Re: [css3-animations] Splitting up concepts of "animations"

On Apr 6, 2010, at 9:15 AM, Tab Atkins Jr. wrote:

> [moving Simon's email directly to this thread, for better organization]
> 
> On Tue, Apr 6, 2010 at 9:04 AM, Simon Fraser <smfr@me.com> wrote:
>> Doesn't this have al the same issues as Hakon's on-enter, on-exit that have
>> been pointed out already?
> 
> As far as I can tell, no (this probably wasn't clear in the original
> email you responded to, but I've hopefully stated things better here).
> 
> The notion of "entry" I use is identical to the notion used in the
> current draft - a play-in animation runs when it first gets added to
> an element that doesn't already have that animation as a play-in.  If
> there are any problems with this, then they apply equally to the
> current Animations draft.  play-out animations are defined similarly -
> "exit" occurs when an element *loses* an animation.
> 
> This isn't based on selectors or anything.  The notion of a selector
> defining state still exists, but only conceptually, as is appropriate.
> Actual 'state' is tracked through the presence or absence of
> animations only.

I think this is the same conclusion that Hakon came to at the end of
<http://lists.w3.org/Archives/Public/www-style/2010Apr/0086.html>:

> So, to conclude, the effects are not tied to "states". Rather, the
> effect trigger when the value of the 'effect' property changes for a
> given element. When this happens, the respective 'on-exit' and
> 'on-entry' effects will be shown.

And my response was that on-exit (or play-out) can have unwanted
side effects if you're allowed to run infinite animations:

<http://lists.w3.org/Archives/Public/www-style/2010Apr/0090.html>

but I think you're disallowing that, in this case? Even so, what if
a play-out animation runs for 1000000000s?

Simon