ARIA mapping - HTML A11Y TF

25 May 2010

See also: IRC log


Rich, Janina, Stevef, Janina_Sajka, Michael_Cooper, [Microsoft]


<richardschwerdtfe> I am on folks

<richardschwerdtfe> not hearing anything

<richardschwerdtfe> ZZZZZZZZZZZZZZZZZZZZZZZZzzzzzzzzzzzzzzzzzz

<richardschwerdtfe> :-)

<cyns> http://cid-8d111c666c91bb15.skydrive.live.com/browse.aspx/Public

<scribe> scribe: janina

We're looking at Cynthia's documents ...

cyns: focus on commands and memory
... this area of the html5 spec seems stable. not changed much in a year
... seems 'command' is 'menuitem'

now looking at second table ...

rich: within a toolbar?

cyns: within a menu tag

rich: separators between them? how rendered
... we currently don't support groups within menus, but ...

cyns: we should do something with this

rich: our recent discussion re groups with a toolbar would be similar

cyns: aria toolbars are nestable?

rich: no
... too many people complained about grouping within toolbar

cyns: html5 also supports nested menus

rich: how does one nav?

cyns: like file

rich: so like 'haspopup'

cyns: i'll look at haspopup
... seems like having elements whose role changes based on context is different from anything we've had before, need to check with ie team on that
... not bad, just new

rich: i looked at new control, see lots for browsers to do

cyns: also labels and legends ...
... html doesn't have href as command -- see button and link as interchangeable
... my questions
... why inputimage not there, input button, link, why not there

rich: yes

cyns: not sure why we need states, attributes plus facets?

rich: why?

cyns: i don't understand it
... seems facet is more or less attribute, and state is attribute value, but not clear mapping
... all of this re-raises whether we map to aria and rely on aria to map OS a11y apis, or
... whether we map directly to a11y apis. Seems sometimes results might be different

rich: seems we should start funneling questions to html-wg now, not wait

cyns: the more i look at facets, the more i think we may need direct mappings to a11y apis
... ex: icon for a menu item

rich: we need that?

cyns: it's in the spec

rich: not just the text for that icon?

cyns: think it may be easier to map to uia than to aria

rich: but wouldn't fit all platforms

cyns: suggest we advise direct mappings from browsers, where appropriate

rich: agree, but
... ex: meter has associated values that may not map
... we can pass off additional properties

cyns: browsers should do best fit mapping

rich: will always be something additional not covered

<Stevef> command is the abstraction behind menu items, buttons, and links.

<Stevef> http://www.whatwg.org/specs/web-apps/current-work/multipage/commands.html#concept-facet

<cyns> http://dev.w3.org/html5/spec/interactive-elements.html#the-command

rich: how do these go into the spec? how integrated

steve: try to fit into current structure
... what's currently there is for conformance, not a detail implementation mapping for each element

rich: ie may use a very different drawing agent from another browser

cyns: which is why we can't talk about the mappings, just describe

rich: don't mind what the browser does, as long as it does its job and we have sufficient a11y data

cyns: yes, ml launches dialog

steve: is it possible for us separate out two tasks here?
... mappings is more work than the basic conformance decisions

cyns: seems easier to me, most is already implemented except for new elements

general discussion of what our task actually is

cyns: how to map element to api is what will be most useful to browser devs; it's also what ian has been saying
... also, none of these should be overwritten.

rich: ordered list, yes; input element, no

cyns: command element defs differ between w3 and what specs ...\

<Stevef> http://dev.w3.org/html5/spec/commands.html

rrsagent make minutes

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2010/05/25 20:56:41 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.135  of Date: 2009/03/02 03:52:20  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Found Scribe: janina
Inferring ScribeNick: janina

WARNING: No "Topic:" lines found.

Default Present: Rich, Janina, Stevef, Janina_Sajka, Michael_Cooper, [Microsoft]
Present: Rich Janina Stevef Janina_Sajka Michael_Cooper [Microsoft]
Got date from IRC log name: 25 May 2010
Guessing minutes URL: http://www.w3.org/2010/05/25-aria-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

[End of scribe.perl diagnostic output]