Re: use cases 'DOM needs a way to get element's computed ARIA role and computed label string'

If you mean for this wiki to be broad enough to cover reflected attributes
and not just computed role / text, then could you include the use-cases
previously discussed about being able to reference a node without needing
an IDREF? Besides the web components examples, there are many other cases
when it'd be a lot more convenient to specify, e.g., aria-activedescendant
or aria-labelledby without putting a unique id on the target element.

If that's out of scope for this wiki, that's fine.

- Dominic

On Fri, Dec 19, 2014 at 1:35 AM, Steve Faulkner <faulkner.steve@gmail.com>
wrote:
>
> thanks Jon, I have set up a wiki page and added your use cases:
> https://www.w3.org/wiki/HTML/accAPI/
>
> --
>
> Regards
>
> SteveF
> HTML 5.1 <http://www.w3.org/html/wg/drafts/html/master/>
>
> On 18 December 2014 at 17:44, Gunderson, Jon R <jongund@illinois.edu>
> wrote:
>>
>>  Steve,
>>
>>
>>
>> Where do you want the use cases defined?  In the bug trackers or some
>> wiki?
>>
>>
>>
>> Use Case #1: Accessibility Evaluation Tools
>>
>> Accessibility evaluation tools like AInspector Sidebar for Firefox and
>> FAE 2.0 that are based on the OpenAjax Accessibility Evaluation Library can
>> use this API as part of the analysis of the web resources for accessibility
>> information.  Currently these tools must implement their own algorithms to
>> compute the label and role values.  The computed values can be compared to
>> markup to identify coding errors when labeling information like
>> aria-labelledby reference invalid or duplicate ids.
>>
>>
>>
>> References:
>>
>> OpenAjax Web Accessibility Evaluation Library
>>
>> http://www.openajax.org/member/wiki/Accessibility
>>
>>
>>
>> Use Case #2: “SkipTo” Javascript Utility
>>
>> “SkipTo” is a javascript utility to provide keyboard navigation  to a
>> list of headings and landmarks for use in navigating the structured content
>> of a web page.  The ability to search the DOM for role (e.g. heading,
>> landmark) and get accessible names through computed label values would make
>> this code immensely simpler (e.g. do not need to replicate code already in
>> the browser for computing these values, similar to get computedStyle).  I
>> am currently working on updating this code to include HTML5 section
>> elements with labels.
>>
>>
>>
>> Reference:
>>
>> https://github.com/paypal/skipto
>>
>>
>>
>> Jon
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> *From:* Steve Faulkner [mailto:faulkner.steve@gmail.com]
>> *Sent:* Thursday, December 18, 2014 10:25 AM
>> *To:* HTMLWG WG
>> *Cc:* Alice Boxhall; David Bolter; Dominic Mazzoni; Richard
>> Schwerdtfeger; Domenic Denicola; Alexander Surkov; W3C WAI Protocols &
>> Formats; James Craig
>> *Subject:* use cases 'DOM needs a way to get element's computed ARIA
>> role and computed label string'
>>
>>
>>
>> Can people who are interested in getting this[1] stuff defined and
>> implemented[2] provide use cases?
>>
>> [1] https://www.w3.org/Bugs/Public/show_bug.cgi?id=27294
>> [2]
>> https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/WToTEn4lveY
>>
>> --
>>
>> Regards
>>
>> SteveF
>>
>>
>>
>

Received on Friday, 19 December 2014 18:59:25 UTC