W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > October to December 2004

RE: Guideline 2.5 Level 3 SC work item

From: Jason White <jasonw@ariel.its.unimelb.edu.au>
Date: Wed, 15 Dec 2004 16:22:10 +1100
Message-ID: <16831.51714.631453.316106@jdc.local>
To: Web Content Guidelines <w3c-wai-gl@w3.org>

Gregg Vanderheiden writes:
 > I think the problem is backward.  
Agreed.
 > 
 > The accessibility problem is twofold
 > 
 > - for cognitive and learning disability - spelling is a problem and picking
 > choices (when there is a short list) is good.  
 > 
 > - for blindness - accessing long lists is a problem and we want to encourage
 > the ability to type in an entry and not have to scroll and pick.  But typing
 > a part of the entry and jumping to it may work just as well.

The rule appears to be: for short lists provide a menu of options; for
long lists provide text entry. Menus are optional for long lists; text
entry is optional for short lists, and redundancy in the interface is
desirable where practicable. Is this accurate?

For HTML and user interface design this is concrete and testable (as
long as one specifies thresholds for distinguishing short from long
lists). However this is also a user agent problem and I wouldn't want
to require the content author to try to control the interface if a
more abstract language is being used in the implementation, or if user
agents implement incremental or text search over u i components (e.g.,
form fields).

Again we run into the interplay between content and user agent. Is
there a way of abstracting this so that it applies in HTML techniques
for example, but isn't a requirement in the guidelines? That is, can
we discern a higher-level concept that can be included in the
guidelines themselves without assuming that the content author
exercises
control over the user interface?
Received on Wednesday, 15 December 2004 05:22:41 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:32 GMT