RE: Taking another round at @summary

Jonas Sicking wrote:
> 
> The question should never be why not to have a feature, but why to
> have it. So my question is why should we have @summary in the spec?

Backwards compatibility, and related policies and standards around the
world that mandate @summary's use - a point that Denis has again brought
forward. This question, as Maciej has already pointed out, has been asked
and answered many times previously - I believe the "Reasons Why" now
numbers 27 individual entries. (http://bit.ly/8PX3Ln) 

> As
> far as I can tell all the problems that @summary aimed to solve, are
> already solved by aria-describedby. And solved better than @summary
> does.

Do you have publicly available user-studies or other proof to back that
assertion? Examples in the wild? 

> 
> If you think you could successfully evangelize @summary to be used,
> why not instead spend that effort to evangelize aria-describedby?

Why must that be a binary choice? Could we not do both?

> 
> And my understanding is that most AT tools are working on implementing
> aria. Firefox already supports it. Don't know what the state is in
> other browsers. So it seems likely that tool support is the smaller
> problem here.

In the interest of backward compatibility with HTML 4, does Firefox
currently support the @summary attribute insofar as it is exposed to
adaptive technology? 
Does Firefox plan to no longer supporting some aspects of HTML 4 moving
forward? 
If the answer is that it does and will continue to support @summary as
described in HTML 4, what overhead exactly does it place on any of the
browsers to continue supporting this attribute in HTML5? 
It seems that browser support is the least of our worries. 

Perhaps if Mozilla believed in the value of @summary (and in making that
data available to all users on demand), that browser could lead the way
and improve on the way it exposes @summary values to all who desire it?

"...users over authors over implementors over specifiers over theoretical
purity."
http://www.w3.org/TR/html-design-principles/#priority-of-constituencies 

Jonas Sicking continued:
> 
> Does Quebec really have a law that mandates @summary to be used? While
> that seems like an utterly stupid law I guess I wouldn't put that past
> a government.

Yes, imagine that, a *legislated* accessibility requirement based on a
Standards Recommendation and Success Criteria Techniques authored by...
the W3C.  Crazy government fools indeed. (Better that they ignore an
International body and go the same route as South Korean and make up their
own...; oh, wait...) Do you realize how offensive that statement really
is? (Did you really just call the Quebec government stupid?)

> 
> South Korea mandates a specific encryption algorithm [1]. Do you think
> we should put this encryption algorithm in SSL because of this?

How does this compare with continuing to support an existing HTML
attribute that has been part of the International and 'open' HTML 4
specification for over a decade? Now you are just trailing off into silly
land... Apples to apples please.

JF

Received on Wednesday, 6 January 2010 09:49:36 UTC