W3C home > Mailing lists > Public > public-fx@w3.org > July to September 2012

Re: [filters] Shading language recommendation

From: Dean Jackson <dino@apple.com>
Date: Thu, 06 Sep 2012 10:48:50 -0700
Cc: David Sheets <kosmo.zb@gmail.com>, "public-fx@w3.org" <public-fx@w3.org>
Message-id: <38794941-60E4-4879-93C7-906E96497179@apple.com>
To: Sylvain Galineau <sylvaing@microsoft.com>

On Sep 5, 2012, at 1:29 PM, Sylvain Galineau <sylvaing@microsoft.com> wrote:

>> Do you not recommend any shading language at all?
> If recommend refers to its definition in RFC 2119, no.
>> Do you not recommend any *specific* shading language?
> If recommend refers to its definition in RFC 2119, no.
>> Do you not recommend any technical means to manage shading language media
>> type divergence?
> I believe the feature should allow for more than one language and such a
> technical mechanism should be defined. (Though I do not have a proposal
> at this time). The spec may also informatively document how such a 
> mechanism can be used with any specific technology the WG deems worth
> documenting. 

For what it is worth, at the moment Apple would object to not requiring support for the variant of GLSL used in WebGL (to be clear, we're already slightly diverging in CSS Filters - not from the language itself though).

However, I do not mind if the first draft of the specification goes out with this topic as an unresolved question or comment. I (hope I) understand Microsoft's position and would like to hear more technical rationale about it, from them and from the community.

The reasoning is that if a shading language becomes part of CSS, it would be much better to adopt the one used by WebGL. We don't want to support more than one language if possible, and feel that there needs to be a common language to all implementations. The rise of WebGL makes this an almost obvious choice. Also, it's an open standard, been shown to work across platforms and a range of device capabilities, and simple enough that it can be converted to platforms that don't natively support it.

Received on Thursday, 6 September 2012 17:49:18 UTC

This archive was generated by hypermail 2.3.1 : Monday, 22 June 2015 03:33:48 UTC