- From: Sylvain Galineau <sylvaing@microsoft.com>
- Date: Sat, 4 Aug 2012 01:17:02 +0000
- To: Tab Atkins Jr. <jackalmage@gmail.com>
- CC: "www-style@w3.org" <www-style@w3.org>
[Tab Atkins Jr.:] > > On Fri, Aug 3, 2012 at 3:15 PM, Sylvain Galineau <sylvaing@microsoft.com> > wrote: > > Looking into this open issue [1], I find that both Firefox (Aurora) > > and IE10 ignore keyframe selectors that are <0% or >100% and simply > > run the animation as if they weren't present. > > > > WebKit (Chrome 21) seems to ignore the entire animation of any > > keyframe selector is out of bounds. > > > > (I have no build of Opera that supports css3-animations at the moment). > > > > I propose we specify the behavior implemented by Firefox and IE; it > > seems more consistent with author expectations of what would happen to > an invalid selector. > > > > [1] https://www.w3.org/Bugs/Public/show_bug.cgi?id=14795 > > Agreed. > > (Though it would have been nice, if this weren't constrained by existing > impls, to have it work similarly to, say, radial gradients stops > positioned less than 0%, where the keyframe is still used to generate a > "virtual" 0% keyframe. But we're past that point now.) > Possibly, though in that case you also have to define what happens to the -10% keyframe selector when the author did define 0%. I find the current FF/IE approach to be both simple and reasonable. And yes, this is all getting much harder to change now.
Received on Saturday, 4 August 2012 01:17:35 UTC