Re: [css-transitions] transition-delay and reversing

On Tuesday 2014-12-30 18:43 -0800, Estelle Weyl wrote:
> Question about transition-delay: 
> 
> in my testing of transition-delay: when a transtion does’t complete or goes in reverse order, such as when a transition occurs on a hover, and the user mouses out, the transtion delay happens in the reverse order. So it waits before going back to the non-hover state.
> 
> It also happens when the transition has completed,  
> 
> Is that intentional? Should it be explicitly stated under ‘transition-delay’ section.

So transitions don't really have a concept of "reverse order"; when
a transition happens from a :hover state to a non-:hover state, it's
just a transition like any other (in response to a style change like
any other), and the delay happens at the beginning.

That said, the spec has rules to make the transition shorter when a
transition is interrupted halfway through and the new end-state is
the same as the old start-state.  (I'm not sure how interoperably
implemented these rules are, but Gecko does implement them.)

Some of this is explained in
http://dev.w3.org/csswg/css-transitions/#reversing
and in the example (currently example 4) a little bit before that
section.

Do you think the spec should try to say more explicitly that
transitions don't have a concept of running backwards?

-David

-- 
𝄞   L. David Baron                         http://dbaron.org/   𝄂
𝄢   Mozilla                          https://www.mozilla.org/   𝄂
             Before I built a wall I'd ask to know
             What I was walling in or walling out,
             And to whom I was like to give offense.
               - Robert Frost, Mending Wall (1914)