- From: Jason White <jasonw@ariel.its.unimelb.edu.au>
- Date: Thu, 26 May 2005 15:24:14 +1000 (EST)
- To: Web Content Accessibility Guidelines <w3c-wai-gl@w3.org>
On Wed, 25 May 2005, Gregg Vanderheiden wrote: > > The text is sent in character codes that are not Unicode. Then, the user > agent paints them to the screen as pixels (as all user agents do). The > difference is that AT can access the characters before they are painted to > the screen and, if they are Unicode, it can render them as voice etc. if > they are not Unicode (e.g. proprietary or encrypted) and are not made > available as Unicode - then AT can do nothing with them. The above is correct and captures the essence of the example. This kind of situation can be very real if the document is stored in an encoding specific to the fonts used in rendering it. Without a mapping to Unicode (or another standard character set/encoding), the user agent can't identify the characters in the document, or can do so at best only partially. Often the font is provided along with the document, as in PDF, which supports an Unicode map for exactly the purpose described here, i.e., identifying the characters used in font-specific encodings.
Received on Thursday, 26 May 2005 05:24:22 UTC