RFC: Kernel changes that may affect desktops

[Dimi Paun]

On Tue, 2009-06-30 at 16:46 +0100, Matthew Garrett wrote:
> > Can't the desktop inform the kernel if it can handle the
> interaction?
> > If not, you can just fallback to the current behavior.
> 
> Somewhat, but you then hit issues like fast user switching potentially
> involving desktops that support this and desktops that don't.

But this is more a theoretical concern than practical.
It would seem to me that you would still end up doing
the "right thing" in a lot more cases than otherwise. 

And fast user switching could be instrumented to tell 
the kernel the right thing.

The current behavior is useful for cases when you don't
want user interaction (either because some desktops do not
have the manpower to do it, or the use case for the box
is such where such interaction is not desired).

-- 
Dimi Paun <dimi at lattica.com>
Lattica, Inc.