W3C home > Mailing lists > Public > public-canvas-api@w3.org > July to September 2011

Re: You Got Your SVG in my Canvas! Mmm, Delicious!

From: Charles Pritchard <chuck@jumis.com>
Date: Thu, 07 Jul 2011 11:03:44 -0700
Message-ID: <4E15F500.80508@jumis.com>
To: Doug Schepers <schepers@w3.org>
CC: "Tab Atkins Jr." <jackalmage@gmail.com>, Richard Schwerdtfeger <schwer@us.ibm.com>, "public-canvas-api@w3.org" <public-canvas-api@w3.org>, public-canvas-api-request@w3.org
On 7/7/2011 10:53 AM, Doug Schepers wrote:
> Hi, Tab-
> Tab Atkins Jr. wrote (on 7/7/11 1:14 PM):
>> On Wed, Jul 6, 2011 at 6:58 PM, Richard 
>> Schwerdtfeger<schwer@us.ibm.com>  wrote:
>>> I agree. and title and description are insufficient for dynamically
>>> changing drawings. This is an SVG deficiency but NOT a canvas
>>> deficiency.
>> I don't understand.  Why are they insufficient?  They seem to
>> adequately solve one of the presented use-cases (which I've begun
>> recording 
>> at<http://wiki.whatwg.org/wiki/Active_Image_Accessibility_Use-cases>).
>>   Can you elaborate either why they don't solve that use-case well, or
>> why that use-case is unimportant to solve?
> Thanks for writing those up.  I've linked to them from the wiki page I 
> created on the topic:
>  http://www.w3.org/WAI/PF/HTML/wiki/Canvas_Accessibility_Use_Cases
> They have a different level of granularity, but I think it's useful to 
> discuss all the concrete use cases we have.
There are several use cases in my post, with less emphasis on "low vision":

> I agree with Rich that title and description are not always enough 
> (though I maintain that they sometimes are); obviously, more 
> information is better, such as may be available and readily 
> AT-consumable through ARIA attributes.  But ARIA attributes can be 
> used with SVG as well.
Rich and I would both like to be working in SVG a11y, we're just stuck 
at the moment, due to our commitments
on these few remaining issues in Canvas a11y / the shadow dom.

There are some difficulties with ARIA/HTML interop in SVG that I'd 
certainly like to discuss at
length over at www-svg and/or the public fx group.
>>> Although important, the bounds of an object are only a piece of all
>>> the things to support interoperability. You need roles,
>>> relationships, state change notifications, and so on. We have those
>>> in canvas now. We do not in SVG.
>> What problems are you trying to solve with those things?  The
>> use-cases I've been able to extract so far don't appear to need any of
>> those.  Do you have more use-cases?
> Yes, please share any use cases on either (or both) of the wiki pages, 
> so we can have them easily available for discussion.  This will 
> ultimately be useful for SVG anyway, regardless of what happens with 
> Canvas.
I've not seen a response to my post about use cases. And the use cases I 
very much are based on the roles and relationships available to the 
canvas shadow DOM.

SVG, at the moment, does not work with foreignObject in the manner that 
canvas does
to the shadow DOM.

For example: <path d="..."><input type="button" /></path> is not how 
things are done in SVG.
Proper ARIA support would help that case, as path could be assigned a 
"button" role.

Received on Thursday, 7 July 2011 18:04:09 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:10:32 UTC