Re: Request for PFWG WAI review of Omitting alt Attribute for Critical Content

Dave Pawson wrote:
> As a document author you might be able to make that judgment.
> As a reader how can you know that that assessment has been made?

Indeed. There is a degree of trust on the part of the user that the
author is doing their best to provide them with appropriate information.
That it is a quality, well designed site etc. I guess, the acid test it
whether the user will continue therefore to use a site or service or
whether they move onto something better that suits their needs.

> The comparison is an author who is lazy and his tool inserts alt="" by
default.
> The other is  the conscientious author who does the same affirmatively
> knowing that the surrounding text provides the same information?

Yes. I suppose there will always be lazy authors and the best that spec
can do is some kind of damage limitation that on top of designing better
tools, education etc.

> 'no text equivalent' as alt text at least says something, perhaps still
> leaving the reader edgy that he/she has missed something.

It is arguable that there could be times when there is 'no text
equivalent' either needed or wanted.

> 'no text equivalent needed' is a bit cocky IMO, even if true according
> to the author; perhaps a shorter version of
> 'image information provided in associated text' is an improvement?

Perhaps, but is that a user agent issue to a degree? What are others
thoughts?

Cheers

Josh