Re: Proposing _NET_WM_STATE_FULLSCREEN_EXCLUSIVE

>> The current text just expects the window to stay on the screen it's currently
>> associated with, but perhaps that's naive?
> 
> i think so. you game will start and set this property THEN map the window... at
> this point it will get associated with a screen by the wm. it's already too
> late at this point. :)

Our current use case (which isn't the only valid one, of course) has an already-mapped window, which is why this didn't come up before. I think we'll correct this in the spec, which will solve the concerns about fullscreen windows on secondary monitors.

--ryan.