I mean sure, that is how some (mostly strategy and tactical) games do it, but for an FPS, figuring out where the buffer should be would be a programmers nightmare. I guess you would have to try to calculate all possible lines of sights a player could have within some buffer time (100-1000ms) and then all players that could in theory enter them… Add physics and it is practically impossible.

Also, corner hack is useful enough and it does not address aimbot. IMO the answer is some combination of human moderation and ability to play with “friends” instead of randos. E.g. you could ask people to like or dislike a player at the end of a match and try to pair players that liked each other in the past.

That does not detect things like wall hack and aim-bots that don’t modify the game state directly.

In XMPP, e2e encryption (just like everything else) is an optional extension. So in practice half the clients don’t support e2ee, half support different version of e2ee (can’t talk to each other) and pretty much all e2ee are likely full of holes since there are too many implementations to review.

In Matrix, e2ee is in a library that all clients can use, so while it is not Signal, it provides decent security.