Re: Comments on CSS Transitions

Boris Zbarsky:
> Sort of. It's a bit weird, because it actually does change the
> specified and computed value as it animates. Last I checked, the
> fit between the model as sort-of described in SMIL and the way
> CSS actually works was pretty poor. It was causing us some major
> implementation headaches.

IIRC in Batik I implemented getting of underlying CSS values by removing
the animated property from the override style sheet, getting the
computed value, then re-adding the property to the override style sheet.
So I’d agree, the model’s not perfect.

I wonder if other implementations ignore that and just do something
special in the cascade instead of messing with the override style?

-- 
Cameron McCormack ≝ http://mcc.id.au/