You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
[web-animations-1] Define interaction with page display #2083
Chrome, Firefox, and Safari all appear to make the timeline continue running when loading the page from the bfcache—that is, they don't fiddle with the times to make the animation resume where it left off. As the issue text notes, doing that would be inconsistent with the relationship between animation times and other timing properties like navigationStart.
So we should either just drop this spec section altogether or make it explicit that browsers shouldn't try to restore animation position. I suggest the former. We could probably make a WPT for this too, if necessary.
@astearns Can I request an async resolution to drop section 8 "Interaction with page display" from Web Animations level 1?
@flackr@graouts Any objection to dropping the "Interaction with page display" section since no-one implements what it says (and it would be weird if they did)?
This seems reasonable to me. Do the pages get the finish events for all of the animations that finished in that time since you navigated away? If not, there's a possibility this will result in a broken page state.
Do the pages get the finish events for all of the animations that finished in that time since you navigated away? If not, there's a possibility this will result in a broken page state.
Yes.
The proposal here is simply to remove special case behavior. In the absence of that, for pages restored from the bfcache it page continues as if it were always loaded so in effect there's just a really big gap between frames. Because the timing model is frame-rate independent, we'll catch up on any events that were scheduled in that gap.
The CSSWG will automatically accept this resolution one week from now if no objections are raised here. Anyone can add an emoji to this comment to express support. If you do not support this resolution, please add a new comment.
Proposed Resolution: Drop section 8 "Interaction with page display" from Web Animations level 1.
Activity
birtles commentedon Aug 3, 2026
So the spec text here is wrong. I created a test here: https://birtles.blog/tests/restoring-animations.html
Chrome, Firefox, and Safari all appear to make the timeline continue running when loading the page from the bfcache—that is, they don't fiddle with the times to make the animation resume where it left off. As the issue text notes, doing that would be inconsistent with the relationship between animation times and other timing properties like
navigationStart.So we should either just drop this spec section altogether or make it explicit that browsers shouldn't try to restore animation position. I suggest the former. We could probably make a WPT for this too, if necessary.
@astearns Can I request an async resolution to drop section 8 "Interaction with page display" from Web Animations level 1?
birtles commentedon Aug 15, 2026
@flackr @graouts Any objection to dropping the "Interaction with page display" section since no-one implements what it says (and it would be weird if they did)?
graouts commentedon Aug 17, 2026
No objection from me.
flackr commentedon Aug 17, 2026
This seems reasonable to me. Do the pages get the finish events for all of the animations that finished in that time since you navigated away? If not, there's a possibility this will result in a broken page state.
astearns commentedon Aug 17, 2026
(waiting on an answer to the question from @flackr)
birtles commentedon Aug 19, 2026
Yes.
The proposal here is simply to remove special case behavior. In the absence of that, for pages restored from the bfcache it page continues as if it were always loaded so in effect there's just a really big gap between frames. Because the timing model is frame-rate independent, we'll catch up on any events that were scheduled in that gap.
Test case: https://birtles.blog/tests/animation-finish-events.html
Test in Chrome and Firefox (no access to Safari I'm afraid).
astearns commentedon Aug 24, 2026
The CSSWG will automatically accept this resolution one week from now if no objections are raised here. Anyone can add an emoji to this comment to express support. If you do not support this resolution, please add a new comment.
Proposed Resolution: Drop section 8 "Interaction with page display" from Web Animations level 1.
astearns commentedon Aug 31, 2026
RESOLVED: Drop section 8 "Interaction with page display" from Web Animations level 1.
[web-animations-1] Drop interaction with page display section