- From: Al Gilman <asgilman@iamdigex.net>
- Date: Thu, 20 Sep 2001 11:19:46 -0400
- To: <w3c-wai-ig@w3.org>
At 09:23 AM 2001-09-20 , Jonathan Chetwynd wrote: >I am grateful for your responses, perhaps part of the problem I'm having is >that ie6 renders the title as a 'tooltip' whereas lynx renders the alt. >This leaves me a little non-plussed. The title is additional information, >however if the image is rendered, then the user first needs the alt tag.... AG:: Here is your essential contribution. You have users who still need the ALT information first, even 'though the image has been seen. To back up the User Agent Accessibility Guidelines, we need to document that this case actually happens. The conventional behavior, of writing an annotation and not a title in TITLE, and showing the additional information, not the naming or summary information as the transient display when the user mouses over, is predicated on the premise that the user has got the idea that is stated in the ALT by viewing the image. This is not so for all. There is language in the User Agent Accessibility Guidelines ensuring that the user has access to all content, regardless of what the author thought was useful, where need be. What this is is another use case we need to work up as a scenario and User Agent techniuqes entry where for your clients, there is a User Agent mode setting (policy) which a) displays ALT in preference to TITLE onMouseOver, and pushes the TITLE into the head of the context menu which is the next logical 'pull' protocol place to get at it. This is your thread, so do you wish to re-post a reply to this message to UA and not IG seeking partners to implement and evaluate this technique? Can it be done in a script as a prototype? That has to do with the way the browesr and the script handle onMouseOver that Jim Ley will know and I don't know. Al PS: Partly you are still muddled because the logical model of the format is muddled. Sorry. It is on file as a problem area in HTML, and we don't have a solution at the moment. A working UA technique that at least shows how it can work would be useful to the HTML folks in planning a migration to a model that works for both authors and user agents following different user policies. [sorry, the search.w3.org service is down, and I don't have this FAQ bookmarked. It's in the early entries in the wai-tech-comments archive and it talks about "why TITLE isn't a title" or some such. Oops! forgot Google. Here it is:] wai-tech-comments@w3.org from July 2001: behavior matters (...) <http://lists.w3.org/Archives/Public/wai-tech-comments/2001Jul/0001.html>h ttp://lists.w3.org/Archives/Public/wai-tech-comments/2001Jul/0001.html PPS: Once again this shows how the needs of the cognitive disabled and the visually disabled are more similar than they are different. People think ALT is there for the blind, but LD users benefit from it, too. >so I am still muddled. >Is ignoring the title attribute maybe simplest? >tx > >jonathan chetwynd >IT teacher (LDD) >j.chetwynd@btinternet.com ><http://www.peepo.com/>http://www.peepo.com "The first and still the best picture directory >on the web" >
Received on Thursday, 20 September 2001 11:16:30 UTC