Re: About longdesc and table summary (attribute) in HTML5

I have taken this thread off the WHATWG list since the issue has been 
discussed to death there and people will likely not want to see it 
discussed again, but since I promise to reply to all substantial feedback 
sent to that list, I still want to send you a reply.

On Thu, 12 Aug 2010, Sailesh Panchang wrote:
>
> The recent decision to banish the longdesc attribute from HTML spurred 
> me to write this. The argument that won the day it seems was "there are 
> no good examples" of longdesc in use.

That's a dramatic over-simplification of the reasoning used in deciding 
that the feature shouldn't be included.


> Similar arguments were made by those who opposed the American with 
> Disabilities Act: no wheelchair users come into my library so why should 
> I make it accessible?

I agree that that is a bogus argument. It is not the argument used in this 
case, however.


The argument against longdesc="", in a nutshell, is that it is not the 
best way of solving the problem.


> Even today authors / developers fail to code forms and data tables as 
> they should be coded with proper accessibility markup. And there are 
> countless such forms and tables out there. So why not banish the FORM 
> and TABLE element?

When used incorrectly, they _are_ disallowed.


> The problem is most developers / authors do not know what AT is, how AT 
> works, how a blind guy navigates a table, what non-visual access is, 
> what audio user interface is. So obviously you cannot expect them to 
> write imaginative and helpful text alternatives.

If we cannot expect authors to write helpful text alternatives, then 
having a feature that asks them to write helpful text alternatives seems 
like poor language design. :-)


> Just a few days ago I saw a page on which every image had an alt and a 
> longdesc ...  even image links had a longdesc. It only demonstrates that 
> the author / developer does not know how to use the attribute and not 
> that the attribute is flawed.

When a tool is consistently misused, one can blame the user, but at the 
end of the day one must also consider how to fix the tool.

Suppose one were to make a door with a pull handle when the door was in 
fact to be opened by pushing. Would we blame the users if they tried to 
pull the door open and failed? Or would we blame the door builder for 
putting on the wrong kind of handle? At the end of the day, does it matter 
who we blame, so long as the door gets fixed so that people don't get 
confused?

The same applies to Web technologies and accessibility. At the end of the 
day, it doesn't matter who is to blame -- if pages aren't being made 
accessible, we need to fix the technologies to make it more likely that 
they get made accessible.


> So what is needed is more education and outreach on the proper use of 
> longdesc. So do not banish it.

We've had ten years of education with no success. Let's try something else 
now. As Rita Mae Brown once wrote, "Insanity is doing the same thing over 
and over again but expecting different results".

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'

Received on Tuesday, 16 November 2010 02:37:54 UTC