W3C home > Mailing lists > Public > public-html-a11y@w3.org > April 2010

RE: RESOLUTION to modify text alternative change proposal and reject WAI CG's consensus recommendation

From: John Foliot <jfoliot@stanford.edu>
Date: Sun, 11 Apr 2010 14:54:10 -0700 (PDT)
To: "'Laura Carlson'" <laura.lee.carlson@gmail.com>
Cc: "'Janina Sajka'" <janina@rednote.net>, "'HTML Accessibility Task Force'" <public-html-a11y@w3.org>, "'Michael\(tm\) Smith'" <mike@w3.org>, "'Michael Cooper'" <cooper@w3.org>, "'Judy Brewer'" <jbrewer@w3.org>
Message-ID: <02fe01cad9c1$82a8da60$87fa8f20$@edu>
Laura Carlson wrote:
> Hi John,
>
> > why struggling over a word is worth the effort
>
> If the word is not "error", there will be no result.

OK, so we call it an ERROR. Now what? What *practical* result does it get 
us?

The image still displays in the browser.

Even calling it a "super-mega-show-stopping-error-to-end-all-errors" still 
gets the same result in the browsers, so it's just a word IMHO; the net 
result is that the image still displays. What have we 'won', what have we 
gained?


> Developers pay
> attention to errors not warnings.

If the final outcome is that the error has zero impact on how the image/page 
renders visually on the screen, then why will they 'pay attention'?  *I* 
know what the problem is, as do you and everyone following this thread, but 
for the average, non-educated content creator, she doesn't see a problem - 
she sees an image on the screen.

What is the *cost* of that problem to the content creator? What is the cost 
to my Dad, uploading pictures to Flickr? To the third-world child, with 
their new green and white $100.00 computer, making a web page about his 
village? Without practical outcomes, this is an exercise in 
angels-on-a-pinhead. What we really call that outcome boils down to two 
things - success or failure.


> >> This isn't about making deals.
> > Sure it is - that's the nature of compromise and consensus
>
> Consensus is about being able to "live with" something.

Right. I can live with changing the *name* of what it is we are talking 
about, if at the end of the day we get all of the other behaviors and 
recommendations in your Change Proposal. Those requirements, to me, are what 
we should be fighting for, not what we name them. Getting MUST language into 
the spec is where the big win is - that editors and validators that 
encounter broken images MUST enforce a teachable moment and provide guidance 
and assistance in remediation/repair, so at the end of the day we have 
neither an ERROR nor a WARNING, we have a complete <img>!


> Some things
> can't be lived with.  Without both @src and a text alternative the
> <img> element is broken. Making them both a  warning doesn't make
> sense.  Both are needed for an image to be complete.

Logically, emotionally, philosophically I am 100% in agreement with you. 
Unfortunately, mechanically this is not true, and we are not going to change 
that reality either. <img src=""> works ('complete' or otherwise) in all of 
the browsers today, and will continue to work so long as the value of @src 
is intact. That ship has sailed long ago.



> Both are on the
> same level.   Lowering standards for one and not the other is
> discrimination. I can't live with that. WAI CG's consensus document
> doesn't live with that.

Laura. This isn’t about _us_ lowering our standards - that has been done for 
us already by the browsers and there is no going back; we won’t get 
draconian fail and honestly that is the only real penalty that would make 
@src and @alt truly equal. (The browsers could use the same argument too 
BTW - that introducing draconian fail is a lowering of their well 
established rendering standards to a point that they cannot accept)

I HATE that as much as you do, as much as we *all* do. But it is a reality 
we must come to accept.

Insisting that we call this an ERROR, *fighting* for it at HTML WG, is a 
battle we will not win; or conversely we will all agree to call it an ERROR, 
but (with a wink and a nod) in true Orwellian fashion we will have a 
situation where all ERRORS are considered equal, but some are more equal 
than others...

JF
Received on Sunday, 11 April 2010 21:54:46 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 04:42:07 GMT