At least there will be a setting for users to allow any app from any dev, which those who know about it will enable right away, and those who don't know about it have no business to use it anyway.

Personally I think the easiest would be to support uBO or at least have a built-in engine that can load filters written for uBO. Anything else is just wasting development time.

Probably impossible if the original code is publicly available since most code oriented LLM just gobble them all.

Later on, when everyone has their own (which are also highly accurate due to tower/network regular syncing) on their phone, I noticed that moving to a new place/classroom/office no longer involve ensuring that we have a wall clock. Nice if there's one, but otherwise we don't care, and if the battery run out, we can't be arsed to buy one and just put it in the shelf.

At this point appliances with accurate clock would only be those connected to the internet (because otherwise TLS won't work if they drifted long enough), dumb devices won't have battery-backed clock because that will cut a single cent of profit which obviously a tragedy that must be prevented at all costs.

In practice this phone's YouTube reviews show it's running dog slow while lacking basic features, just like pretty much any other recently released feature phones at its price point.