W3C home > Mailing lists > Public > www-svg@w3.org > October 2007

Re: ARIA and the user's rôle

From: Doug Schepers <schepers@w3.org>
Date: Fri, 05 Oct 2007 08:56:03 -0400
Message-ID: <47063463.90103@w3.org>
To: ~:'' ありがとうございました。 <j.chetwynd@btinternet.com>
Cc: SVG List <www-svg@w3.org>

Hi, Jonathan-

In an earlier post, I explicitly included 2 links for those not familiar 
with the background of Aria.  I feel certain that had you read those, 
most of your questions would have been answered; refer specifically 
Aaron's FAQ, written in very plain language, with entries such as [1], 
[2], and [3].

I appreciate your strong advocacy, but I have to politely insist that 
you familiarize yourself with the technical material before challenging 
its premises (particularly if you're going to direct the email straight 
at me).  I'll explain briefly inline, but I expect you to do your own 
homework.  The SVG WG has very limited resources, and way too much to 
do, and you have to meet us halfway.

(As a side note, when bright lights like Dave Raggett and Anne van 
Kesteren *agree* on something, and when there is a huge body of work 
done by equally bright people, and there is a market push to get this 
all implemented in a timely manner in 2 of the major browsers... and as 
a kicker, when it relates to /accessibility/, that red-headed stepchild 
of software development... you might presume that there might be some 
substance to it.  It's a win, lighten up.)

~:'' ありがとうございました。 wrote (on 10/5/2007 2:37 AM):
> as you will know I am interest in why W3 working groups fail to engage 
> the naive user in creating specifications.
> Outcomes include such questions as:
> why CSS user style sheets aren't in general popular use?

Since SVG is a presentation layer with a level of abstraction above the 
implicit semantics of a text-only format like HTML (for which CSS was 
designed), user style sheet would be all but useless (other than 
possibly specifying, say, the appearance of links and the default zoom 
level), a topic we've discussed many times before.  As for why they 
aren't used widely for HTML, I could only speculate, but it's not in the 
scope of SVG.

> why is the easy to use publishing software, corporate owned?

Because it was written by those corporations?  Because they spent hordes 
of money (supplied by consumers) developing it so that it was easy to 
use?  What kind of silly question is that?

> etc
> in the case of ARIA much work has been contributed concerning adding 
> semantic information to documents.

That's an oversimplification.  It's not merely semantic information, but 
categorized contextual behaviors.

> Where are the test cases and documentation describing how the naive 
> user, possibly with (mild?) learning disabilities,  will engage usefully 
> with the final products?

ARIA, in its present form, is largely a taxonomy of widget roles and 
behavior, so any user would engage with ARIA-informed content in much 
the same way as they would with existing controls common on the Web and 
on desktops.  That's the point.  It lets you create custom controls that 
have the same semantic context and behavior as existing native controls, 
which is impossible today.

> how is it proposed that the non-developer author aria compliant documents?

Please read the FAQ.

> Has any user testing (whatsoever) been undertaken?

I'm sure that some research has been done regarding how a variety of 
users interact with common controls, but I'm not qualified (nor do I 
have the time or resources) to produce pointers to that research.

But the SVG WG is not the one doing the ARIA research, so this isn't 
really the place to ask most of your questions.  We're merely talking 
about including the 'role' attribute, to take advantage of the huge 
efforts that others have done.


-Doug Schepers
W3C Staff Contact, SVG, CDF, and WebAPI
Received on Friday, 5 October 2007 12:56:23 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 8 March 2017 09:47:12 UTC