[whatwg] SVG: Accessible Forms

[Charles McCathieNevile]

On Wed, 02 Sep 2009 15:38:00 +0600, ~:'' ありがとうございました  
<j.chetwynd at btinternet.com> wrote:

> Chaals,
>
> thanks for yet another well considered and easy-to-read response*!
> your comments around ARIA and SVG are noted.
>
> however you fail to address the central issue, which as Filipe Sanches  
> wrote me in a private email, and which I believe correctly states the  
> case:
>
>> Implementing it is not trivial and that is why this kind of thing  
>> should be available as a general module. Preferably as part of the spec  
>> in order to get it into every UA.
>
> my query was intended to address this issue in particular. ie as to why  
> whereas accessible html forms had been around for about a decade,  
> accessible svg forms remain extremely hard to implement.

Because for SVG there was nothing formal like ARIA until now. There was  
the idea foreshadowed in the svg-access note[1] a decade ago of using  
metadata to tag objects - this dates from the very early days of what  
became ARIA, in discussions with Lisa Seeman of UBAccess and Rich  
Schwerdtfeger of IBM who followed this work through the spec editing  
process that eventually got us ARIA.

There was also work on using Xforms, for example the experimental work of  
X-smiles and the work of Sebastian Schnitzenbaumer. Unfortunately that  
never got traction for a variety of reasons (not least the failure of  
XHTML, meaning that little effort was made to develop better tools for the  
public to create content).

Effectively this was because there were not enough people working actively  
on making SVG accessible - so we are not much further ahead than we were 8  
years ago. Given the new and renewed interest in SVG by people who  
understand accessibility, I expect to see much more progress in the next 2  
years.

> SVG1.1 has now been implemented by many UA and browsers, but still miss  
> this vital  "user"  functionality.

Yes. So we are only now arriving at the stage of making this really  
possible. With ARIA in the spec, and with at least Opera working on  
implementation, it should become possible.

> There is a structural fault in the W3C process, and that is why process  
> are copied into this thread.

I think that should be a seperate thread - fixing the process and fixing  
the SVG problem are things that need to be done by different people.  
Mixing the two is likely to lead to more discussion about process and less  
action on anything :(

cheers

Chaals

-- 
Charles McCathieNevile  Opera Software, Standards Group
     je parle français -- hablo español -- jeg lærer norsk
http://my.opera.com/chaals       Try Opera: http://www.opera.com