Positive on Declarative Web Push #1190
Conversation
activities.yml
Outdated
| id: declarative-web-push | ||
| issue: 1176 | ||
| rationale: | | ||
| Mozilla is happy with the overall idea of Declarative Web Push. We believe having a declarative way to subscribe to push notifications will help users by reducing power usage on mobile devices and also help web authors by reducing required code size. We also believe that this introduces a way for browsers to suppress potentially unwanted push notifications from tracking users by conditionally removing its ability to use local state for mutable notifications. |
There was a problem hiding this comment.
So... I don't believe Apple's claims about power usage here. It's obviously less power consumption to not run code, but I don't find it convincing that instantiating and running a small SW is significant relative to activating the device radio, the speaker/haptics, or other costs that come with notifications.
What I do think is that this model of notifications is probably better overall than what we originally specified. Having a clear path to a notification, with the option to intercept and update application state, is useful. Having the option be truly optional in the sense that the browser might choose not to invoke it, is also good.
I do think that this is an incomplete solution, in that having no declarative way to dismiss a notification using this mechanism is problematic. Mostly from the perspective of not enabling the creation of annoyances.
As for the rest of the text in this chunk, I would suggest that you post the other stuff on the issue, rather than here. It's enough to just say that we think this is good.
There was a problem hiding this comment.
Actually do we need a PR at all or is adding a tag to the issue enough?
There was a problem hiding this comment.
Commenting in the issue and adding a label is enough. PR is needed if we want to include a description/rationale in the dashboard.
Closes #1176