RE: [css3-3d-transforms] z-index and transform-style:flat, z position follow-ups

Really attaching the testcase this time.

> -----Original Message-----
> From: Sylvain Galineau
> Sent: Friday, October 29, 2010 2:26 PM
> To: www-style@w3.org
> Subject: [css3-3d-transforms] z-index and transform-style:flat, z
> position follow-ups
> 
> Following up on http://lists.w3.org/Archives/Public/www-
> style/2010Sep/0253.html.
> 
> 1. transform-style flat and painting order
> 
> In the attached testcase, three overlapping sibling elements in a
> transform-style:flat
> container are translated to various Z positions. Their CSS painting
> order is A, B, C i.e.
> C above B above A.
> 
> In Safari, B is above A and C by virtue of its higher Z position. C is
> behind A due to
> its negative Z position. This remains true regardless of the z-index
> that's specified,
> if any.
> 
> In other words, the Z translation overrides the painting order.
> 
> As Simon pointed out in his message:
> 
> # When transform-style is 'flatten', you can consider 3d transforms to
> be just a painting effect
> # (like 2d transforms), and to not affect the rendering order. (Note
> that WebKit does not currently
> # behave this way; it depth-sorts sibling elements with 3D transforms
> in this case).
> 
> Are we saying that in this case the rendering should be the same as one
> gets with a browser that
> does not support 3D Transforms ?
> 
> 2. Z position
> 
> # The Z position of an element is computed from the center of the
> border box.
> 
> Always, regardless of transform-origin ?
> 
> 
> 
>