W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > July to September 2012

Re: Canvas and ARIA alternatives

From: Ramón Corominas <listas@ramoncorominas.com>
Date: Wed, 01 Aug 2012 22:02:32 +0200
Message-ID: <50198B58.5040908@ramoncorominas.com>
To: Ian Sharpe <isforums@manx.net>
CC: 'WAI Interest Group' <w3c-wai-ig@w3.org>
Hi, Ian and all.

 > Apologies if this is obvious to others but could somebody please
 > explain to me why anyone would choose to use the canvas element
 > to handle user input over the designated input elements?
 > Have I missed something?
 > I appreciate that designers tend to have a disproportionate influence
 > over user interfaces and understandably so. And I understand that
 > this in > conjunction with advances in technology has driven the
 > development of a miriad of custom widgets which has then required
 > additional support through ARIA for accessibility purposes. But
 > this just seems unnecessary to me and is only going to lead to
 > increased complexity and reduced accessibility in my view.

You are making the assumption that "input elements" are something that 
everyone should use exactly the same way, with almost no room for new UI 
ideas or innovation. But many designers (and interaction designers) want 
to experiment and create new user experiences, or simply want to mimic 
user interfaces that already exist in other environments that cannot be 
achieved through "normal" inputs.

 > I'm all for freedom to be creative up to a point but surely there's a
 > time when somebody has to push back on the basis that if such
 > approaches are to be adopted, accessibility is going to be
 > significantly compromised.

Not necessarily. For example, I can imagine 3D interfaces that are not 
achievable through normal HTML & CSS (I'm not saying that this is a 
"common" case, but it can happen). Let's say that we want to create a 
F1-devoted website with a 3D representation of a F1 car, where users can 
select different parts of the car to change or view their parameters; 
when the user selects a part, the camera pans and rotates to show its 
detail in the 3D view (and maybe the data in a separate region of the 
page). As far as I knoew, this kind of interface could only be achieved 
using a canvas or similar approach. And it can be perfectly accessible, 
too. For example, using WAI-ARIA we can tell the screen reader that it 
is a simple select or radiogroup, and we can also provide keyboard 
accessibility. So, why should the designer discard the 3D interface?

And, thinking in the future of HTML5+CSS3+JS games this situation will 
happen very often, so it is better to be prepared and provide solutions 
and tools for accessibility. In my opinion, we should focus our effort 
on ensuring that these type of interfaces can still be made accessibile, 
and not on "prohibiting" their usage. Prohibitions or restrictions only 
lead to the wrong idea that accessibility is bad for design, which in 
turn leads to lower interest in accessibility. If there is a conflict, 
designers will not discard their creative ideas, they will discard 
accessibility instead.

If, in contrast, we say: "hey, of course you can do that and still be 
accessible, just do it this way...", many designers will accept the 
challenge and create more accessible things.

And, maybe, some of them will become our best accessibility evangelists, 
trying to demonstrate their skills and showing the world the amazing 
things that can be done.

Received on Wednesday, 1 August 2012 20:03:31 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:36:40 UTC