Fwd: Is disabling ctrl-alt-backspace really such a good idea? - no.

[Thomas Jaeger]

Remco wrote:
> Every program that hangs but doesn't release grabs is a problem. You
> could certainly implement some kind of solution to that, but only
> after that solution is implemented, C-A-B or equivalents should be
> disabled. Not before.
I know that this is possible, but the question is how common this
situation is.

> Every program that makes the system so slow that it becomes unusable
> is a problem. This can't be solved. Any buggy program can take so much
> CPU/RAM/IO that your mouse pointer moves only every 10 seconds and
> your key presses only register after that same time. Only a simple key
> press that kills the program can solve this. Or a hard reset. We
> shouldn't remove the reset button functionality either.
If an application is hogging resources, you'll want to kill that
application, not X.

> This means that C-A-B or equivalents can never be disabled.
I think should be pretty clear by now that they can :)

> Concrete examples: Windows games run in Wine. Some will leak memory or
> whatever. You can't file bug reports against those games. And even if
> you could, you would still experience the problem until it was solved.
Okay, now we're getting closer.  There's no reason you can't make wine
handle those situations more gracefully.  But really, if you're doing
highly experimental things like running windows games in wine, it's not
unreasonable to expect that you do the tiny amount of extra of work if
you want enable C-A-B.