W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > July to September 2003

Re: Possible problem with Fahrner Image Replacement

From: Charles McCathieNevile <charles@sidar.org>
Date: Tue, 15 Jul 2003 18:58:47 +0200
Cc: Joe Clark <joeclark@joeclark.org>, WAI-IG <w3c-wai-ig@w3.org>
To: Brandon Bowersox <brandon@ojctech.com>
Message-Id: <9A1B3922-B6E5-11D7-AC3E-000A958826AA@sidar.org>

I think problem 4 - the case of low-vision users, is the one where this 
technique is really problematic. And not for people with seriously low 
vision, who use a zoom of 10x on a screen magnifier, but the large 
number of people with moderately poor eyesite who take off their 
glasses (because they give you a sore nose) and enlarge the text 50% or 

In cases like Joe's example, the problem is barely evident - browsers 
can enlarge the image as well as the text - but they can't reflow 
images. This holds true for text in SVG as well, until version 1.2 
which includes reflowing text. But heavily used this technique can make 
life somewhat frustrating - putting 15% of the text outside the window 
is enough to really slow down reading, and requires people to start 
manipulating the browser a lot. Combine it with a motor disability 
(even a minor one) and you have a problem at the level of reducing 
efficiency and concentration for a lot of people.

I think the long-term solution is to use smarter things than HTML - SVG 
1.2 is not yet the accepted standard for SVG, but it is getting market 
acceptance and does provide more possibilities for solving the problem. 
In the short term, this is the sort of case where an alternate "clean" 
version can be readily generated automatically (change or disable the 
document stylesheet) on client or server side, if the user knows to do 
it. Which leaves the disabled user needing to learn more than others to 
make the same technology work - not something unusual, but not ideal. 
Of course, we don't live in an ideal world...

just some thoughts.


PS iCab makes a mess of reading the example, that is apparently due to 
bugs in its CSS implementation. Without the floating it might get it 

On Tuesday, Jul 15, 2003, at 17:09 Europe/Zurich, Brandon Bowersox 

> Thanks for organizing this testing--I look forward to the results and 
> the possibility that designers can use images to control the display 
> of text without hindering accessibility.
> One question--it seems there are 4 main problems with the age-old 
> method of using images to display text:
> 1) The text will only be accessible if the designer includes useful 
> ALT text within the IMG tag.
> 2) Even so, text inside an IMG ALT tag is not indexed by search 
> engines and not searchable by users with a browser's Find feature.
> 3) Adding ALT text is not sufficient for AREAs inside a client-side 
> imageMAP (redundant text should be added 
> http://www.w3.org/TR/WCAG10-HTML-TECHS/#client-side-redundant-text)
> 4) Images with ALT text are not legible for low-vision users because 
> the text is not crisp as it scales larger (even with the fancy zooming 
> Opera does).
Charles McCathieNevile                          Fundación Sidar
charles@sidar.org                                http://www.sidar.org
Received on Tuesday, 15 July 2003 12:59:19 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 13 October 2015 16:21:24 UTC