Intent to Implement and Ship: notificationclose event

[Nicolas Satragno]

Contact emails

nsat...@chromium.org

pe...@chromium.org

Spec

https://notifications.spec.whatwg.org/#close-steps

Summary

Fire a notificationclose event on a service worker when the user closes a persistent notification.

Motivation

Letting web developers know when a notification has been dismissed by the user allows them to close it on other devices, or stop the service the notification was being displayed for. This feature also allows collecting metrics.

Interoperability risk

Chrome will be the first browser to implement this feature. A firefox developer has expressed interest on exposing the event to Web developers as well [1].

Compatibility risk

This change only adds a new event, so it won’t break any existing websites.

Ongoing technical constraints

None.

Will this feature be supported on all six Blink platforms (Windows, Mac, Linux, Chrome OS, Android, and Android WebView)?

The feature will be supported in all platforms where notifications are supported (currently Windows, Mac, Linux, ChromeOS and Android).

OWP launch tracking bug

http://crbug.com/581299

Link to entry on the feature dashboard

https://www.chromestatus.com/features/5706570994286592

Requesting approval to ship?

Yes

[Rick Byers]

This is definitely important for the web, and the spec change / API surface area looks pretty low risk to me.

Still, I see the spec commit just landed yesterday, and I don't see any discussion / GitHub issues.  To what extent has the design been discussed with engineers from other vendors?  Did you intend to link to something with your "[1]"?

We won't block shipping on lack of feedback, but we do want to make sure the other vendors have had a reasonable opportunity to review the design and provide feedback before we ship.  Of course we get ~10 weeks from landing until stable, so there's still plenty of time to change things if we get feedback after landing.

Thanks,

   Rick

[Nicolas Satragno]

Hello,

On Tue, Jan 26, 2016 at 1:58 PM, Rick Byers <rby...@chromium.org> wrote:

This is definitely important for the web, and the spec change / API surface area looks pretty low risk to me.

Still, I see the spec commit just landed yesterday, and I don't see any discussion / GitHub issues.  To what extent has the design been discussed with engineers from other vendors?  Did you intend to link to something with your "[1]"?

>

 

Looks like I failed to copy the last line of my draft. Here it is:

[1] https://github.com/whatwg/notifications/issues/31

 

We won't block shipping on lack of feedback, but we do want to make sure the other vendors have had a reasonable opportunity to review the design and provide feedback before we ship.  Of course we get ~10 weeks from landing until stable, so there's still plenty of time to change things if we get feedback after landing.

So far Anne from Firefox LGTM'd the spec change [2].

[2] https://github.com/whatwg/notifications/pull/61

Thanks for the quick response,

[Rick Byers]

Ah, I failed to find the pull-request - sorry!  I see there's a bunch of history here, with Firefox having a similar API in the past.  It looks like there's still a good discussion about what else might be needed for great metrics, but I think that's orthogonal.

LGTM1 to ship.

[Dimitri Glazkov]

LGTM2

[Jochen Eisinger]

lgtm3