#AskChrome We learned this week from Firefox that 99% of users don't accept notification requests, yet I'm still hearing "the Web has to compete with Native". How will you ensure the 103 features on the Fugu roadmap don't end up in the same position as Push Notifications?
-
-
Replying to @davatron5000
I'll add this to the agenda for the leadership panel - https://docs.google.com/forms/d/e/1FAIpQLSeLx2GAturxsSF4GXHOvlYXUJAeO7uDA57XKUdpNqrTN-oFug/viewform?usp=sf_link … - however my understanding from the survey is that 99% is for people being prompted onload, not that 99% don't accept when in context. need to read again though.
1 reply 0 retweets 9 likes -
Replying to @Paul_Kinlan @davatron5000
I think that's the correct understanding.
@beverloo or@b1tr0t can confirm. But to the main point, yes, we *must* compete. One way to coud be to ape native app model for push permissions where it's tied to install. Looking forward to seeing Mozilla Desktop PWA support some day.1 reply 0 retweets 4 likes -
Replying to @slightlylate @Paul_Kinlan and
Anyhow, I do think we gave notification prompts too much rope. A lot of that is down to internal churn on the Chrome team, but I'm excited about some of the experiements happening now. I expect this to get a lot better in short order w/o removing Notifications from the platform.
1 reply 0 retweets 1 like -
Replying to @slightlylate @Paul_Kinlan and
While you're thinking about that, have a look at this use case, where a push is used to update a local sw cache without a notification being triggered. Lots of great offline first use cases here, and connection with websub.https://adactio.com/journal/14511
1 reply 0 retweets 0 likes -
Replying to @kevinmarks @Paul_Kinlan and
We thought hard about that case in the design of both SWs and Notifications. If you do a silent update more than once or twice, we eventually show a notification on your behalf to let the user know work happened and allow them to revoke the permission.
1 reply 0 retweets 0 likes -
Replying to @slightlylate @kevinmarks and
One question we haven't resolved is: if the system app model allows work to happen in the background in this way w/o UI, should the web? We can throttle it, and we can restrict to only happen from networks/IPs users has talked to specific servers from before...haven't built it.
2 replies 0 retweets 0 likes -
Replying to @slightlylate @Paul_Kinlan and
There's a related one there with geolocation and geo fencing - could you have a permission so the sw can get notified by the location api (with the same kind of granularity that latest android has for foreground/background)
1 reply 0 retweets 0 likes -
Replying to @kevinmarks @Paul_Kinlan and
Great question! We've ended up with a relatively small number of geofencing requests that aren't implicitly spammy -- there was one for unlocking rental cars that was pretty good, but not many others -- and the concern there is silent tracking...
2 replies 0 retweets 0 likes -
Replying to @slightlylate @kevinmarks and
We shelved work on geofencing over privacy, but unsatisfying to not have an answer. Took a while to see it from a different perspective, but eventually
@beverloo (and I think@owencm) realised we could flip it around. The result is Notification Triggers:https://github.com/beverloo/notification-triggers …1 reply 0 retweets 1 like
Still behind a flag, but my hope is that we'll be able to expand from timed triggers to geolocation triggers at some point soon. Heading to Origin Trial soon, but available behind a flag now.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
& Web Standards TL; Blink API OWNER
Named PWAs w/
DMs open. Tweets my own; press@google.com for official comms.