W3C home > Mailing lists > Public > public-web-perf@w3.org > October 2013

[minutes] Web Performance WG Teleconference #120 2013-10-23

From: Jatinder Mann <jmann@microsoft.com>
Date: Wed, 23 Oct 2013 23:42:45 +0000
To: "'public-web-perf@w3.org'" <public-web-perf@w3.org>
Message-ID: <c7c3b2769b734b3ba0db8ef7a207ea4f@BLUPR03MB065.namprd03.prod.outlook.com>
Meeting Summary:



1.     Specs moving to PR and CR

Philippe has started the process of moving User Timing and Performance Timeline to PR and rAF to CR. There will be a transition request next week.



2.     Performance implements EventTarget

We had initially discussed updating Navigation Timing L1 errata to change the Performance object to implement EventTarget. Though this doesn't have a compatibility impact, it will have a conformance impact and require the spec to go back to Last Call, which isn't desirable for a Recommendation spec. Instead we think it may be better to re-define the Performance object in the Navigation Timing L2 and deprecate the Navigation Timing L1 once the L2 spec has reached Recommendation. Seeing that the L2 spec already re-defines all of the L1 timing attributes in the PerformanceEntry object, this seems like a natural progression. I will follow up on the mailing list on this proposal.



3.     Specs moving to FPWD

As Resource Priorities and Beacon are relatively stable compared to the other new specs, we are moving to publish those specs as FPWD. This will also encourage more people to review those specs and provide feedback.



4.     Progress on new specs prior to TPAC

We haven't made much progress on some of our older specs, including Navigation Error Logging, Resource Error Logging, HAR, and JavaScript Injection. If all editors could clean up their specs prior to TPAC, we can have a more concise and targeted discussion on the remaining open issues on those specs.



5.     Discussions topics for HTML WG F2F at TPAC

We currently only have two topics of discussion with the HTML WG at TPAC: getting link rel=prerender included in the HTML5 spec and showing the Resource Priorities spec to the group. If you have any additional topics you want to consider discussing, please respond to this thread.



6.     Moving test cases to github

To match other working groups, we are planning on moving our test cases to the github repository. Philippe has pointed to some documentation in this mail thread,  http://lists.w3.org/Archives/Public/public-web-perf/2013Oct/0052.html. For now, we will keep the specs still in the existing repository, but can consider moving that over to github in the future as well.




W3C Web Performance WG Teleconference #120 2013-10-23





IRC log: http://www.w3.org/2013/10/23-webperf-irc



Meeting Minutes: http://www.w3.org/2013/10/23-webperf-minutes.html



Attendees

Philippe Le Hegaret, Jatinder Mann, Ganesh Rao


Scribe

Jatinder Mann



Agenda

1.     Discuss existing spec progress

2.     Discuss new spec progress

3.     Move to github repository

4.     TPAC logistics

--------------------------------------------------------------------------------



Minutes:
Existing specs

<plh> https://lists.w3.org/Archives/Member/chairs/2013OctDec/0056.html

<plh> https://lists.w3.org/Archives/Member/chairs/2013OctDec/0072.html

<plh> so, RAF, User Timing, and Performance Timeline should be moving forward

plh: We should have a issues list for our specs prior to TPAC.

<plh> http://www.w3.org/2013/08/cr-issues-performance-timeline.html

<plh> http://lists.w3.org/Archives/Public/public-web-perf/2013Aug/0106.html

<plh> performance.getEntries() == performance.getEntries() returns false

performance.getEntries()[0] == performance.getEntries()[0]

performance.getEntries() == performance.getEntries()

Jatinder: IE11 gives me false on both tests

performance.getEntries()[0] == performance.getEntries()[0]

Jatinder: Returns true in Chrome

performance.getEntries() == performance.getEntries() is false for the entire list

Jatinder: Seems like we should test both cases.
... I made updates to Performance Timeline based on feedback from Philippe, namely note to wiki page with latest known entryTypes and a new Vendor Extensions section, http://lists.w3.org/Archives/Public/public-web-perf/2013Oct/0053.html

Plh: Let's plan to do the transition request sometime next week.

Jatinder: Navigation Timing L1 Errata: Update Performance object to implement EventTarget, http://lists.w3.org/Archives/Public/public-web-perf/2013Oct/0046.html

plh: NTL1 change would have no compatibility hit, but will have a conformance hit. We would have to go to Last Call again and the whole process again.
... Alternatively, we can just define this change in NT L2 and deprecate NT L1. Resource Timing would be tied to NT L2 though.

Jatinder: Seeing that we have defined all the attributes in the NT L2 spec, I am okay with deprecating NT L1 in favor of NT L2. IE11 already supports NT L2. I'll email the list.
... High Resolution Time L1 Errata: Move navigationStart reference to note, http://lists.w3.org/Archives/Public/public-web-perf/2013Oct/0049.html

Plh: I'll make the update to the Errata.

Jatinder: Page Visibility L1 had a spelling mistake on visibilitychange, which was fixed in the latest draft. Do we need to update errata or can this change just be pushed out?

Plh: Let me follow and see what I can do about this.

ACTION Plh to fix visibilitychange typo in Page Visibility L1 spec.

<trackbot> Created ACTION-109 - Fix visibilitychange typo in page visibility l1 spec. [on Philippe Le Hégaret - due 2013-10-30].
Progress on new specs

Jatinder: I have been making changes based on feedback on Resource Priorities. I don't believe we have any significant feedback changes remaining, but there may be some. Can we at least publish this spec as a FPWD?

<plh> Resolved: Publish resource priorities as FPWD

Jatinder: I haven't looked at the Beacon spec in a while, but I didn't think there was a lot of feedback on that spec. I may need to strengthen the processing model. Can we publish this spec as FPWD as well?

<plh> Resolved: Publish beacon API as FPWD

Jatinder: We had quite a bit discussion on Navigation Error Logging and Resource Error Logging, and I think there are changes that remain to be made. I'll ping Arvind and I'll try to help make changes.
... I believe for Prerender we decided that we should try to get the link = prerender text added to the HTML5 spec. Is there a bug open on that?

Plh: We updated the wiki page pointed on the HTML5 spec. We will want to ask HTML WG to put this into the spec.
... Why topics did you want to talk to with HTML WG?

Jatinder: Mainly the prerender spec and we can introduce the Resource Priorities spec.
... I believe we still have plans to document HAR as a note. We should try to make progress on that
... Based on the last discussions, I'm not sure if we are yet committed to JavaScript Injection. We should close on whether we want to pursue it or not.
... Since TPAC is only two weeks away, why don't we talk about this spec at TPAC.

Plh: Mark will be at TPAC.

Jatinder: We'll be sure to talk about this with Mark and Alois in the room.
Moving repository to github

Jatinder: Philippe shared documentation here: http://lists.w3.org/Archives/Public/public-web-perf/2013Oct/0052.html
TPAC 2013

plh: I will add the agenda on a wiki page
Received on Wednesday, 23 October 2013 23:43:15 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:04:37 UTC