RE: [css3-animations] animation-reverse: none | all | even | odd

[Lea Verou:]
> 
> Let me start by saying that I understand it's probably too late to change
> this, but I think the change I propose is considerably better than the
> existing syntax, so it might be worth doing so.
> 
> `animation-direction` currently [1] accepts 4 values: normal | reverse |
> alternate | alternate-reverse. Attached is a graphic that demonstrates all
> four, in a linear progression over time.
> 
> As you maybe have noticed, their effect basically is:
> alternate: Reverse even iterations
> reverse: Reverse all iterations
> alternate-reverse: Reverse all iterations, then reverse even iterations
> again
> 
> As you can see, alternate-reverse is hard to understand, because it's
> basically a double negative. The end result of `alternate-reverse` is
> basically the same as `alternate`, with the difference that odd iterations
> are reversed instead of even ones. So, this gives us these four possible
> things we can do with `animation-direction`, respectively:
> normal
> reverse even iterations
> reverse all iterations
> reverse odd iterations
> 
> Therefore, a property that would capture this better, and would be easier
> to understand, would be called `animation-reverse` (or `animation-
> iteration-reverse` although that implies an `animation-iteration`
> shorthand which is misleading) and have the states:
> none (corresponds to animation-direction: normal) all (corresponds to
> animation-direction: reverse) even (corresponds to animation-direction:
> alternate) odd (corresponds to animation-direction: alternate-reverse)
> 
> Not only this is more elegant and easier to understand, but is also easier
> to extend in the future (e.g. to reverse only every third iteration), if
> such a need arises.
> 
> Thoughts?
> 
> [1]:
> http://www.w3.org/TR/2012/WD-css3-animations-20120403/#animation-
> direction-property
> 
Yes, this does feel much cleaner and simpler. There already is content using 
animation-direction though. And here we get back in part of the prefix controversy 
- 'Prefixes are there to enable this kind of change!'/'We shouldn't go arbitrarily 
change what people are already using without a good reason!' - and as this is one 
of the specs the group wants to unprefix yesterday this co-editor is torn. We should
talk about this at the f2f next week.

Received on Saturday, 5 May 2012 19:39:38 UTC