- From: Davide P. Cervone <dpvc@union.edu>
- Date: Mon, 21 May 2012 20:49:22 -0400
- To: Cameron McCormack <cam@mcc.id.au>
- Cc: Dirk Schulze <dschulze@adobe.com>, Vincent Hardy <vhardy@adobe.com>, <public-fx@w3.org>, Peter Krautzberger <p.krautzberger@googlemail.com>
> I think generating the spoken text is an interesting approach! The > results there seem much better than the reading out of the words > within the <pre>. Thanks. I was pleasantly surprised at how well it turned out so far. There are still a number of issues that will need to be worked out, but I think it is a promising approach. > I wonder though if your approach of making the existing math subtree > aria-hidden="true" and inserting the equivalent spoken text in a > <div> positioned off the page would get in the way of ATs that do > know how to understand MathML and present it to the user. I can't > think of a good way around that -- feature testing for the abilities > of the AT is impossible I guess. I suspect you are right, we will not be able to tell anything about the screen reader from within javascript. I think we will have to allow users some control over the process, since they are the ones who know what screen reader they are using and what needs they have. If they have a screen reader that understands MathML (if these are developed in the future), then they can use the MathJax menu to change the renderer to the NativeMML one, which will output MathML rather than HTML or SVG. Then their screen reader will be able to read that (though if their browser doesn't understand it, it would not have a good visual representation). There may also need to be a setting to control whether the hidden text is generated or not. I haven't thought a lot about what might be needed in terms of user controls, but I suspect we will need to get assistance from the accessibility community. > (I'd love to know what the accessibility community thinks is the > state of the art for presenting math in web pages accessibly, by the > way.) I would be interested in that as well. Davide
Received on Tuesday, 22 May 2012 00:50:12 UTC