Re: ARIA as stop-gap (was Re: Next steps for the ARIA syntax discussion)

"Steven Pemberton" <steven.pemberton@cwi.nl> wrote on 05/28/2008 03:23:33 
PM:
> Let me just say upfront that I think ARIA is needed, and will be needed 
> for some time yet. But let me also answer your question about why we 
> should strive for it to be a stop-gap in the long term in order to get 
> even better accessibility.

We definitely agree about striving for it to be stop gap. The less ARIA is 
needed the better. But I'm extremely skeptical about getting there based 
on what I see today. Realistically, to get beyond the need for ARIA will 
require many things, such as:
1. Getting the major vendors to agree to support this next great thing
2. Agree to the details
3. Get browser vendors to implement it in a compatible way
4. Get users to update to new browsers so authors can use the new 
standards

ARIA got around this adoption problem by implementing in 1 key chain of 
technology along the way (Dojo -> Mozilla -> ATs) and trying to help 
anyone that wanted to join the work. The key was that current web pages 
still work in legacy browsers even when ARIA is not supported -- they just 
won't be accessible. But then, JS widgets aren't accessible to start with 
so nothing lost. Having one major working implementation, with docs, was a 
great way to encourage the larger community to adopt and implement.

Without getting into the details of XForms, or XBL, or whatever, getting 
traction for any of those things will be very, very difficult. How are you 
actually going to get websites to move to these great new things if it 
will break web pages on current browsers? These things don't even 
gracefully degrade. No one wants to write 2 web sites.

If the strategy to get browser & author adoption is just technological 
elegance, you'll never get off the ground. Authors won't write to it 
without browser support and some browser will always lag behind unless 
they're forced into supporting its used in important places. Typical 
chicken & egg.

Again ARIA didn't have this problem because it can just be added to 
current websites without breaking them. We could just start implementing 
it without breaking working stuff.

Almost everyone has worked hard on stuff that didn't live up to potential 
because the adoption strategy was broken. The failure pattern is way too 
predictable, and it's painful to watch.

So in the end I agree we should strive toward removing the need for ARIA. 
I would love to hear any proposal that includes a solid adoption strategy 
that doesn't require luck and good faith in all vendors caring about the 
web. We're probably just dreaming now, and I don't see the point. 

- Aaron