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

On 6/30/2011 2:16 PM, Doug Schepers wrote:
> Tab Atkins Jr. wrote (on 06/30/2011 02:07 AM):
>>
>> While I like the general shape of several of your ideas,
>
> Thanks.
Doug, I'd like to point out that drawImage(svgElement) as well
as style="background: url(#cssCanvasId)" are available in WebKit,
something similar is available in Firefox, and IE has had support
for background css items since at least IE6 via ActiveX.

They're not -tight- integration, but they are a start. The CSS Canvas
object introduced by WebKit may be something that can be
incorporated into FX in the near future. Mozilla has a more general
element() method for CSS.
http://lists.w3.org/Archives/Public/public-fx/2010OctDec/0067.html

There are some difficult issues when it comes to managing the
resolution of the backing bitmap / bitmaps.

http://lists.w3.org/Archives/Public/public-fx/2010OctDec/0066.html
https://lists.webkit.org/pipermail/webkit-dev/2011-March/016130.html
http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2010-December/029475.html

I am all for Canvas+SVG harmony; I think FX can get there sooner
via the CSS Canvas concept put forward by WebKit. It's quite similar
to HTML Canvas, a little different.

Two issues I worry about with "waiting" on Canvas+SVG harmony:

one, is that it's going to take awhile, possibly years. That's no good
for a11y in the present.

two, HTML+SVG harmony is still a little tricky. I think the FX
group is correct in starting with CSS integration.

I'd like to point out that, should Canvas a11y be handled appropriately
now, such techniques will likely fit well into Canvas+SVG profile.

For example: CSS transforms and Canvas a11y -should- work
automatically. CSS 3d transforms and Canvas -do- currently
work, and that's neat.

The DPI attributes exposed through window.screen help with those issues 
in HTML:
http://msdn.microsoft.com/en-us/library/ms535868(v=vs.85).aspx

SVG is a bit more dynamic in its needs.

>
>
>> I must make
>> the same requests as I did of other people - we need to know precisely
>> what problems are to be solved before we can decide how to solve them.
>
> I don't understand.
>
> I'm happy to answer your question, because I think there are many 
> advantages for general use, as well as for accessibility, but it's not 
> clear what you consider a "problem to be solved."  I've listed several 
> benefits on my blog post, but it seems like you're looking for 
> something else.
>
> Can we get a bullet-point listing of the kinds of problem sets that 
> you feel are appropriate for addressing, along with concrete 
> examples?  I've read all of your posts on this topic, and I don't 
> understand what exactly you're asking for.  I'm sure the details are 
> hidden there, but the wordiness makes my eyes slide right over them.
Tab provided two replies, when I requested the same:

short intro:
http://lists.w3.org/Archives/Public/public-html/2011Jun/0424.html
longer description:
http://lists.w3.org/Archives/Public/public-html/2011Jun/0420.html


-Charles

Received on Thursday, 30 June 2011 21:42:00 UTC