W3C home > Mailing lists > Public > www-svg@w3.org > May 2011

RE: [SVGIntegration] Too many referencing modes

From: David Dailey <ddailey@zoominternet.net>
Date: Fri, 13 May 2011 08:12:58 -0400
To: "'Boris Zbarsky'" <bzbarsky@MIT.EDU>, "'Doug Schepers'" <schepers@w3.org>
Cc: "'Tab Atkins Jr.'" <jackalmage@gmail.com>, "'www-svg'" <www-svg@w3.org>
Message-ID: <000901cc1167$18e3fcc0$4aabf640$@net>
The privacy issues discussed in the thread at https://bugzilla.mozilla.org/show_bug.cgi?id=628747 seem similar to those involved in allowing <canvas> to access content from an SVG source. I gather the convergence from the Mozilla team is that disallowing <img> to take its SVG source from different domains resolves those issues amicably. Would it not then make sense to allow the same solution to be applied to <canvas>? It would provide much-needed access to pixel data within SVG's though I would be even happier with a method like getImageData that were directly usable from within SVG (and not having to use HTML for this purpose).

Cheers
David

-----Original Message-----
From: www-svg-request@w3.org [mailto:www-svg-request@w3.org] On Behalf Of Boris Zbarsky
Sent: Friday, May 13, 2011 1:14 AM
To: Doug Schepers
Cc: Tab Atkins Jr.; www-svg
Subject: Re: [SVGIntegration] Too many referencing modes

On 5/13/11 12:26 AM, Doug Schepers wrote:
>> From Gecko's perspective, "Animated Mode" is NOT acceptable for
>> <html:img>, and not acceptable for <svg:image> for the same reasons.
>
> I've heard differently from other browser vendors. What's your rationale?

See https://bugzilla.mozilla.org/show_bug.cgi?id=628747

If other browser vendors choose to expose their users to that sort of 
privacy leak, I guess that's up to them.  But the spec shouldn't 
recommend using this mode for things like <html:img> without 
highlighting the possible issues that result.

Did this never get brought up on this list?  The plan was to bring it up 
here...

>> We also have no plans to do different things for <img> and background
>> images (hence there is no point in "Static Mode" from our point of view;
>> it's not like we prevent animated GIFs or APNG in CSS backgrounds, so
>> why would we prohibit declarative animation of SVG?).
>
> Okay. Again, others have been of a different opinion

OK.  Again, I have no problem with the mode existing so much (I can 
always just ignore its existence as an implementor); I have problems 
with the spec recommending, for reasons that are unclear to me, that 
animated SVG not be supported in CSS background images.

-Boris
Received on Friday, 13 May 2011 12:13:30 GMT

This archive was generated by hypermail 2.3.1 : Friday, 8 March 2013 15:54:48 GMT