<jamesn> scribe: Detlev

<marcjohlic> https://www.w3.org/2002/09/wbs/35422/htmltechs20121005/results

Starting with using aria-label for link purpose

Marc: thought that aria-labelledby might be the better choice for link purpose

Jon: what about aria-descrobedBy - worls with JAWS and VO

James: that would be a diffent technique

<marcjohlic> http://www.w3.org/WAI/GL/wiki/Techniques/ARIA

Marc: Yes there are other techniques for aria-describedby

Link to SR tests: http://www.3needs.org/en/testing/code/aria-labelledby-for-link-purpose.html

target of aria-labelledby would need to be in the accessible tree (add tabindex -1)

Will update tests to reflect that

James: do the same tests for aria-describedBy

Will do

Marc: Is not the broader goal to identify sufficient techniques *provided* that AT will eventually support it?
... It's then up to authors to decide what to implement

James: agrees
... likes the example in technique, starts with same text as the link

jon: The typical use case would be the Read more.. links

Adam: could SC be met also by ajacent text?

Ajust example to not pass 2.44 without aria-label?

James: or apply to 2.4.9?
... In any case, its difficult to fail 2.4.4.

Marc: Shall we add another example or modify?

Adam: adjust slightly so there is no visual adjacency

James: move into separate paragraphs

Jon: often Read more links *arte* on new lines

Marc: happy to go forward - include UA/AT support notes?

James UA notes point out where it *doesn't* work.

Marc: Should "Start with same text" goe into the description?

James: Expected behaviour of aria-label / aria-labelledby: remove the link text and replace it with content referenced by the attribute
... inclusion in accessible tree of element referenced may only make a difference when more than one element is referenced - seems an IE quirk
... SR results should go into UA notes of Wiki

and opage be linked for otgher cases

James: move to No. 3 of the survey (Loretta being not on the call)

Failure of 4.1.2 due to using Using Abstract Roles

James: no real code examples so that people won't copy and use it

Detlev: working cde useful

James: put note into code
... referring to Detlev's comment - adding abstract roles does not meet the purpose of using ARIA

Detlev: if it doesn't do any harm, it should not be failure

James: agrees, leave it to larger audience to possibly quash it later

John & Marc: seem to agree it is a valid failure

Sorry: Jon

Jon: Validator will flag it

James: but invalidiy won't make it a failure
... using abstract roles will indiacte lack of understanding of developer

<jamesn> i'll take those comments and come back with a new technique in a week or 2

thanks you wrequickwer than I

Move to nexx topic:

Using aria-activedescendant to allow keyboard operability

Adam: was discussed last week, main debate whether it falls into 2.1.1
... results seem to indicate, it does

James: Jon, any opinion?

<marcjohlic> http://www.w3.org/WAI/GL/wiki/Using_aria-activedescendant_to_allow_keyboard_operability

James: does aria-activedescendant help meeting 2.1.1?

Jon: aria-activedescendant supports focus handling when not everything is focusable

Detlev: question about behaviour aria-activedescendant

Jon: you would have to use tabindex to make elements focusable - with aria-activedescendant you can provide info to SR users without actually firing focu events

<Loretta> (Thought I'd drop in and watch while waiting for my meeting...)

Adam: focused elements will need visible keyboard focus (2.4.7)?

Jon: Yes you need companion techniques to ensure that visual focus is handled as well

Adam: Not clear if Sufficinet Techniques can provide a11y for parts of the disabled popolation
... Is this not sufficient on its own?

James: Will be difficult to make clear what techniques have to be used in combination to create accessible content - also needs scripting

<Loretta> I think active descendant will be sufficient for 4.1.2, but not for keyboard access.

Jon: Suggestion would be to add a class attr to list item California may be named "focus" so it is clear that visual formatting is also required

Adam: agrees with Loretta that 4.1.2 is appropriate SC
... should the technique combine 2.1.1 and 4.1.2?
... aria-activedescendant refeers to properties of the element, not the focus
... Title of technique needs to changed

James: what do you think of my suggested title?
... may be "Move" ad the top does the job

James "Move" seems to create a redirect to the new one

Addam: will do if all agree with that

Adam: Change done!

<Loretta> Yes, use "Move" to rename.

Adam: The title would be in the way of combining 2.1.1 and 4.1.2

James: aria-label-techniques may not be easily combined
... If we combine them the technique and internal diffwerentiation gets too complex - separating them out makes it easier, may be merged later

<jongund> need to go

Adam: Will write up aria-activedescendant technique for 4.1.2
... will follow opinion of group whether we should combine 2.1.1 and 4.1.2 or not
... If combined will a generic title like "Using aria-activedescendant" be OK?

James: Has only one clear use case anyway

Adam: could write a new one to combine later

Or draft both options

James: Reading the intent again think that Lorettta is right about the role of focus

Adam: ACTION: Will re-write and try to combine technique to map onto 2,1,1 and 4.1.2

bye

This is scribe.perl Revision: 1.137  of Date: 2012/09/20 20:19:01  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Found Scribe: Detlev
Inferring ScribeNick: Detlev

WARNING: No "Topic:" lines found.

Default Present: +49.404.318.aaaa, James_Nurthen, Marc_Johlic, Detlev, Adam, Jon_Gunderson
Present: +49.404.318.aaaa James_Nurthen Marc_Johlic Detlev Adam Jon_Gunderson
Got date from IRC log name: 15 Oct 2012
Guessing minutes URL: http://www.w3.org/2012/10/15-html-techs-tf-minutes.html
People with action items: 

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


WARNING: No "Topic: ..." lines found!  
Resulting HTML may have an empty (invalid) <ol>...</ol>.

Explanation: "Topic: ..." lines are used to indicate the start of 
new discussion topics or agenda items, such as:
<dbooth> Topic: Review of Amy's report