- From: Al Gilman <asgilman@iamdigex.net>
- Date: Thu, 05 Oct 2000 22:24:30 -0400
- To: <w3c-wai-ua@w3.org>
- Cc: kasday@acm.org, Bruce_Bailey@ed.gov
[Len, Bruce: the reference thread is at http://lists.w3.org/Archives/Public/w3c-wai-ua/2000OctDec/thread.html#4 ] According to Bruce Bailey's report http://lists.w3.org/Archives/Public/w3c-wai-ig/2000OctDec/0012.html if you can force the text up in size you have solved the P2 problem. Icons don't seem to be the same severity of problem. This would sound as though, modulo a possible P3 gap, you are probably covered as is. I am copying Len and Bruce on this as I feel a bit over my depth. I don't have that good a background in low vision. Al On second thought that bit about icons is not so strange. Icons are analog encoding. This means you would expect them to degrade gradually on low-pass filtering where all you get is an outline, more or less. Text doesn't look like it sounds. You have to correctly identify the letters. Fuzz text and all the letters without ascenders or descenders look alike; that's an awful lot of them! Symbolic codes have better error rejection for low error rates. That's why digital and FM before it deliver high fidelity. But like the little girl with the little curl; when they are bad... watch out!
Received on Thursday, 5 October 2000 22:02:43 UTC