W3C home > Mailing lists > Public > public-fx@w3.org > January to March 2013

Agenda requests for FXTF call

From: Dirk Schulze <dschulze@adobe.com>
Date: Wed, 27 Feb 2013 10:23:19 -0800
To: "public-fx@w3.org" <public-fx@w3.org>
Message-ID: <193C60D1-3A53-43C5-B162-7F46895DA36B@adobe.com>
1. Transformation Matrix proposal
2. CSS3 Transforms
3. Filter Effects
4. CSS Masking


1. Transformation Matrix proposal [1]

On request of the SVG WG, I made a proposal for a Transformation matrix interface. I would like to discuss this proposal and have an agreement how to continue from here.
I looked at current specifications, proposals and libraries and took them into account for this proposal, including SVGMatrix[2], WebKitCSSMatrix[3]/MSCSSMatrix[4], Processing. I asked on the mailing list and got one mail with great feedback. For SVG a proposal like this gets urgent, for CSS Transforms in the next level as well.


2. Pseudo stacking context for 2D transformations

During the SVG F2F we found a possible spec incompatibility between CSS3 Transforms[5] and CSS Compositing and Blending[6]. Each use of transform property (value different then 'none'), creates a new stacking context. CSS Compositing and Blending takes the backdrop up to the last stacking context creation into account. Transforms are heavily used in SVG, which makes it nearly impossible to use blending on content. Is it possible to use the term 'pseudo stacking context' for 2D Transforms? Checking the WebKit code, we do not isolate the drawing for 2D transforms. I assume that other implementations do the same.


3. Filter Effects

- @support rule for filter effects extension [7]
It would be great to check for the existence of certain filter effect extension support in UAs. As an example, support for CSS Shaders could be checked by:

@support filter(webgl) {
}

This was suggested during the SVG F2F and I would like to resolve on the filter function for the @support rule.

- Make filter regions unbound
This proposal came up from time to time. I do no know if we have a resolution on having unbound filter regions. By default they are 10% bigger then the object bounding box. Which is to small for a lot of use cases. Increasing the number does not help all the time.

- Publication of new WD
I am preparing the spec for creating a new WD. It would be great if we can resolve on publishing a new WD.


4. CSS Masking[8]
- Make mask regions unbound
This has the same topic as for Filter Effects. It might be less controversial to make masking regions unbound.

- Use CSS properties x, y, width, height or delay?
The CSS WG resolved on allowing x, y, width and height being CSS properties. So far we do not have a specification text that describes these properties. Should the use of the attributes as presentation attributes be delayed to the next level of the spec?

Greetings,
Dirk


[1] https://dvcs.w3.org/hg/FXTF/raw-file/tip/matrix/index.html
[2] http://www.w3.org/TR/SVG/coords.html#InterfaceSVGMatrix
[3] http://developer.apple.com/library/safari/#documentation/AudioVideo/Reference/WebKitCSSMatrixClassReference/WebKitCSSMatrix/WebKitCSSMatrix.html
[4] http://msdn.microsoft.com/en-us/library/windows/apps/hh453593.aspx
[5] http://dev.w3.org/csswg/css3-transforms/
[6] https://dvcs.w3.org/hg/FXTF/rawfile/tip/compositing/index.html
[7] https://dvcs.w3.org/hg/FXTF/raw-file/tip/filters/index.html#support-condition-filter
[8] https://dvcs.w3.org/hg/FXTF/raw-file/tip/masking/index.html
Received on Wednesday, 27 February 2013 18:23:46 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 27 February 2013 18:23:46 GMT