[minutes] 20111012 Web Performance WG Teleconference #51

Meeting Summary:



1.      TPAC 2011 Schedule

a.       Agenda

We have proposed the following agenda and schedule for the Web Performance meeting:

1:00pm - 1:30pm - Operational Discussion.

1:30pm - 2:30pm - Navigation Timing to Recommended.

2:30pm - 2:45pm - Break

2:45pm - 3:45pm - RequestAnimationFrame Open Issues.

3:45pm - 5:00pm - New Performance Investments for this Working Group.



Please share any feedback on this schedule. With an afternoon meeting, we can potentially speak with the WebApp Security WG in the morning about the cross-origin security discussions in the Timing specification space. This can help close the security discussion, if it has not been already closed prior to TPAC.



2.      Page Visibility Feedback

a.       IFrame Visibility

The consensus on the conference call was that the spec currently tries to simplify the spec by defining everything in a visible page in the foreground tab as being visible, and everything else as hidden. Currently, there are other edge cases, like a non-minimized browser window that is fully obscured by other windows appearing as visible though it is hidden. The IFrame and fully obscured window cases may be considered in a future revision in the spec or as non-normative cases. This issue will continue to be discussed on the mailing list.



b.       Visibility on Navigate Away

The consensus on the conference call was that visibility state and the visibilitychange event should not change when the page is being navigated away from. This is mainly because we do not understand a use case where this information would be useful.



Action Items (from last week):
ACTION-51 - Look at NT references to HTML5 and see if those parts of the spec are stable. [on Zhiheng Wang].
ACTION-52 - Look into the failing IE tests in Navigation Timing [on Jatinder Mann].
ACTION-53 - Update resource timing constants from integers to strings [on Jatinder Mann].
ACTION-54 - Update the resource timing abstract [on Jatinder Mann].



Detailed Notes:



Web Perf Teleconference #50 10/12/2011



IRC log: http://www.w3.org/2011/10/12-webperf-irc


Meeting Minutes: http://www.w3.org/2011/10/12-webperf-minutes.html



Attendees

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


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

Philippe Le Hegaret, Jatinder Mann, Arvind Jain, Karen Anderson, Shishir Agrawal



Scribe

Jatinder Mann



Contents

Agenda

1.       Page Visibility Feedback

2.       TPAC 2011 Schedule



--------------------------------------------------------------------------------
IFrame visibility for the Page Visibility spec
Jatinder: Our options are to either mirror IFrame visibility with that of the top level document or giving the more granular IFrame visibility including the applied styling and offscreen location.
... I have found there are two interesting cases where the granular IFrame visibility may be useful: UA extension applying a display:none style to cross-origin iframes or ad being scrolled off screen and can throttle work. The first case does seem like an edge case. The spec today makes a simplifying assumption that anything in the visible page is visible.
... The goal is to reduce background interference. This goal is relatively well met with the current spec, where pages in background tabs are throttled.
Arvind: There are some cases that we haven't covered, like fully obscured window. Likewise, this can be another case that can be covered in the next cut of the spec.
Jatinder: If we do support this in the spec, I feel it should not be a normative requirement.
Arvind: I will respond to the mailing list.
... The next topic was whether the visibility event should be fired and state changed for the navigate away case.
Jatinder: I think we should make an assumption that we should either fire the event and change the state or not fire the event or change the state.
... Before the unload event, we are still visible, so we shouldn't report as hidden then. After the unload event, the page should be unloaded, so there shouldn't be any action that can be taken by knowing the page is hidden. We should follow up on the mailing list to understand why this behavior would be desired.
TPAC Schedule
Jatinder: As for TPAC, the proposed timeframe will be to meet between 1-5pm Tuesday afternoon. This gives us the morning to potentially chat with the WebApp Security WG. If we need more time, we can move the meeting to the morning.