W3C home > Mailing lists > Public > wai-xtech@w3.org > April 2008

Re: [html4all] some reflections on @alt usage

From: Charles McCathieNevile <chaals@opera.com>
Date: Mon, 28 Apr 2008 15:00:00 +0200
To: "Laura Carlson" <laura.lee.carlson@gmail.com>, "W3C WAI-XTECH" <wai-xtech@w3.org>, HTML4All <list@html4all.org>
Message-ID: <op.uabnqawrwxe0ny@widsith.local>

On Mon, 28 Apr 2008 14:00:58 +0200, Laura Carlson  
<laura.lee.carlson@gmail.com> wrote:

> Optional alt is a way to codify and bless bad tools.

I don't think so.

There is a lot of nonsense that Ian had in drafta about when you can omit  
alt - circumstances where it is necesary and he is just wrng. There are  
other badly constructed examples (his use of HTML5 markup with figure and  
legend should have alt="" since the content is already there, not omit the  
alt). All of the stupid advice about times when alt can be ommitted needs  
to be rewritten.

There are arguments given in the debate that are about blessing bad tools,  
an those specific arguments should be shot down in flames.

The spec should point out that not having alt is wrong. But having some  
default value that is meaningless, or that collides with a meaningful  
value like alt="" is even more wrong since it not only breaks the user  
experience like no alt does, but it also becomes harder to test and breaks  
existing rules, advice, and tools.

If we get these issues fixed in the spec then we are ready to start  
addressing the actual question of whether missing alt should be a  
validation error.

Optional alt is about a belief that validation is more important to people  
(in particular tool developers) than accessibility. If that turns out to  
be true, then it makes sense. If that turns out to be false, then it  
doesn't. But we have to have some research - both sides of the argument  
are currently based on gut feeling and instinct. There is a real issue  
here, and there is very little real information being provided to settle  
the question rationally.

So my preferred approach is to fix the rest of the rubbish in the spec,  
and do some research on the real issue, and *then* talk about making a  
decision...

cheers

Chaals

-- 
Charles McCathieNevile  Opera Software, Standards Group
     je parle français -- hablo español -- jeg lærer norsk
http://my.opera.com/chaals   Try Opera 9.5: http://snapshot.opera.com
Received on Monday, 28 April 2008 13:00:42 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:51:35 UTC