Re: [w3c/gamepad] gamepadconnected/disconnected events & non-"fully active" documents (#149)

I think the desired goal is to move towards a world where features which disable BFCache are few and far between.

As a compromise when it comes to introducing additional complexity versus not introducing unexpected developer-visible behavior is that events should not be queued, but coalesced. That is, if a gamepad is connected or disconnected while the page is not fully active (but the page would otherwise have received the event) then that event should fire when the page becomes fully active. On the other hand, if a gamepad is connected and then disconnected or the other way around then the paired events should be dropped since the "end state" when the page becomes fully active again is no change from the state when it became not fully active.

For the input events currently being discussed by the TAG I think the same principle should apply and only events corresponding to the difference between the last observable state and the current observable state be fired.

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/w3c/gamepad/issues/149#issuecomment-896289032