[whatwg] Move RequestAnimationFrame steps into HTML?
Cameron and I are editors of the Timing control for script-based
animations spec, more commonly known as the spec for
requestAnimationFrame. This spec has some outstanding feedback from folks
like Anne that needs to be addressed at a basic editorial level that I
haven't had bandwidth to address. It also needs to integrate more tightly
into the HTML's rendering model to get proper timing. I think that adding
the appropriate hooks to both specs will be complicated and I know I don't
have the bandwidth to do this correctly, so I propose that we simply move
this algorithm into HTML itself and ask that the HTML editors (aka Hixie)
take over this part of the spec. If this doesn't happen, I'm afraid that
the spec will languish and it'll be hard to correctly specify the various
things that are supposed to coordinate with each other to produce a
smoothly functioning and consistent system.
Specifically, the spec defines the following WebIDL:
partial interface Window {
long requestAnimationFrame(FrameRequestCallback callback);
void cancelAnimationFrame(long handle);
};
callback FrameRequestCallback = void (DOMHighResTimeStamp time);
and a relatively simple processing model that should integrate tightly into
HTML 8.4.1.2 Processing Model's Update the rendering step:
https://dvcs.w3.org/hg/webperf/raw-file/tip/specs/RequestAnimationFrame/Overview.html#processingmodel
I'm happy to provide any help with technical issues here, but can't promise
to actually edit anything useful.
Does this sound useful to folks?
- James
Re: [whatwg] Move RequestAnimationFrame steps into HTML?
On Wed, 17 Sep 2014, James Robinson wrote: Cameron and I are editors of the Timing control for script-based animations spec, more commonly known as the spec for requestAnimationFrame. This spec has some outstanding feedback from folks like Anne that needs to be addressed at a basic editorial level that I haven't had bandwidth to address. It also needs to integrate more tightly into the HTML's rendering model to get proper timing. I think that adding the appropriate hooks to both specs will be complicated and I know I don't have the bandwidth to do this correctly, so I propose that we simply move this algorithm into HTML itself and ask that the HTML editors (aka Hixie) take over this part of the spec. I'd be happy to do this. I've filed a bug to track it: https://www.w3.org/Bugs/Public/show_bug.cgi?id=26839 This will actually help substantially with resolving this issue also: https://www.w3.org/Bugs/Public/show_bug.cgi?id=26636 Do you have a log of the issues that are outstanding on this spec? -- Ian Hickson U+1047E)\._.,--,'``.fL http://ln.hixie.ch/ U+263A/, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Re: [whatwg] Move RequestAnimationFrame steps into HTML?
On Wed, Sep 17, 2014 at 3:53 PM, Ian Hickson i...@hixie.ch wrote: On Wed, 17 Sep 2014, James Robinson wrote: I'd be happy to do this. I've filed a bug to track it: https://www.w3.org/Bugs/Public/show_bug.I can't speak for the editors of that spec, but I'm pretty sure thcgi?id=26839 https://www.w3.org/Bugs/Public/show_bug.cgi?id=26839 This will actually help substantially with resolving this issue also: https://www.w3.org/Bugs/Public/show_bug.cgi?id=26636 Do you have a log of the issues that are outstanding on this spec? http://lists.w3.org/Archives/Public/public-web-perf/2014Jul/0019.html and http://lists.w3.org/Archives/Public/public-web-perf/2014Jun/0035.html describe two open issues with the spec. The resolutions for both are (I believe) pretty simple and described in those threads. https://www.w3.org/Bugs/Public/show_bug.cgi?id=26440 and https://www.w3.org/Bugs/Public/show_bug.cgi?id=26636 (which you mentioned) are issues that would be a lot easier to resolve if we had the processing model for this system described in one place. The Web Animation spec defines a fuzzy hook into the requestAnimationFrame processing model here: http://w3c.github.io/web-animations/#script-execution-and-live-updates-to-the-model, but that should probably be tightened up as it doesn't specify many details exactly such as the order things happen in different frames or exactly what sorts of things can happen between sampling a WebAnimations animation and running a requestAnimationFrame callback. I believe that the editors (+cc Shane) of that spec would really appreciate having a better model to hook in to in order to precisely define these things. Thanks, - James -- Ian Hickson U+1047E)\._.,--,'``.fL http://ln.hixie.ch/ U+263A/, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Re: [whatwg] Move RequestAnimationFrame steps into HTML?
Web animation is --to my knowledge-- only for the manipulation of the dom's children element and to some extent the dom itself. What of the canvas? Is there any interest in simplifying or and adding on to the requestAnimationFrame? Like making it a constructor or making Daemon object to better PUI--programmer user interface--? E-S4L N-S4L J-S4L On Sep 17, 2014, at 7:02 PM, James Robinson jam...@chromium.org wrote: On Wed, Sep 17, 2014 at 3:53 PM, Ian Hickson i...@hixie.ch wrote: On Wed, 17 Sep 2014, James Robinson wrote: I'd be happy to do this. I've filed a bug to track it: https://www.w3.org/Bugs/Public/show_bug.I can't speak for the editors of that spec, but I'm pretty sure thcgi?id=26839 https://www.w3.org/Bugs/Public/show_bug.cgi?id=26839 This will actually help substantially with resolving this issue also: https://www.w3.org/Bugs/Public/show_bug.cgi?id=26636 Do you have a log of the issues that are outstanding on this spec? http://lists.w3.org/Archives/Public/public-web-perf/2014Jul/0019.html and http://lists.w3.org/Archives/Public/public-web-perf/2014Jun/0035.html describe two open issues with the spec. The resolutions for both are (I believe) pretty simple and described in those threads. https://www.w3.org/Bugs/Public/show_bug.cgi?id=26440 and https://www.w3.org/Bugs/Public/show_bug.cgi?id=26636 (which you mentioned) are issues that would be a lot easier to resolve if we had the processing model for this system described in one place. The Web Animation spec defines a fuzzy hook into the requestAnimationFrame processing model here: http://w3c.github.io/web-animations/#script-execution-and-live-updates-to-the-model, but that should probably be tightened up as it doesn't specify many details exactly such as the order things happen in different frames or exactly what sorts of things can happen between sampling a WebAnimations animation and running a requestAnimationFrame callback. I believe that the editors (+cc Shane) of that spec would really appreciate having a better model to hook in to in order to precisely define these things. Thanks, - James -- Ian Hickson U+1047E)\._.,--,'``.fL http://ln.hixie.ch/ U+263A/, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
