- From: Ian Hickson <ian@hixie.ch>
- Date: Thu, 16 Jul 2009 23:15:29 +0000 (UTC)
- To: Charles McCathieNevile <chaals@opera.com>
- Cc: Steven Faulkner <faulkner.steve@gmail.com>, HTMLWG WG <public-html@w3.org>, wai-xtech@w3.org
On Thu, 16 Jul 2009, Charles McCathieNevile wrote: > On Sat, 27 Jun 2009 08:22:28 +0200, Steven Faulkner > <faulkner.steve@gmail.com> wrote: > > > > 3. rephrase the question: what would need to be added to the canvas > > API so that the UI controls in the example could have programmtic > > focus (like an area on an image map) name, role and state information > > provided. > > This is pushing into the area of solutions to actually difficult > problems posed by canvas, which I think involves building something like > an ARIA-tree in the DOM. In my ideal vision this would be backed by some > actual fallback like SVG, with real markup objects, but I imagine that > may of the use cases for canvas mitigate against that anyway. > > As Jonas noted elsewhere, to a certain extent this leaves us relying on > authors to do the right thing, rather than having it happen > automatically. But in some cases (like the table stuff, which took > almost no time to get re-implemented at Opera by one designer) it's not > complicated and best practice can be spread reasonably easily. In > others, it is hard - but as has been noted many times, getting a chance > at making a few important things accessible is better than rejecting the > baby with the bathwater of misuse and ineffective attempts. At least if > you rely on accessible content to take part in society, which many > people do. So what should the spec say? -- Ian Hickson U+1047E )\._.,--....,'``. fL http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Received on Thursday, 16 July 2009 23:17:07 UTC