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

RE: Tests 59,

From: Gregg Vanderheiden <gv@trace.wisc.edu>
Date: Tue, 21 Dec 2004 12:06:09 -0600
To: "'Chris Ridpath'" <chris.ridpath@utoronto.ca>, "'WAI WCAG List'" <w3c-wai-gl@w3.org>
Message-ID: <auto-000190771699@spamarrest.com>

Hi Chris,

 If I understand this correctly - then I don't think this is quite right.
I see what you are trying to do -- but that is not what was agreed to by the
group.  We can't make the test be something other than the guidelines.   And
I'm not sure what you propose is right even then.   

What we need to do at this point is take it back to the full group and have
a discussion.  I can't say yea or nay on something that is different than
the last group consensus.   Propose - yes.  But that is all.

So lets take this up with the full group. What you are talking about has
many implications including re-defining alt text and creating a compound
"alt text" "title" etc group of rules with interaction.  (E.g. what goes in
ALT differs depending on whether element has title attribute etc). 

So I think we have to stop here and pick it up when our full group meets
again. 

In the meantime, please summarize what you think should be done with each
attribute for each type of element that has ALT attribute.

Thanks 

 
Gregg

 -- ------------------------------ 
Gregg C Vanderheiden Ph.D. 
Professor - Ind. Engr. & BioMed Engr.
Director - Trace R & D Center 
University of Wisconsin-Madison 


-----Original Message-----
From: Chris Ridpath [mailto:chris.ridpath@utoronto.ca] 
Sent: Tuesday, December 21, 2004 9:32 AM
To: Gregg Vanderheiden; 'WAI WCAG List'
Subject: Re: Tests 59,

Hi Gregg,

Thanks for your thoughtful comments.

Alt text is one of the core means of making content accessible so I want to
make sure we get this right. Please forgive me if I belabor this issue.

Test 11 ensures that any text in the image will appear in the alt text.
http://www.w3.org/WAI/GL/WCAG20/tests/test11.html

I propose modifying test 59 so it uses test 11 as a prerequisite and test 11
takes precedence.

If your submit button is an image and it contains text then that text
becomes the Alt text. This should cover most image submit buttons and it's
what most people do now.

This narrows our discussion to submit buttons that use images containing no
text. This is likely a small minority of submit buttons.

The Alt text for these images should not be the purpose of the button. It
should not be "submit" or "button" or any combination of these 2 words. The
purpose, (a submit button) is abundantly clear, both explicitly (through the
HTML code) and implicitly (its position within the form) so no need to put
that in the Alt text.

The Alt text should be the purpose of the form. You provided good examples
as "search" or "find". Other examples could be "purchase" or "renew
license".

I propose modifying test 59 so it reads:
"INPUT element with type of "image" must have Alt text that identifies the
purpose or function of the form."
(This only applies to images that don't have text in them.)

Our current techniques state "This label should indicate the button's nature
as a submit button" and use "submit" as an example of good Alt text. This is
a common practice that should be discouraged.

I note that previously in this thread, John Slatin argued that "button"
should not be included in the Alt text.

I've created test 192 that checks for "button" and "image":
INPUT element with type of "image" must not have redundant Alt text.
http://www.w3.org/WAI/GL/WCAG20/tests/test192.html

These proposals should ensure that submit buttons using images are
accessible. However, we can improve accessibility...

Gregg noted:
> If the search button is a magnifying glass or binoculars
> its alt text should be 'search' or  'find'  not 'magnifying
> glass' or 'binoculars'
>

This is today's thinking but allow me to take you to the fantasy world of
tomorrow. Open your mind and explore the vast possibilities of what may be.

In the world of tomorrow, the purpose of the button is in the INPUT
element's title attribute. The Alt text describes the image.

Our fantasy world contains user agents that render, if present, the title
attribute. A screen reader would say "search" when encountering code like:
<input type="image" name="submit" src="button.gif" alt="magnifying glass"
title="search" />

This is allowed under the current UAAG and HTML specs.

It means the user has more semantic information available - they can find
out what the image looks like. Increased semantic information is a good
thing

> The alt text was never intended to provide a
> description of the image.
>
Break the chains! Surge toward the new frontier of increased web
accessibility. Use Alt text to describe the image and use title attributes
for functional purposes.

Using Alt text for multiple purposes is confusing and should be avoided.

Back in our hum-drum world of today, I see there's a lot of discussion on
the lists regarding this issue so perhaps we should let this simmer for a
bit before making any firm decisions.

Cheers,
Chris

----- Original Message ----- 
From: "Gregg Vanderheiden" <gv@trace.wisc.edu>
To: "'WAI WCAG List'" <w3c-wai-gl@w3.org>
Sent: Monday, December 20, 2004 5:22 PM
Subject: RE: Tests 59,


>
> Hi Chris,
>
> The alt text was never intended to provide a description of the image.
Just
> its function.   A 'stand in' if you will if the image is missing.    The
> test is
> "If you replace the image with the text - would everyone do the same
thing -
> receive the same information - etc.  as best as possible."
>
> If the search button is a magnifying glass or binoculars its alt text
should
> be 'search' or  'find'  not 'magnifying glass' or 'binoculars'
>
>
> If the image is meant to convey an image - then a description is
> appropriate.
>
> If it was meant to convey data - then that is what is appropriate.
>
> If it was meant to convey a function then that is what should be there.
>
> "alt" text was not called "desription" or "Desc".  It was not meant to be
a
> description per se.    Just when appropriate.
>
>
> Gregg
>
Received on Tuesday, 21 December 2004 18:06:18 GMT

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