W3C home > Mailing lists > Public > public-web-perf@w3.org > March 2017

Minutes [was: Re: [3/1/17] webperf group call @ 10AM PST]

From: Xiaoqian Wu <xiaoqian@w3.org>
Date: Fri, 3 Mar 2017 00:16:13 +0800
To: public-web-perf <public-web-perf@w3.org>
Message-Id: <B7AF59B0-0761-444E-855C-C33DC6F386CA@w3.org>

The minutes of this week's webperf group call are available:
https://www.w3.org/2017/03/01-webperf-minutes.html <https://www.w3.org/2017/03/01-webperf-minutes.html>

Next call will be 3/22/2017.

Plain text version:

Using User Timing at scale

Dan: working on React.. CDT integration was great
… there is no way for a library to tell if user is recording or not, we're afraid of wasting memory
… we're calling clearmarks/measures all the time, but it means we can't harvest it later
… our main target is CDT
… i.e. in dev mode
an example -> https://‌facebook.github.io/‌react/‌docs/‌optimizing-performance.html#profiling-components-with-chrome-timeline
a) sounds like there is a need for API in devtools for in "record mode"
b) you should log some data to performance timeline.. perhaps with lower fidelity
Nic: we monitor React today, we look at routechange events
Dan: we expose UT for every component in dev mode
Beacon: Enforce payload limits via Fetch API

Enforce payload limits via Fetch API: https://‌github.com/‌w3c/‌beacon/‌pull/‌39
Performance Timeline: issues#66

timeline: https://‌github.com/‌w3c/‌performance-timeline/‌issues/‌66
Q: is this L2 or L3 feature?
we'll revisit next time.
Navigation Timing: Set `name` to document address

Set `name` to document address: https://‌github.com/‌w3c/‌navigation-timing/‌pull/‌61
shipping in M57 in Chrome, stable ~mid march
we'll pause landing it until we have confidence in it
Administrative stuff

next: charter update
Todd: we need/want charter update from MSFT end
… I'll review and merge, first pass looks good
we'll followup with plh@ on next steps
xiaoqian: we probably need a test for prerender.
AI: let's start a new issue on the repo
we'll wait for PR until we resolve ^
Todd: there are tests in the WPT tree that require user input
L2 CR for UT: https://‌github.com/‌w3c/‌user-timing/‌issues/‌19
Todd: I'll try to get the updated tests.
We'll try to triage UT2 issues before next call
next call: March 22nd


> On 1 Mar 2017, at 7:59 AM, Ilya Grigorik <igrigorik@google.com> wrote:
> Hangout: https://hangouts.google.com/hangouts/_/chromium.org/webperf-wg <https://hangouts.google.com/hangouts/_/chromium.org/webperf-wg>
> --- Agenda:
> Using User Timing at scale -- Dan Abramov
> Beacon
> Enforce payload limits via Fetch API: https://github.com/w3c/beacon/pull/39 <https://github.com/w3c/beacon/pull/39>
> Updates to webplatform tests
> Performance Timeline
> Feature detection note: https://github.com/w3c/performance-timeline/pull/70 <https://github.com/w3c/performance-timeline/pull/70>
> Should we tackle takeRecords as part of L2? See https://github.com/w3c/performance-timeline/issues/66 <https://github.com/w3c/performance-timeline/issues/66>
> Navigation Timing
> Set `name` to document address: https://github.com/w3c/navigation-timing/pull/61 <https://github.com/w3c/navigation-timing/pull/61>
> Administrative stuff
> Charter update: https://github.com/w3c/charter-webperf/pull/34 <https://github.com/w3c/charter-webperf/pull/34>
> Adopt Long Tasks, Server Timing?
> Move Page Visibility L2 to PR?
> See https://github.com/w3c/page-visibility/issues/25#issuecomment-262263862 <https://github.com/w3c/page-visibility/issues/25#issuecomment-262263862>
> Move HR-Time L2 to PR? 
> One missing test: https://github.com/w3c/hr-time/issues <https://github.com/w3c/hr-time/issues>
> Missing TAO test is blocking RT L1..
> https://github.com/w3c/resource-timing/issues/92 <https://github.com/w3c/resource-timing/issues/92>
Received on Thursday, 2 March 2017 16:16:27 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 2 March 2017 16:16:27 UTC