[minutes] 2012-04-25 Web Performance WG Teleconference #70

Meeting Summary:



1.      Navigation Timing

a.       test_timing_attributes_order.htm

Karen and Boris are working on fixing the test_timing_attributes_order.htm test case. They appear to have identified the fix, but don't currently have it implemented.



b.       test_document_readiness_exist.htm

This test fails in IE10, however, passes in IE9. Karen will investigate and determine if there is an IE10 bug.



c.        Navigation Timing 2

The working group decided to take Section 4.4 of Performance Timeline out of the Performance Timeline spec and into a Navigation Timing 2 speak. Further, the additions that Jatinder suggested to include PerformanceNavigationTiming will be added to Navigation Timing 2. Jatinder has ACTION 102: Move the data from Section 4.4 in PerformanceTimeline and add the PerformanceNavigationTiming entry into Navigation Timing 2 spec.



2.      Performance Timeline

a.       Remove Section 4.4

The working group has decided to move Section 4.4 to the new Navigation Timing 2 spec.



b.       Performance Timeline to CR

As there has not been any feedback or open issues on the remaining Performance Timeline sections for quite some time now, the working group would like to move this spec to CR. Philippe will review the mailing list and ensure nothing is missed.



3.      Resource Timing

a.       Resource Timing to CR

As there are no open issues or feedback on this spec, the working group would like to move this spec to CR. Philippe will review the mailing list and ensure nothing is missed.



b.       Reference Implementations

The working group would like to have two reference implementations before removing prefixes.



4.      User Timing

a.       User Timing to CR

As there are no open issues or feedback on this spec, the working group would like to move this spec to CR. Philippe will review the mailing list and ensure nothing is missed.



5.      High Resolution Time

a.       Non-normative notes sections updated.

Per Simon's feedback, the non-normative notes text has been updated.



b.       High Resolution Time to CR

As there are no open issues or feedback on this spec, the working group would like to move this spec to CR. Philippe will review the mailing list and ensure nothing is missed.



6.      requestAnimationFrame

a.       window.animationStartTime

The working group brought up the topic of synchronizing multiple animations and found that animationStartTime to be an interesting solution to that problem. Further, the working group was interested in seeing if animationStartTime could be used to synchronize animations across same-origin sub-documents.



7.      Page Visibility

a.       visibilitychange event processing model

Jatinder is prepping a fix to update the visibilitychange processing model to be more clear.



b.       Tracking Unload/Suspending Scenarios in ISSUE-8

The specification contains draft text on including the unloading and suspending scenario in the Page Visibility spec, as tracked by ISSUE-8: http://www.w3.org/2010/webperf/track/issues/8. Once this issue has been closed, we can decide to remove or keep the text supporting the unloading/suspending scenarios.


8.      F2F at Velocity June 2012

The WG agreed that it does not make sense to meet at Velocity, as the WG proceedings will not be open to the public and the conference will already cover Web Perf specification topics (e.g., sessions on requestAnimationFrame and Navigation Timing). The working group agreed to complete the working in shipping the current specifications before meeting to discuss the next set.


Detailed Notes:



Web Perf Teleconference #70 4/25/2012



IRC log: http://www.w3.org/2012/04/25-webperf-irc


Meeting Minutes: http://www.w3.org/2012/04/25-webperf-minutes.html



Attendees

Present for Navigation Timing, Resource Timing and User Timing (4-5PM EST/1-2PM PST)

Jatinder Mann, Philippe Le Hegaret, Tony Gentilcore, Arvind Jain, James Simonsen


Present for Page Visibility, Efficient Script Yielding, Display Paint Notifications (4-5PM EST/2-3PM PST)

Meeting cancelled.



Scribe

Jatinder Mann



Contents

Agenda

1.     Review feedback on all specifications

--------------------------------------------------------------------------------
Navigation Timing Updates

Jatinder: Test_timing_attributes_order.htm test case needs to be fixed. Karen has prepped a fix, however, it is incomplete. I will try to fix it before handing it off to Boris.
... The test_document_readiness_exist.htm is failing in IE for the reasons Karens mentioned in this mail thread: http://lists.w3.org/Archives/Public/public-web-perf/2012Apr/0062.html. Considering we added this test case here to ensure the HTML5 dependencies were stable, should we update the test case or find other ways to test the stability of the dependencies?

<plh> http://www.w3.org/2012/04/navigation_timing_cr_results.html

plh: Test passes in IE9 and fails in IE10.

Jatinder: That shouldn't be the case. I will look into this test case.
PerformanceTimeline

Arvind: There was a discussion on adding a PerformanceNavigationTiming. Should we add that in PerformanceTimeline or NavigationTiming2?

plh: We can add to Navigation Timing 2, that seems like the logical approach.

Arvind: I was hoping to add additional attributes to Navigation Timing 2.

Jatinder: I can remove data from Section 4.4 in PerformanceTimeline into this new spec and make the PerformanceNavigationTiming entry.

Action Jatinder Move the data from Section 4.4 in PerformanceTimeline and add the PerformanceNavigationTiming entry.into Navigation Timing 2 spec.

<trackbot> Created ACTION-101 - Move the data from Section 4.4 in PerformanceTimeline and add the PerformanceNavigationTiming entry.into Navigation Timing 2 spec. [on Jatinder Mann - due 2012-05-02].
Resource Timing

<plh> ACTION: plh to add wording on getting implementations before removing prefixes [recorded in http://www.w3.org/2012/04/25-webperf-minutes.html#action01]

<trackbot> Created ACTION-102 - Add wording on getting implementations before removing prefixes [on Philippe Le Hégaret - due 2012-05-02].

Jatinder: We have no action items or remaining feedback for this spec.

plh: Should we move to CR for this spec?

Jatinder: The spec seems stable enough that we should move it forward.

Tony: Let's make sure that we have some text that says we do not remove prefixes until we have two reference resources.
User Timing

Jatinder: This spec has no feedback or action items. Shall we move this spec to CR as well?

plh: We can include this spec with the others.
High Resolution Time

Jatinder: I have prepped the changes requested by Simon from Opera. They all involve changes in the notes.
... Otherwise, there are no remaining feeback or action items.

James: I recall there was a discussion of the origin of the high resolution time timebase of being navigationStart for each document. This would be unique for all subdocuments. For synchronizations reasons, should we make same origin subdocuments share the navigationStart of the parent?

Jatinder: Actually, if a same origin subdocument wanted that information, it can already access the parent's navigationStart by using parent.performance.timing.navigationStart, for example.
... I think for the timing specs it makes sense to have each document keep its unique origin, as the timeline looks better. Recall, that the parent only has the subdocument on its timeline, not the resources for the subdocument. However, I think you bring up a valid concern for animation synchronization. I had sent a previous mail about window.animationStartTime as a way to synchronize multiple animations.

James: Yes, the same origin subdocument can access its parent's navigationStart; that solves most concerns. I think the animationStartTime is interesting; I will talk to JamesR about this.

Jatinder: Considering we have no other actions, can we move this spec to CR along with the other specs?

James: I'm fine with that.

Arvind: That's okay with me.

Tony: We are working on an implementation and this one seems simple.

Plh: Okay.
Page Visibility

Jatinder: Boris and James Graham had requested that I clarify the processing model for the visibilitychange event. I have prepped a change for that and will push it out soon.
... We still have that one remaining open issue on the unload/pagehide/pageshow events. Has Google decided on their point of view?

Arvind: We are currently in discussions on this topic and will have reply with our stance soon.
Web Performance F2F Meeting

Jatinder: There was a dicussion on the mailing list on whether or not to meet at the Velocity conference in June of this year.

Arvind: I agree that Jason's point is valid; the conference attendees can't attend the working group meetings and the conference will have these subjects covered. I don't think we should meet then.

Jatinder: I also agree that we should complete these specs first, as they are almost done, and then meet F2F to discuss the next set of work we want to do.