W3C home > Mailing lists > Public > w3c-wai-ua@w3.org > July to September 2002

Re: Comments on the User Agent Accessibility Guidelines 1.0 doc (21 August 2002 draft)

From: Jon Gunderson <jongund@uiuc.edu>
Date: Mon, 09 Sep 2002 09:12:33 -0500
Message-Id: <>
To: Brian Kelly <b.kelly@ukoln.ac.uk>, w3c-wai-ua@w3.org

Responses are indicated by JRG:
At 03:05 PM 9/8/2002 +0000, Brian Kelly wrote:

>Some comments on the User Agent Accessibility Guidelines 1.0 (21 August
>2002 draft).
>Apols if these have been addressed previously.
>The document does not appear to address the printing of resources.  A
>page may be rendering correctly on screen, but is corrupted when
>printing (as has happened in Netscape).  It would appear that a user
>agent could comply with the UAAG, but provide inaccessible hardcopy

JRG: COuld you provide more information here.  I am not sure what you mean 
by inaccessible hardcopy.

>The LONGDESC attribute is poorly supported in user agents and so there
>is very limited deployment experience.  Should the guidelines mandate
>support for it and give guidence on how it should be used?

JRG: UAAG requires access to all "conditional content".  LONGDESC is one 
example of conditional content and access is required at the P1 
level.  (see checkpoint 2.3)

>Guideline 3 - All configuration not to render some content ...
>I think there is an issue concerning the granularity of this control.
>For example, it may be necessary to switch off JavaScript to disable
>certain features (e.g. animated text), but JavaScript may be needed for
>other reasons.
>I guess related to this is the ease of disabling features.  For example,
>blinking or animated text could be implemented in various ways (e.g.
>proprietary HTML tags, through CSS, through JavaScript, through animated
>GIFs,  through Java, etc.).  An end user would want to switch off the
>animation, and not CSS, Javascript, etc. as they won't necessary know
>about these technologies.
>Guideline 9 - Provide navigation mechanisms
>Does this mean that a compliant user agent should (must) support <LINK>
>e.g. the nexp, previous, up, etc attributes.  I think this would be
>great.  This element has been aound for a long time, but user agent
>supporty is very limited.  If this is the intention, shouldn't the
>document mention it explicitly?

JRG: We could certainly add this to the techniques document.  But it is 
currently on an explicit requirement of the document.  Navigation is an 
important, but complex issue.  One of the limitation of the HTML 4.01 
specification is that in many cases it does not state what behaviors should 
be associated with an element.  In the case of LINK, it implies behavior 
but does not state what the browser should do to comply.  This would help 
UAAG alot since it could point to a spec and say do this.

Thanks for your comments,
Received on Monday, 9 September 2002 10:11:01 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:49:32 UTC