W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > January to March 2005

RE: [techs] Alt Text Tests 192 and 195

From: Gregg Vanderheiden <gv@trace.wisc.edu>
Date: Mon, 24 Jan 2005 19:29:34 -0600
To: "'Chris Ridpath'" <chris.ridpath@utoronto.ca>, "'WAI WCAG List'" <w3c-wai-gl@w3.org>
Message-ID: <auto-000218602606@spamarrest.com>

There is a lot here to discuss I think.  Some of it will be hard to
determine.

1 - We can't map techniques to success criteria unless the criteria requires
them.  

2 - I don't think some of these rise to level 1 criteria.    So I think we
still have a problem. 

3 - we are going to have to look real hard to determine what level some of
these go at - or if they are just good advice.   And some are controversial
which tends to make one wonder if they are critical.  

Hmmmmmm

Thanks for bringing all these to the surface.   I think we need a "holding
tank" to put things in that we are unsure of so that we don't have them "IN"
when we aren't sure but also don't lost them by putting them out.  

Maybe we should have a list of all techniques and a separate list of just
those that are approved for inclusion at this time.  
 
Gregg

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


-----Original Message-----
From: w3c-wai-gl-request@w3.org [mailto:w3c-wai-gl-request@w3.org] On Behalf
Of Chris Ridpath
Sent: Monday, January 24, 2005 10:08 AM
To: Gregg Vanderheiden; 'WAI WCAG List'
Subject: Re: [techs] Alt Text Tests 192 and 195


> Currently we don't have any SC to tie these to.   So I don't think we can
> require them unless we add new SC to levels 2 or 3.
>
There was a suggestion a while back (Sept. 2, 2003) to create a guideline or
success criteria that addressed exactly this sort of thing. Recently, Wendy
proposed [1] to drop this suggestion because it was covered by guideline
1.1, level 1, SC 4. I've mapped these tests to this success criteria for
now.

Success criteria 4 in guideline 1.1 specifically refers to "non text
content" and these tests deal with text so I think my mapping to that SC may
be incorrect. Perhaps a more accurate mapping would be to guideline 3.1,
level 3, SC 2 or 3. [2].

> Do these really affect accessibility - or just reduce chattiness and
> annoying repetition?
>
I interpret the guidelines to mean that chattiness and annoying repetition
are accessibility problems. For example, guideline 1.1, level 1, SC4
specifically addresses this by stating you must mark some things so they can
be ignored. We've also discussed this issue in regard to text in images used
as buttons. Some text in the image should (must) be ignored because it is
"decorative" or "incidental" or "not relevant". I think there's general
agreement that accessibility improves when some content is excluded. But
where do we draw the line?

Test 195 "don't use source anchors starting with 'link to' or 'go to'"
addresses the issue of reducing chattiness and annoying repetition. Most
authors are not going to object to this - it means they actually have to
write less. It's a simple rule that can be easily applied and tested for
compliance. This issue will also come up when we discuss tests for link text
so we should try to get a good decision now.

Test 192 "don't label your submit buttons as 'submit'" may upset some
authors. It's a common practice to label form submit buttons as "submit" and
getting people to change is always difficult. The issue here though is not
just annoying repetition but meaning. The button has much more meaning if it
says something about the form and that will benefit a larger audience. This
is a simple rule that can be easily tested for and will improve
accessibility but the burden on authors will be greater.

> These seem like good ideas - but I am afraid to put every good idea
> in as an SC or our guidelines will get very long.
>
It's a difficult task trying to get the guidelines right but I hope these
detailed questions have helped rather than hinder the process.

If the tests are a good idea and should be covered by the guidelines then I
suggest we could:

1) broaden guideline 1.1, L2, SC4 so text as well as non-text should be
ignored, or
2) add another SC as suggested earlier [1], or
3) map them to guideline 3.1, L3, SC2 or 3

Cheers,
Chris
Received on Tuesday, 25 January 2005 01:29:44 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:35 GMT