W3C home > Mailing lists > Public > www-style@w3.org > May 2012

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

From: Sylvain Galineau <sylvaing@microsoft.com>
Date: Sun, 6 May 2012 14:09:18 +0000
To: Lea Verou <leaverou@gmail.com>, Tab Atkins Jr. <jackalmage@gmail.com>
CC: www-style list <www-style@w3.org>
Message-ID: <3C4041FF83E1E04A986B6DC50F0178290A34A5D7@TK5EX14MBXC263.redmond.corp.microsoft.com>

[Lea Verou:]
> 
> On 6/5/12 15:56, Tab Atkins Jr. wrote:
> > I disagree, actually.  I think that both "reverse" and "alternate" are
> > very clear names.  "alternate-reverse" isn't ideal, but it's not bad
> > either, and I don't know how to say it more cleanly.
> `reverse` is, but I disagree about the others. `alternate` could mean many
> things (what alternates? how? which iterations are reversed?), and without
> reading the spec, I’d imagine it’s hard to guess what it does just from
> the syntax.
> `alternate-reverse` is the worst. The naming is completely detached from
> what it does (reversing odd iterations).
> `animation-reverse` (or `animation-iteration-reverse`) is straightforward,
> not prone to misinterpretation and much more extensible.
> > Yes, Animations is no longer in the "we can make aesthetic changes"
> > stage.  It left that a long time ago, we just didn't finish the spec
> > before that happened, like we're supposed to.
> In general, yes. But I’d argue that some properties are not used as much
> in the wild as animations in general. `animation-direction`, `animation-
> play-state` and `animation-fill-mode` fall in that bucket.
> However, without data, we’re all just speculating.
> 
We're not all 'just' speculating :) The number of sites using it in production is
small, but it includes popular sites such as Skype's, Airbnb's or Ars Technica.
And the web snapshot I'm looking at is a few months old.

I agree with Tab that the nature of this change is essentially cosmetic (as a rule
people who would rather not take this kind of change will call it cosmetic; those 
who would will call it usability...). As such I think he and I believe it has a much 
lower priority at this stage than exposing the reverse and alternate-reverse capabilities
which we added to the spec very recently. And while we could make the case that we 
shouldn't go to Last Call without these values - and the group essentially did just that 
a few weeks ago - some of us are comfortable going to Last Call with the current syntax
for this property. This is where I lean as well. I'm as prone to agonizing over the 
perfect vs. the merely good as anyone else but this is one of those cases where what we 
have is in fact good enough; and the benefits of the change are not clear or measurable 
enough to matter. But we'll talk about it this week.



Received on Sunday, 6 May 2012 14:09:53 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:53 GMT