- From: Ian Hickson <ian@hixie.ch>
- Date: Sun, 28 Feb 2010 06:51:15 +0000 (UTC)
- To: Richard Schwerdtfeger <schwer@us.ibm.com>
- Cc: public-canvas-api@w3.org, "public-html-a11y@w3.org" <public-html-a11y@w3.org>
On Thu, 25 Feb 2010, Richard Schwerdtfeger wrote: > > So, perhaps what would be easier for all of you is to simply have an > attribute that says include the subtree in the navigable document > structure when canvas is rendered? This is not the case now. This way > you can't twist the definition of adom around to say that it is a > compliance statement. After this we provide direction on how ensure the > subtree supports accessibility. > > Right now HTML 5 has a whopper hole in it in that you can't use what is > in the subtree to navigate the canvas rendering. > > something like navigatesubtree. I would not call it fallback because it > may not me. Is there a fifth case that I haven't thought of, in addition to these four?: - content is empty (reading has no effect) - content is accessible augmentation of <canvas> - content is the only accessible alternative to the <canvas> - there is no accessible alternative and the content says so If so, what is this fifth case? We don't need navigatesubtree="" for any of the four cases above, but if there is a fifth case I haven't considered, then that's a different matter. -- Ian Hickson U+1047E )\._.,--....,'``. fL http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Received on Sunday, 28 February 2010 06:51:53 UTC