Hi Rich,
Thanks for driving this.
I can't speak for Mozilla as a whole, but thinking of all our
discussions and the feedback that has come from people inside and
outside this group, this approach seems like a reasonable compromise to
me, and I'd like to hear what others think.
Nits:
For the introductory paragraphs:
" When possible a facility must me" should be "be"
" accessible binding" needs to be defined. "Binding" might get confused
with XBL etc.
" The determination of the disposition of the <canvas> subtree can be
determined at load time." could be "can be done at load time."
" When set to true it indicates that the canvas subtree is to be use as
a direct accessible subtree of canvas" could be " When set to true it
indicates that the canvas subtree is to be used as a directly keyboard
operable accessible subtree of canvas".
" The fallback content can come over the wire" could be "Progressive
enhancement can be achieved since the fallback content...", and " a
script can determine whether canvas is renderable" could be "a script
can use feature detection to determine...".
In the proposed text for html5:
I would say something about the adom indicating that the sub-tree of
canvas is intended to interact (be kept in sync?) with the canvas
rendering. I would not prescribe that the " rendering of the subtree is
controlled by script through the canvas API".
Regarding the last sentence " Add the following definition to the HTML 5
glossary:" was there more here?
cheers,
David
On 18/02/10 2:20 PM, Richard Schwerdtfeger wrote:
>
> Folks,
>
> I would like to take my adom proposal change to canvas to vote. Do you
> support it? I have a due date of February 25. I need to get the last draft
> in spec. ready format for the weekend and after the last meeting we appear
> to have consensus barring some word smithing. We still have a dependency
> on Steve to address action item 19 regarding the caret.
>
> http://lists.w3.org/Archives/Public/public-canvas-api/2010JanMar/0163.html
>
> Janina wants to do a straw poll early next week as there is a Feb. 25 due
> date for this. It would seem that the Opera example would be an excellent
> best practice for some content. We would simply set adom to true when
> canvas is supported. The author needs to ensure visual focus is drawn on
> the canvas.
>
> Rich
>
> Rich Schwerdtfeger
> Distinguished Engineer, SWG Accessibility Architect/Strategist
>