Intent to Ship: Unprefixed CSS Animations

102 views
Skip to first unread message

Alan Cutter

unread,
Mar 17, 2015, 3:39:27 AM3/17/15
to blink-dev

Contact emails

alanc...@chromium.org

Spec

http://dev.w3.org/csswg/css-animations-1/

Summary

This proposal is to ship the unprefixed animation-* properties and @keyframes at-rule.

Chrome has shipped the -webkit-* prefixed versions of these since 2009.

Link to “Intent to Implement” blink-dev discussion

https://groups.google.com/a/chromium.org/d/msg/blink-dev/BytHPljnifk/nF91Krd22DcJ

Is this feature supported on all six Blink platforms (Windows, Mac, Linux, Chrome OS, Android, and Android WebView)?

Yes.

Debuggability

sa...@chromium.org is currently working on DevTools support for animations.

Timeline slowdown: http://blog.chromium.org/2015/01/chrome-41-beta-new-es6-features-and.html

Cubic-bezier timing function editor: https://www.youtube.com/watch?v=TBDzHRa6B5I

Compatibility Risk

Low, every other engine has shipped unprefixed CSS Animations.

The specification is still evolving, for example it has been agreed that all properties are valid for animation though no browser has yet implemented this. We will continue to update CSS Animation’s behaviour as the specification and other browser implementations evolve.

One risk is that sites may be using -webkit-filter in @-webkit-keyframes and unprefixed filter in @keyframes. Because we don’t yet support the unprefixed filter property those animations will cease to animate filter. Such sites are unlikely to become broken in this scenario as filter is primarily a cosmetic effect.

-webkit-filter is present in 0.8% of all animations (including transitions): https://www.chromestatus.com/metrics/css/timeline/animated/413

OWP launch tracking bug?

https://code.google.com/p/chromium/issues/detail?id=154771

Link to entry on the feature dashboard

https://www.chromestatus.com/feature/6121990213599232

PhistucK

unread,
Mar 17, 2015, 3:47:47 AM3/17/15
to Alan Cutter, blink-dev
This is great!
Perhaps -webkit-filter can also be unprefixed shortly? Someone mentioned this is not a lot of work, I think.


PhistucK

To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+...@chromium.org.

Dimitri Glazkov

unread,
Mar 17, 2015, 11:43:19 AM3/17/15
to Alan Cutter, blink-dev
LGTM.

Chris Harrelson

unread,
Mar 17, 2015, 4:06:45 PM3/17/15
to Dimitri Glazkov, Alan Cutter, blink-dev
+1 to PhistucK's comment. Alan, is fixing that feasible?

Alexis Menard

unread,
Mar 17, 2015, 4:36:58 PM3/17/15
to Chris Harrelson, Dimitri Glazkov, Alan Cutter, blink-dev
LGTM

Douglas Stockwell

unread,
Mar 17, 2015, 5:21:11 PM3/17/15
to Chris Harrelson, Dimitri Glazkov, Alan Cutter, blink-dev
'-webkit-filter' is certainly on the list of properties to unprefix, it's non-trivial as there already exists a 'filter' property which targets SVG elements.

I do not think we should block unprefixed animations on filter. The animation numbers are low and unlike transform (which we did block on) it does not affect layout.

- Doug

Chris Harrelson

unread,
Mar 17, 2015, 7:44:41 PM3/17/15
to Douglas Stockwell, Dimitri Glazkov, Alan Cutter, blink-dev
Ok.

LGTM2

Philip Jägenstedt

unread,
Mar 17, 2015, 11:00:45 PM3/17/15
to Chris Harrelson, Douglas Stockwell, Dimitri Glazkov, Alan Cutter, blink-dev
LGTM3

Erik Dahlström

unread,
Mar 18, 2015, 4:45:18 AM3/18/15
to Chris Harrelson, Dimitri Glazkov, Alan Cutter, Douglas Stockwell, blink-dev
On Tue, 17 Mar 2015 22:21:03 +0100, Douglas Stockwell
<dstoc...@chromium.org> wrote:

> '-webkit-filter' is certainly on the list of properties to unprefix, it's
> non-trivial as there already exists a 'filter' property which targets SVG
> elements.

The Filter Effects spec[1] requires that 'filter' (which should be merged
with '-webkit-filter') applies to all elements, the current implementation
status is that '-webkit-filter' doesn't apply to svg elements, and
'filter' doesn't apply to html elements. Fixing that is a requirement for
unprefixing as far as I'm concerned, otherwise existing web content will
break. See [2] for a list of issues that remain to be fixed.

> I do not think we should block unprefixed animations on filter.

I agree completely.


[1] http://dev.w3.org/fxtf/filters/
[2]
https://code.google.com/p/chromium/issues/list?can=2&q=label%3ACr-Blink-CSS-Filters

--
Erik Dahlstrom, Web Technology Developer, Opera Software
Co-Chair, W3C SVG Working Group
Reply all
Reply to author
Forward
0 new messages