- From: Tab Atkins Jr. <jackalmage@gmail.com>
- Date: Sat, 3 Apr 2010 09:59:47 -0700
- To: Håkon Wium Lie <howcome@opera.com>
- Cc: www-style@w3.org
On Sat, Apr 3, 2010 at 1:06 AM, Håkon Wium Lie <howcome@opera.com> wrote:
> One difference between this proposal and the current specs is that it
> matters where the 'effect' property is set. For example, in the
> current specs, the following two examples have identical results (I
> believe):
>
> [example 9]
>
> a {
> color: blue;
> transition: color 1s;
> }
>
> a:hover {
> color: red;
> }
>
> or:
>
> [example 10]
>
> a {
> color: blue;
> }
>
> a:hover {
> color: red;
> transition: color 1s;
> }
Small comment; I haven't yet fully digested the implications of the
rest of this.
The two examples aren't equivalent. The element has be in a
'transition-ready' state when the property that you're transitioning
"starts to change". Your first example transitions color both ways,
but your second only transitions when the element is in :hover and the
color changes (possibly through losing :hover, since the "starts to
change" bit happens before :hover is actually lost).
Don't know how this might affect the effect proposal, or if there are
any implications of changing the model as you describe (where, I
think, having the effect set on either the starting or ending state
works?).
~TJ
Received on Saturday, 3 April 2010 17:00:40 UTC