Action: AsCT algorithm and CSS

Hi,

Two weeks ago, I got an action item to "walk the AsCT algorithm" for CSS.

The first condition for CSS to be counted as an 
accessibility-supporte content technology is that it should be 
supported by assistive technology. For this purpose, I looked for 
test results for CSS support in AT. I have collected my current 
findings in the attachment (ScreenReaderCSSSupport.html). The test 
that I found date from 2 to 4 years ago and don't cover certain areas 
of CSS very well (I hoped to find more and clearer test resuls for 
CSS layouts and for image replacement, for example). I am not ready 
to draw conclusions from these tests.
There is also an API for CSS (SAC: Simple API for CSS: see 
http://www.w3.org/Style/CSS/SAC/ and http://www.w3.org/TR/SAC/), but 
this is not the same thing as the "accessibility API" mentioned in 
the conformance section.

The second condition for CSS to be counted as an AsCT is that 
accessibility-supported host user agents for the technology are 
available to users. CSS is "supported natively in widely-distributed 
host user agents that are also accessibility-supported":
* Internet Explorer supports CSS (although not fully, like many other 
browsers) and is accessible with several screen readers on Windows;
* Firefox supports CSS and can be accessed with Window-Eyes (5.5+) 
and JAWS (7.1+) on Windows, and possibly also with Orca on GNOME 
desktops (haven't checked the details, e.g. need for FireVox and 
self_voicing.py, and there are no test results for this combination 
in the attachment mentioned above) and with LSR/Linux Screen Reader 
on GNOME (haven't checked the details; LSR is shipped with the Ubuntu distro);
* Safari supports CSS and can be accessed with VoiceOver on Mac OS X.

So some details still need to be filled in.
(Also, it is not sufficient to look only at JAWS, Window-Eyes and IBM 
HomePage Reader because the most commonly used screen readers/talking 
browsers for Windows in a country like the US is not necessarily the 
most commonly used screen reader/talking browser elsewhere.)

The above focuses on screen readers, but other AT also need to be 
taken into consideration, e.g. talking browsers, TTS used by some 
people with dyslexia, and zoom software.

Even though this survey is not as complete as I would like, I am 
sending it now because I will need to concentrate on other things 
until the middle of next week.

Best regards,

Christophe


-- 
Christophe Strobbe
K.U.Leuven - Departement of Electrical Engineering - Research Group 
on Document Architectures
Kasteelpark Arenberg 10 - 3001 Leuven-Heverlee - BELGIUM
tel: +32 16 32 85 51
http://www.docarch.be/ 

Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm