attribute that will override aria-label?

JN: You can have aria-label and aria-labelledby defined on the same element and can refer to one label as well as itself.. there are cases where this is needed.
... So there are more advanced cases.
... So there are more advanced cases.

<AWK> Scribe: David

<AWK> Authors should be aware that aria-label may be disregarded by assistive technologies in situations where aria-labelledby is used for the same object, and in general aria-label text is voiced instead of other label text, not in addition to it.

CW: developers think about aria label like the title attribute to add a bit more infomation or context

DM: Yes

<jamesn> Authors should be aware that use of aria-label will override any native naming such as alt on images or label associated with a form field using the for attribute

LGR: this warning could be described in the description section

Resolution: accepted as amended aria label

Using aria-label for link purpose

<Ryladog> +1

<AWK> #s 4,7,8 remain on this survey.

<MichaelC> trackbot, end meeting

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.138 (CVS log)
$Date: 2013-07-23 16:38:34 $

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/generic/too generic/
Succeeded: s/specifi/specific/
Succeeded: s/MathMl/MathML/
Succeeded: s/agreees/agrees/
Succeeded: s/f /if/
Found Scribe: joshue
Inferring ScribeNick: Joshue
Found Scribe: Joshue
Inferring ScribeNick: Joshue
Found Scribe: David
Inferring ScribeNick: David
Scribes: joshue, David
ScribeNicks: Joshue, David
Default Present: David_MacDonald, Peter_Korn, Joshue, Michael_Cooper, Bruce_Bailey, Andrew_Kirkpatrick, James_Nurthen, hhalpin, +1.978.443.aaaa, Kathy, Ryladog, Marc_Johlic
Present: David_MacDonald Peter_Korn Joshue Michael_Cooper Bruce_Bailey Andrew_Kirkpatrick James_Nurthen hhalpin +1.978.443.aaaa Kathy Ryladog Marc_Johlic

WARNING: Replacing previous Regrets list. (Old list: Kerstin, Probiesch)
Use 'Regrets+ ... ' if you meant to add people without replacing the list,
such as: <dbooth> Regrets+ Kerstin_Probiesch

Regrets: Kerstin_Probiesch
Agenda: http://lists.w3.org/Archives/Public/w3c-wai-gl/2013JulSep/0025.html
Got date from IRC log name: 23 Jul 2013
Guessing minutes URL: http://www.w3.org/2013/07/23-wai-wcag-minutes.html
People with action items: 

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


[End of scribe.perl diagnostic output]