RE: [Proposal] Checkpoint 10.4 highlight requirement and image maps

If we think about this issue from the perspective of the user, the goal is
to inform the user of the active elements on the screen.  It really doesn't
matter if the image is an HTML image map or SVG image, as that is behind the
scenes.  The user needs to know what elements are active.  In the case of an
image map, highlighting the entire map indicates that it has active areas,
and highlighting the link areas shows where those areas are.

It would seem that an SVG map, with multiple objects, might have "filler"
objects that do not link to anywhere, to tie together those that do.  The
language of the checkpoint should be generic enough that it indicates that
the need is to indicate to the user those areas or components of a rendered
image which will link to other content.

Denis Anson

-----Original Message-----
From: w3c-wai-ua-request@w3.org [mailto:w3c-wai-ua-request@w3.org]On
Behalf Of Ian B. Jacobs
Sent: Tuesday, July 03, 2001 1:05 PM
To: w3c-wai-ua@w3.org
Cc: dean@w3.org
Subject: [Proposal] Checkpoint 10.4 highlight requirement and image maps


Hello,

In the 22 June 2001 UAAG 1.0 [1], provision 4 of checkpoint 10.4
("Highlight
special elements") reads:

   "For an image map, the user agent must highlight the image map as
    a whole and should allow configuration to highlight each enabled
    region."

I discussed this provision and the following question with Dean Jackson
of the SVG Working Group: Should we consider an SVG with links in it an
image map?

We both felt that an SVG image with links in it was a kind of image map,
but that it was different from image maps as used in HTML, and we did
not
think that provision 4 should apply to SVG-like content. My summary of
the
difference is this:

 - In SVG, an image is composed of objects, each of which is rendered
   independently, in document order.

 - The case we meant to address in 10.4 was for images that are not
composed
   of objects rendered individually (e.g., a GIF image with active
regions
   specified with AREA elements).

I propose that the above provision be edited to reflect a difference
between
composed and monolithic images and video.

<NEW>
For images or animations that have enabled graphical regions (e.g., HTML
image maps or SVG images that include links), the user agent must
highlight each region that corresponds to an independently rendered
element.

 Note: For instance, in HTML, a user agent rendering a PNG image
 as part of an image map must highlight the image as a whole (and
 may highlight each individual region specified in the MAP element).
 For an SVG image that includes 15 links, each rendered independently
 according to the SVG specification, the user agent must  highlight all
 15 links.
</NEW>

We should also include some HTML and SVG examples in the Techniques
document.
Your contributed examples are welcome.

I am not thrilled with this language, but I support the distinction.
Improvements to the language welcome.

 _ Ian


[1] http://www.w3.org/TR/2001/WD-UAAG10-20010622/

--
Ian Jacobs (ij@w3.org)   http://www.w3.org/People/Jacobs
Cell:                    +1 917 450-8783

Received on Wednesday, 4 July 2001 08:04:01 UTC