- From: Henri Sivonen <hsivonen@iki.fi>
- Date: Thu, 10 Apr 2008 21:47:42 +0300
- To: Jon Gunderson <jongund@uiuc.edu>
- Cc: "Gregory J. Rosmaita" <oedipus@hicom.net>, Al Gilman <Alfred.S.Gilman@IEEE.org>, Dan Connolly <connolly@w3.org>, Michael (tm) Smith <mike@w3.org>, Chris Wilson <Chris.Wilson@microsoft.com>, HTML WG <public-html@w3.org>, W3C WAI-XTECH <wai-xtech@w3.org>
On Apr 10, 2008, at 17:06, Jon Gunderson wrote: > > It should also be noted just because HTML5 is being developed and > implemented it does not mean all developers will use it. If people > use HTML5 and it supports accessibility those resources will not > need to use ARIA (Hurrah for those developers!). Other developers > for various reasons will choose not to use HTML5 technologies and > will need to use ARIA to make their resources more accessible. ARIA has been put forward as a mechanism for retrofitting accessibility to existing DOM scripting-based systems. Legacy systems using DOM scripting consist predominantly of text/html and of SVG to a minor extent. HTML 5 is the spec for processing text/html even if authors aren't knowingly using HTML5. It follows from browser software architecture that application/xhtml+xml inherits the above-DOM processing from text/html. Which other languages besides text/html, application/xhtml+xml and SVG is ARIA expected to cater for? Surely potential future languages that have not yet been implemented in browsers in a legacy form should be developed to be natively accessible and not to need retrofitted accessibility in the first place. (MathML isn't used in a way that needs ARIA except perhaps for the live region features.) -- Henri Sivonen hsivonen@iki.fi http://hsivonen.iki.fi/
Received on Thursday, 10 April 2008 18:49:03 UTC