Re: [css-compositing] Request to move Compositing and Blending spec to CR

On Thu, Dec 12, 2013 at 6:47 PM, James Robinson <jamesr@google.com> wrote:

>
>
>
> On Thu, Dec 12, 2013 at 6:09 PM, Rik Cabanier <cabanier@gmail.com> wrote:
>
>> The paragraph in question describes a compositing model that is not used
>>> by canvas or HTML or CSS or SVG any other web technology and that there are
>>> no plans to use in any web technology.  in way way do you think that it
>>> *is* helpful to have in a web standard?
>>>
>>
>> There are plans. CSS compositing was going to use clip-to-self which is
>> why it was deferred to level 2; it required too many changes which would
>> slow down progress.
>>
>
> I think that would be a bad idea, for the reasons in my first reply.
>

Applications implement clip-to-self compositing today and don't seem to
suffer from the problems you're describing.
For instance, knockout groups in PDF have this behavior. Photoshop and
After Effects do this under the hood as well.


> Even if that was not the case, it's good to call out that there are
>> different compositing modes and that canvas is using a particular one.
>>
>
> Do it on a blog or in a book or something other than in a web standard
> spec, in that case.  We don't put all text that is potentially useful in
> specs especially when it's describing behavior that is specifically *never*
> used on the web.
>

No, the spec should not refer to blogs. Also, this is not 'potentially'
useful as the absence of this description has caused confusion in the past.

Another very important reason is also that if this property/behavior is
included in the spec, the W3C patent policy will apply.


> That section of the spec has already come in handy when talking to browser
>> developers (ie the skia team) and the graphics driver (ie NVidia)
>>
>