W3C home > Mailing lists > Public > public-fx@w3.org > October to December 2011

Re: Merging CSS Shaders proposal into Filter Effects

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Sun, 30 Oct 2011 17:48:36 -0700
Message-ID: <CAAWBYDDcMxNUdYG5k2d0PJJ1-0yH7jtotAdD1DWrCEoJ058_vQ@mail.gmail.com>
To: Dean Jackson <dino@apple.com>
Cc: public-fx@w3.org
On Sun, Oct 30, 2011 at 5:34 PM, Dean Jackson <dino@apple.com> wrote:
> On 30/10/2011, at 5:19 PM, Tab Atkins Jr. wrote:
>> On Sun, Oct 30, 2011 at 5:04 PM, Dean Jackson <dino@apple.com> wrote:
>>> The publication of the first draft of Filter Effects is underway and expected in the next few weeks.
>>>
>>> I propose that the CSS Shaders proposal from Adobe, Apple and Opera be merged into the Editor's Draft of the Filters spec. This won't be part of the first officially published draft.
>>>
>>> Does anyone have strong objections to this? I expect one contentious part of the spec is the choice of shading language, but I think that's been called out.
>>
>> I object to this.  Shaders is way more contentious than just that
>> issue, and I'd prefer not to slow down Filters (which are generally
>> okay) by pulling in something with serious security concerns still.
>> (Note: I love Shaders and want them to proceed.)
>
> Would you be ok with two versions of the spec? One with and one without shaders. We don't have a well-known versioning identifier like "CSS3", but I guess this could be a 1.0 and 2.0.

That just seems vulnerable to things drifting out of sync.  It's also
bizarre to work on a level 2 simultaneously with level 1 when they're
both at the same maturity level.


> The alternative would be a separate specification completely, but this is a little weird because it's using the same property.

I don't think that's particularly weird.  The Image Values spec
defines new types of values for a handful of properties.  Shaders is a
pretty unique and large type of value, and can survive being defined
in its own spec.  It would be like having Gradients be their own spec,
which would be perfectly appropriate (it just happened that we had a
good spec with very little else in it that could take it).

~TJ
Received on Monday, 31 October 2011 00:49:33 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 31 October 2011 00:49:33 GMT