2 ms·
The problem is far beyond this scope, but one thing that made a huge difference for me was Android introducing the (mandatory) idea of "Channels" to notificatio
by rektide 6y ago
The problem is far beyond this scope, but one thing that made a huge difference for me was Android introducing the (mandatory) idea of "Channels" to notifications.
Totally up to the developer how they want to use the feature, but the idea is that there's bunch of different channels & that users can pick & choose from among them gives the user some control, some ability to decide for themselves what is important.
I introduced an issue for Web Notifications to add similar: https://github.com/whatwg/notifications/issues/154 https://github.com/whatwg/notifications/issues/154
In general though things are a mess. I'm talking about adding more complexity. Websites used to prompt for adding notifications, which was a pain, but now they put some HTML on screen instead to prompt for notifications, since to be friendly/secure it was decided user-interaction would be required to bring up the prompt for notifications. So websites still have harassing notifications prompts, just on the page instead of in the browser. And cookie and gdpr notifications. And a chat agent-bot pops up to ask if you have any questions. Ugh it's silly. So many affordances that all compete with one another. No standard place where they can reside, & be viewed or ignored as the user would prefer. The whole permissions model on the web has this problem: that permissions are interactive demands, versus something more passive (like living in the awesome-bar).
- TeMPOraL 6y ago> In general though things are a mess. That's because we're dealing with malicious actors. They actively fight any attempt to make things better. > Websites used to prompt for adding notifications, which was a pain, but now they put some HTML on screen instead to prompt for notifications, since to be friendly/secure it was decided user-interaction would be required to bring up the prompt for notifications. The popups for adding notifications have an additional reason for being: it's because if you deny the request that's displayed by the browser/system API, the site can't ask you again. Meanwhile, if you deny their in-site/in-app popup, they get to try again. And again. And again. > So websites still have harassing notifications prompts, just on the page instead of in the browser. That's because they want to harass you. That's the whole point. And that's why I call these actors malicious. > And cookie and gdpr notifications. These are another cases in point. The way cookie laws and GDPR are structured, if you're not doing anything shady or malicious, you don't have to show any popup! You only need the popup if you really badly want to do something most of your users would prefer you don't. The problem boils down to: most sites and apps you encounter want to abuse you in one way or another (most commonly, funnel you into a deal that's good for them but not necessarily for you, learn everything they can about you and pass it on to third parties, and/or shove ads in your face). This is not solvable at a protocol, or web standard level. It's, unfortunately, a cat and mouse game.