Re: alt and authoring practices

It is not my interpretation, it was what we were taught in grammar  
school.  Alt means replace.  if you are going to describe, it needs to  
be done in another way.  Yes, the guidelines are rong.  Their  
"interpretation" has lead to far too many alt="blue ball"s.  I have  
not encountered an image worth putting on the web asside from some  
little dressy things to aid in cognition or even less than dressy  
things for that purpose which cannot be replaced.  red left arrow  
means left correct? ruler can be replashed with ---------- correct????

On Apr 21, 2008, at 3:37 AM, Steven Faulkner wrote:

Hi David,

> replace not describe.

This may be your interpretation of what constitutes an approriate alt,
but it is not always the interpretation that the w3C web content
accessibility guidelnes 1 & 2 recommends. The  correctness of "replace
not describe" depends  upon the context that the image is used in and
whether the information that is in the image can be "replaced" using


On 16/04/2008, David Poehlman <> wrote:
> replace not describe.
> ----- Original Message -----
> From: "Dave Singer" <>
> To: "Anne van Kesteren" <>; "Steven Faulkner"
> <>
> Cc: "James Graham" <>; <>;
> <>; <>
> Sent: Wednesday, April 16, 2008 12:43 PM
> Subject: Re: alt and authoring practices
> At 13:29  +0200 16/04/08, Anne van Kesteren wrote:
>> On Wed, 16 Apr 2008 13:23:06 +0200, Steven Faulkner
>> <> wrote:
>>> I don't quite follow the logic, but that is probably due to my
>>> incapacity to understand, but I am pretty sure you are making a
>>> worthwhile point and will cogitate on it further.
>>>> With nobody having data of usage on the Web the position of the
>>>> editor seems more reasonable to me.
>>> And that is your prerogative as a member of the working group, I
>>> myself do not place faith in the editor as being all seeing and all
>>> knowing in the absence of data.
>> If my reasoning is correct the position of the editor is supported
>> by logic which is why his point seems more correct to me. Not
>> because he's the editor.
>> (You assume a minority case is likely to occur more often and the
>> editor assumes a majority case is likely to occur more often.)
> I do wonder if we are trying to pack too much into one attribute.
> Really, this is brainstorming and may be a bad idea, but are we
> trying to pack "what is the alt string" and "how trustworthy is the
> alt string" into the same attribute, when it can't be done?  the rest
> is somewhat in jest...
> maybe we need a second attribute alt-trust-level:
> 0   the string is empty or may as well be, or missing:  it's worthless
> 5   the string contains facts even a stupid program could work out
> from the image itself (e.g. width and height)
> 10 the string contains facts that were deduced automatically with
> some effort from the image itself
> 15 the string contains automatically collected ancillary data not
> found in the image (e.g. time of capture, camera)
> 20 the string contains human-entered data of a basic descriptive  
> nature
> 25 the string contains a rather detailed description of the image
> 30 the string contains an analysis of the meaning of the picture as
> well as its description
> 100 the string is a doctoral thesis, analyzing the image from every
> possible direction, including references to mythological, symbolical
> and historical references, history of the place/people shown,
> analysis of their health, state of mind, an aesthetic analysis of the
> composition, an analysis of the technical competence, and so on
> i'm guessing some people here think everyone should achieve level  
> 20.  :-)
> by the way, can one provide alt strings in multiple languages and/or
> scripts?  what would happen if someone tried level 100?
> --
> David Singer
> Apple/QuickTime

with regards

Steve Faulkner
Technical Director - TPG Europe
Director - Web Accessibility Tools Consortium |
Web Accessibility Toolbar -

Jonnie Appleseed
with his
Hands-On Technolog(eye)s
reducing technology's disabilities
one byte at a time

Received on Monday, 21 April 2008 14:42:43 UTC