W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > July to September 2007

Re: User Testing of Accessiblity Features

From: Tim <dogstar27@optushome.com.au>
Date: Wed, 29 Aug 2007 01:40:27 +1000
Message-Id: <005738d59dd17464dae0ed8203d6be99@optushome.com.au>
Cc: Andrew Kirkpatrick <akirkpat@adobe.com>
To: WAI Interest Group list <w3c-wai-ig@w3.org>

Is there really any need for spacer images anymore and empty alt tags? 
I removed all my empty alt tags years ago.

It was an old html 3.2 pre stylesheet solution to separate links by 
using tiny spacer.gif files between them. CSS allows you to visually 
put links next to each other on a page by using a hidden html element 
to separate the links. The html code is  <a class="noshow">:</a> . In 
the stylesheet for the noshow class, the CSS rule declares .noshow{ 
display:none;} A browser with stylesheets turned off will display : 
separating the links.

I hope you are testing the under-used longdesc tag <img 
longdesc="../Access/index.html#heretictic"> A tag which links to text 
information on the image. This is a potentially great addition to the 
alt short description of the image or link, you can link to a whole 
page of text on an image with longdesc or even a sound file reading of 
the text.

One last point but not for screen readers. Title might not be read by 
screen readers, but can be important for images as I recall Internet 
Explorer will only display the alt text on mouseover if there is no 
title text. The Firefox and Opera browsers only display the title text 
on mouseover and do not use the alt tag for this purpose. If you want 
the alt tag text visible on mouseover, the title tag is also necessary.

I hope I am on topic. I recently reviewed Australian university sites 
for validity and accessibility compliance and made a list of common 
errors and missing accessibility features. Are you testing for other 
what I think are important features like the noscript alternative for 


On 29/08/2007, at 12:57 AM, Andrew Kirkpatrick wrote:

> The main issue that I see related to the need for the alt attribute on
> images that have null alt is assistive technology support.  Screen
> readers like JAWS make attempts to read _something_ for images when
> there is no alt.  This is very important for linked images, but of
> questionable value for images that are not.
> Naturally, the extent of the value of the image varies on the type of
> the image, from spacer images (unimportant) to images bearing important
> content or text (critical), but if there is no alt attribute value then
> there isn't going to be information to read whether there is a null alt
> or not.
> The value of requiring alt on images has been that it is easy to tell 
> if
> a person is totally unaware of certain accessibility issues by the
> absence of alt attributes on images that need them.  I don't think that
> requiring the alt attribute on all images should be sacred - 
> equivalents
> for meaning images and linked images (including inputs with type=image)
> are required for accessibility, but I don't think it matters for other
> images whether the empty alt is there or not, so long as assistive
> technology ignores the image as they would with alt="".
> This seems like a pure technology question - I'm not sure what needs
> user testing here?
>> -----Original Message-----
>> From: w3c-wai-ig-request@w3.org
>> [mailto:w3c-wai-ig-request@w3.org] On Behalf Of Lachlan Hunt
>> Sent: Monday, August 27, 2007 10:15 PM
>> To: joshue.oconnor@cfit.ie
>> Cc: HTML WG; 'WAI Interest Group list'; www-archive
>> Subject: Re: User Testing of Accessiblity Features
>> Joshue O Connor wrote:
>>>  Hi Lachlan,
>>> On the IRC I see that you say:
>>>> # [14:25] <Lachy> I did ask Joshue O Connor in that email
>> if he could
>>>> help out, since he offered his services for such things, but never
>>>> heard back about it
>>> The user testing is taking place tomorrow. Testing of Header/id
>>> combinations, @summary and @longdesc. I think I mentioned
>> that previously.
>> I was aware that you would be testing those features, but I
>> think I was specifically referring to this email [1], where I
>> wanted to do testing of alt="" vs. no alt, etc.  Will that
>> issue be tested too, or left till another time?
>> http://lists.w3.org/Archives/Public/public-html/2007Aug/0577.html
>> --
>> Lachlan Hunt
>> http://lachy.id.au/
The Editor
Heretic Press
Email dogstar27@optushome.com
Received on Tuesday, 28 August 2007 15:41:09 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 13 October 2015 16:21:36 UTC