Add UMA for ServiceWorker FirstMeaningfulPaint

Add UMA for ServiceWorker FirstMeaningfulPaint

BUG=727599
Review-Url: https://codereview.chromium.org/2908323002
Cr-Commit-Position: refs/heads/master@{#475760}
Committed: https://chromium.googlesource.com/chromium/src/+/a55da7b3ac651bebd147bb11f734546ac533ebcb

[horo, 2017-05-30 06:51:01]

The CQ bit was checked by horo@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-05-30 06:51:14]

Dry run: CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-05-30 06:53:08]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-05-30 06:53:09]

Dry run: Try jobs failed on following builders:
  ios-device on master.tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/ios-device/builds...)
  ios-device-xcode-clang on master.tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/ios-device-xcode-...)
  mac_chromium_compile_dbg_ng on master.tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_comp...)
  mac_chromium_rel_ng on master.tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[horo, 2017-05-30 07:08:15]

Description was changed from

==========
Add UMA for ServiceWorker FirstMeaningfulPaint

BUG=
==========

to

==========
Add UMA for ServiceWorker FirstMeaningfulPaint

BUG=727544
==========

[horo, 2017-05-30 07:17:36]

Patchset #1 (id:1) has been deleted

[horo, 2017-05-30 07:17:41]

The CQ bit was checked by horo@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-05-30 07:17:56]

Dry run: CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[horo, 2017-05-30 07:22:05]

horo@chromium.org changed reviewers:
+ falken@chromium.org

[horo, 2017-05-30 07:22:06]

falken@
Could you please review this?

[horo, 2017-05-30 09:15:40]

Description was changed from

==========
Add UMA for ServiceWorker FirstMeaningfulPaint

BUG=727544
==========

to

==========
Add UMA for ServiceWorker FirstMeaningfulPaint

BUG=727599
==========

[falken, 2017-05-30 09:22:05]

lgtm

[horo, 2017-05-30 09:30:04]

horo@chromium.org changed reviewers:
+ mpearson@chromium.org

[horo, 2017-05-30 09:30:04]

mpearson@
Could you please review histograms.xml?

[kinuko, 2017-05-30 09:30:30]

lgtm

[kinuko, 2017-05-30 09:30:31]

lgtm

[commit-bot: I haz the power, 2017-05-30 10:14:44]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-05-30 10:14:45]

Dry run: This issue passed the CQ dry run.

[tdresser, 2017-05-30 12:06:47]

tdresser@chromium.org changed reviewers:
+ tdresser@chromium.org

[tdresser, 2017-05-30 12:06:48]

A few non-blocking drive-by thoughts:

The kinds of analysis I'd like to do on this data would be much easier if we
also recorded a separate histogram for cases without a service worker present.
Could we add parallel non-serviceworker variants?

Will UKM enable us to eventually get rid of the site specific hacks here?

[Mark P, 2017-05-30 18:34:02]

histograms.xml lgtm
Note that I did not the review the actual metrics *code*.  I assume your primary
reviewers did that.

--mark

[horo, 2017-05-31 02:53:18]

On 2017/05/30 12:06:48, tdresser wrote:
> A few non-blocking drive-by thoughts:
> 
> The kinds of analysis I'd like to do on this data would be much easier if we
> also recorded a separate histogram for cases without a service worker present.
> Could we add parallel non-serviceworker variants?
> 
> Will UKM enable us to eventually get rid of the site specific hacks here?

Thank you for the comment.

We already have NavigationToFirstMeaningfulPaint UMA for all pages.
https://chromium.googlesource.com/chromium/src/+blame/ebe52af6c0285feef6f776d...
And also we have NavigationToFirstMeaningfulPaint UKM.
https://chromium.googlesource.com/chromium/src/+blame/ebe52af6c0285feef6f776d...

But I think we should have these service worker metrics to improve the service
worker performance.

[horo, 2017-05-31 02:53:23]

The CQ bit was checked by horo@chromium.org

[commit-bot: I haz the power, 2017-05-31 02:53:41]

CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-05-31 02:58:53]

CQ is committing da patch.

Bot data: {"patchset_id": 20001, "attempt_start_ts": 1496199203622840,
"parent_rev": "910c7d9dcd9571744626dcd19f3ec3dfad2daf5d", "commit_rev":
"a55da7b3ac651bebd147bb11f734546ac533ebcb"}

[commit-bot: I haz the power, 2017-05-31 02:59:03]

Description was changed from

==========
Add UMA for ServiceWorker FirstMeaningfulPaint

BUG=727599
==========

to

==========
Add UMA for ServiceWorker FirstMeaningfulPaint

BUG=727599

Review-Url: https://codereview.chromium.org/2908323002
Cr-Commit-Position: refs/heads/master@{#475760}
Committed:
https://chromium.googlesource.com/chromium/src/+/a55da7b3ac651bebd147bb11f734...
==========

[commit-bot: I haz the power, 2017-05-31 02:59:05]

Committed patchset #1 (id:20001) as
https://chromium.googlesource.com/chromium/src/+/a55da7b3ac651bebd147bb11f734...

[tdresser, 2017-05-31 14:21:16]

On 2017/05/31 02:53:18, horo wrote:
> On 2017/05/30 12:06:48, tdresser wrote:
> > A few non-blocking drive-by thoughts:
> > 
> > The kinds of analysis I'd like to do on this data would be much easier if we
> > also recorded a separate histogram for cases without a service worker
present.
> > Could we add parallel non-serviceworker variants?
> > 
> > Will UKM enable us to eventually get rid of the site specific hacks here?
> 
> Thank you for the comment.
> 
> We already have NavigationToFirstMeaningfulPaint UMA for all pages.

Yeah, it's just easier to compare two distributions if you record both
distributions, instead of recording one distribution, and the aggregate of both
distributions. The UMA recommendations state to avoid derived metrics if
possible, and without deriving the distribution for pages without service
workers, you can't make comparisons between the two distributions.

>
https://chromium.googlesource.com/chromium/src/+blame/ebe52af6c0285feef6f776d...
> And also we have NavigationToFirstMeaningfulPaint UKM.

Should we have a bug to get rid of the inbox specific UMA then?

>
https://chromium.googlesource.com/chromium/src/+blame/ebe52af6c0285feef6f776d...
> 
> But I think we should have these service worker metrics to improve the service
> worker performance.

[horo, 2017-06-01 05:40:16]

On 2017/05/31 14:21:16, tdresser wrote:
> On 2017/05/31 02:53:18, horo wrote:
> > On 2017/05/30 12:06:48, tdresser wrote:
> > > A few non-blocking drive-by thoughts:
> > > 
> > > The kinds of analysis I'd like to do on this data would be much easier if
we
> > > also recorded a separate histogram for cases without a service worker
> present.
> > > Could we add parallel non-serviceworker variants?
> > > 
> > > Will UKM enable us to eventually get rid of the site specific hacks here?
> > 
> > Thank you for the comment.
> > 
> > We already have NavigationToFirstMeaningfulPaint UMA for all pages.
> 
> Yeah, it's just easier to compare two distributions if you record both
> distributions, instead of recording one distribution, and the aggregate of
both
> distributions. The UMA recommendations state to avoid derived metrics if
> possible, and without deriving the distribution for pages without service
> workers, you can't make comparisons between the two distributions.

Created a cl: https://codereview.chromium.org/2916033002

> >
>
https://chromium.googlesource.com/chromium/src/+blame/ebe52af6c0285feef6f776d...
> > And also we have NavigationToFirstMeaningfulPaint UKM.
> 
> Should we have a bug to get rid of the inbox specific UMA then?

Filed: https://crbug.com/728484