16:44:09 <RRSAgent> RRSAgent has joined #webperf
16:44:09 <RRSAgent> logging to http://www.w3.org/2016/10/26-webperf-irc
16:44:30 <xiaoqian> RRSAgent, make log public
16:54:23 <igrigorik> hello!
16:56:26 <n8s> n8s has joined #webperf
16:58:03 <n8s> n8s has joined #webperf
16:58:53 <igrigorik> Hangout for today's call: https://hangouts.google.com/hangouts/_/chromium.org/webperf-wg
17:01:27 <n8s> anyone else having issues joining the call? I just see a message that says "requesting to join the video call..."
17:01:36 <yoav> me too
17:01:42 <xiaoqian> me too
17:02:05 <yoav> tried switching browsers, btu that didn't help
17:02:57 <igrigorik> hmm
17:02:57 <igrigorik> sec
17:03:27 <toddreifsteck> toddreifsteck has joined #webperf
17:04:51 <igrigorik> yoav, try now
17:04:56 <plh> regrets for today. conflicting meeting :(
17:06:16 <igrigorik> Agenda: https://lists.w3.org/Archives/Public/public-web-perf/2016Oct/0004.html
17:06:45 <igrigorik> ben: we want to know how long between user action (e.g. click) to response on the screen (e.g. show a spinner)
17:07:27 <igrigorik> ... 1) hardware input to event running
17:07:39 <igrigorik> ... 2) event running to processed in app code
17:07:45 <igrigorik> 3) processed in app code to visible
17:08:19 <igrigorik> ... today we guess by setTimeout + raf as an estimate on how long it takes for something to get painted
17:08:42 <igrigorik> ... all three are equally important to surface
17:09:11 <igrigorik> ... there are cases where we have significant delays; cases where we force costly rendering operations that delay paint
17:09:32 <igrigorik> ... if we're trying to create an API that measures input, we should try to capture all three phases
17:09:38 <igrigorik> ... total time should be "duration of the event"
17:09:56 <igrigorik> ... for paint, timing up to the end of layout may be acceptable -- close enough approximation
17:11:11 <igrigorik> ... shubhie/tim gave pushback on the 3rd, but there are many opportunities for dev's to screw this up (it's not just up to the browser)
17:11:53 <igrigorik> yoav: not clear on what the 3rd part is, what do you want to measure? how does the browser correlate between handling event and paint
17:12:42 <igrigorik> ben: the 100ms response cycle places hard constraints on what you can do; you can't do network requests.. first phase is to show a response locally (e.g. spinner)
17:13:25 <igrigorik> ... a good input API should allow you to distinguish between events that do and do not have a response (e.g. mousemove may not have direct response)
17:16:12 <igrigorik> ig: above definition is the ~optimistic first opportunity to provide visual response; it's possible for the code not _not_ do this, but that's their own "fault"
17:17:10 <n8s> igrigorik, do you see vladan.bugzilla trying to join the call? He's also a coworker and he said he's seeing the "waiting to join the call" message
17:17:27 <igrigorik> n8s: apologies, just approved
17:17:45 <n8s> igrigorik, awesome, thanks!
17:19:52 <igrigorik> yoav: we also need to think about the thresholds (e.g. 50ms may be too agressive)
17:20:15 <igrigorik> ben: breaking out sub-parts is also a good idea, but we should provide total duration also
17:20:23 <xiaoqian> present+ igrigorik, Ben_Maurer, Wesley, vladan, yoav, toddreifsteck, n8s, xiaoqian
17:21:04 <igrigorik> ben: there needs to be a way for an event to tell the browser if this event should be considered for this API
17:21:20 <igrigorik> ... today our heuristic is dirtying the DOM
17:23:15 <igrigorik> ig: how about debounce/rate-throttlign pattern
17:24:06 <igrigorik> ben: maybe we could treat ~setTimeout as an input event; maybe treat the next paint event
17:25:15 <igrigorik> yoav:  you could polyfill parts of this?
17:25:53 <igrigorik> ben: yes, but it's tricky and requires careful instrumentation -- e.g. all addeventlisteners, custom fields on event, etc. In practice, tricky to deploy and enforce
17:26:24 <rniwa> rniwa has joined #webperf
17:26:34 <igrigorik> todd: if you're office.com and have 40 teams.. having a standard API would be very helpful to solve the coordination problem
17:27:58 <igrigorik> AIs: Ben to draft summary
17:28:49 <igrigorik> ----
17:29:00 <igrigorik> Long tasks / Long input: same?
17:29:21 <igrigorik> Yoav: I think I'm convinced where LT will be triggered but LI won't and vice versa
17:31:08 <igrigorik> todd: LT is essential for identifying what the likely impact will be; Input is actually rare..
17:31:17 <igrigorik> ... in real world you migth only enable LT for subset of users in production
17:31:37 <NicJansma> NicJansma has joined #webperf
17:32:15 <igrigorik> ig: we also discussed blocked flag on LT?
17:32:28 <cristi> cristi has joined #webperf
17:32:34 <NicJansma> Is the call still happening on Hangouts?  (my screen is stuck on Requesting to join the video call...)
17:33:06 <igrigorik> n8s: input api: here's the thing that mattered and it was blocked;  LT is good for identifying potential problems, you'll probably sample
17:33:18 <yoav> NicJansma: Hi! :)
17:33:30 <igrigorik> NicJansma: apologies, had to approve.. still getting a hang of it
17:33:36 <igrigorik> AI: continue discussion on github
17:34:13 <igrigorik> ---
17:34:15 <igrigorik> Beacon: https://github.com/w3c/beacon/issues/29#issuecomment-255882261
17:35:10 <NicJansma> No worries!
17:37:36 <igrigorik> toddreifsteck: current Chrome quota implementation is interesting... passes current spec language, but probably a little strict
17:37:51 <igrigorik> ... an SPA can have a lifetime of days
17:42:11 <igrigorik> yoav: perhaps the quota should only be enforced once onunload starts? that would solve the SPA case
17:44:14 <igrigorik> NicJansma: we're implementing sendBeacon (Soasta), we might have more telemetry
17:45:50 <igrigorik> https://www.chromestatus.com/metrics/feature/timeline/popularity/495
17:46:26 <igrigorik> AI: take this discussion to Github
17:47:55 <igrigorik> https://github.com/w3c/web-platform-tests/pull/4024#discussion_r84792665
17:48:18 <igrigorik> toddreifsteck: probably a low priority..
17:48:52 <igrigorik> NicJansma: agree, if we get an error we'll probably just fallback strategy
17:49:29 <igrigorik> toddreifsteck: I'd like to followup with Travis
17:49:54 <igrigorik> ig: let's pull in Anne  as well
17:50:11 <igrigorik> AI: let's take this for review with Travis and Anne
17:50:45 <igrigorik> https://github.com/w3c/beacon/issues/37
17:51:24 <igrigorik> ig: I dropped Worker support from spec
17:51:28 <igrigorik> toddreifsteck: we're OK with that
17:51:51 <igrigorik> bitly.com/w3c-webperf-status
17:52:10 <toddreifsteck> Ilya, if you feed me links to URLs, I can get them to the owner of the DOM page if it isn't properly filtering for any of them.
17:52:41 <igrigorik> https://lists.w3.org/Archives/Public/public-web-perf/2016Oct/0004.html
17:53:39 <igrigorik> ig: any objections to publishing L3's for Performance Timeline L3, Resource Timing L3, Navigation Timing L3, User Timing L3
17:53:52 <igrigorik> .. no objections. we'll publish L3's.
17:54:58 <igrigorik> .. i'll create "v2" branches for each and gh-pages  will track L3's
17:55:21 <igrigorik> transitions to L2
17:55:22 <igrigorik> https://github.com/w3c/user-timing/issues/19
17:56:24 <igrigorik> ig: spec is ready, I suggest we push it to CR and work on tests in parallel
17:56:27 <igrigorik> toddreifsteck: sgtm
17:56:33 <igrigorik> ... no objections from others
17:56:42 <igrigorik> https://github.com/w3c/page-visibility/issues/25
17:57:09 <plh> +1 re L2 branch and L3 in gh-pages
17:57:52 <igrigorik> igrigorik: spec ready to go, suggest we push it to CR
17:58:13 <igrigorik> ... sgtm from Todd and rest
17:59:18 <igrigorik> next call, Nov 16th
17:59:56 <xiaoqian> present+ NicJansma, nolanlawson
18:00:00 <xiaoqian> RRSAgent, make minutes
18:00:00 <RRSAgent> I have made the request to generate http://www.w3.org/2016/10/26-webperf-minutes.html xiaoqian
18:01:41 <igrigorik> plh: yeah. email?
18:01:56 <plh> I was thinking a quick hangout
18:02:08 <plh> if you have time
18:02:24 <plh> but email might be best for you
18:02:42 <igrigorik> in meetings for next few hours.. let's do email?
18:02:46 <plh> oki
18:03:10 <plh> I'll send you an email with what I think my todo list is
18:03:21 <plh> and you'll correct it as needed
18:03:31 <igrigorik> sgtm, thanks!
18:32:44 <n8s> n8s has joined #webperf
18:50:28 <yoav> yoav has joined #webperf
18:57:03 <rniwa_> rniwa_ has joined #webperf
20:03:35 <rniwa> rniwa has joined #webperf
20:20:39 <Zakim> Zakim has left #webperf
20:29:39 <yoav> yoav has joined #webperf
21:29:13 <bkardell_> bkardell_ has joined #WebPerf
22:39:51 <n8s> n8s has joined #webperf
22:42:56 <n8s> n8s has joined #webperf
23:03:26 <cristi> cristi has joined #webperf