Re: different scope of view in UI design on Canvas

Frank is correct about "easier" click handling:
<button><canvas></canvas></button>

That one is quite easy to do.


I don't think we can say what most authors will do, in the future tense.
At this point, the concept of "desktop quality" web applications is 
still rather new.

We can look at mobile applications, and the hundreds of thousands of 
applications on sale for iPhone.

Though I do not have the numbers in front of me, I have read that a 
significant number of app vendors migrate
from the UI components provided by Apple, to their own UI components, 
once they've released a few applications.

Accessibility hooks are often lost in that transition, simply out of 
not-knowing they exist.

Most canvas authors currently do their own click handling, I've rarely 
seen <button> used in practice.
We use it temporarily, when introducing new components to our UI. 
Basically, we use it as a prototyping feature.


-Charles

On 9/24/2011 11:05 AM, Frank Olivier wrote:
>
> IMO, most authors will use canvas to create a single UI control* 
> (positioning it in html/css like any other input control) – this makes 
> the click handling story much easier, as you are not attempting to do 
> click handling for a complex layered windowed UI.
>
> *There could, of course, be several of these controls on the page.
>
> *From:*Charles Pritchard [mailto:chuck@jumis.com]
> *Sent:* Saturday, September 24, 2011 11:03 AM
> *To:* paniz alipour
> *Cc:* Frank Olivier; Canvas; Richard Schwerdtfeger; Steve Faulkner; 
> Benjamin Hawkes-Lewis; Cynthia Shelly
> *Subject:* Re: different scope of view in UI design on Canvas
>
> http://lists.w3.org/Archives/Public/public-canvas-api/2010JanMar/0147.html
>
> There are three methods for handling clicks: using a bitmap (trading 
> cpu for ram), using an object with bounding box coordinates,
> or using the browser's hit testing and layering the elements atop of 
> the canvas.
>
> Paniz, very few authors are using Canvas for elements, at this time.
>
> -Charles
>
> On 9/24/2011 10:57 AM, Frank Olivier wrote:
>
> Not sure I understand your question…
>
> I sent an example to the canvas a11y discussion a while ago – 
> basically, the author has to handle onclick events on the canvas 
> element, and then set the state of the radio/checkbox so that the 
> screen reader can read that. Likewise, when the screen reader changes 
> the radio/checkbox state, the author show redraw the canvas element to 
> indicate the new state.
>
> Thanks
> Frank
>
> *From:*paniz alipour [mailto:alipourpaniz@gmail.com]
> *Sent:* Saturday, September 24, 2011 1:51 AM
> *To:* Charles Pritchard; Canvas; Richard Schwerdtfeger; Steve 
> Faulkner; Frank Olivier; Benjamin Hawkes-Lewis; Cynthia Shelly
> *Subject:* different scope of view in UI design on Canvas
>
> Hello All,
>
> I was thinking about a problem if you had a button a checkbox a radio 
> button in Canvas ,
>
> How would you handle click events of each one of them? I need to know 
> different scope of view and your thought in programming.
>
> please tell me.
>
> Thanks in advance
>
> -- 
> Paniz Alipour
>