Restarting Xserver from cli

[Bart Silverstrim]

Mario Vukelic wrote:
> On Wed, 2009-03-18 at 15:44 -0400, Bart Silverstrim wrote:
>> Why are you reaching for control-alt-del? In windows, you can 
>> what...change your password, run task manager (right click task bar for 
>> that),...log off...? 
> 
> And lock the screen.

OOOHH, see, I got into the habit of clicking an icon on my panel at the 
top of the screen to do that...just like you're talking about another 
platform's usability habit to hit that? And it's too much of a chore to 
unlearn, so what's the next step? Hack your system to emulate Windows 
more because users have slippy fingers?

Again...how often is this even HAPPENING?!

>> Good thing X is totally integrated with the Linux kernel the way Windows 
>> is. As I recall the more integrated Microsoft made it, the more flak 
>> they got for making it easier for drivers to ruin their security model.
> 
> Yeah, except that AFAICT this has nothing whatsoever to do with catching
> a key combo in the kernel and advising X to do something sensible (as
> long as it still can). Though IANAKD (kernel dev)

Because X isn't part of the kernel, maybe?

>> Never had to. It's always worked that way. 
> 
> So. Now that's a really weak argument.

Probably...maybe...if I knew what you snipped there. Don't feel like 
checking at the moment.

>> If this is in a future 
>> release, it's because someone wants to change something that works fine 
>> the way it is, and now wants to add a new "feature" that will make it 
>> just a little more of a bitch when moving between distros.
> 
> Or has someone else but only experts in mind ...

If there's one thing evident in Linux over the years, it's that the 
developers didn't have someone in mind when making their design choices 
except themselves.

>> Blame yourself if you do that. 
> <snip>
>> If the C-A-B is killing users, set an option to disable it, not vice-versa.
> 
> It's fine not to have any clue about usability design, but it's not fine
> to use that as a basis to make usability decisions.

Good leap to assume I know nothing of usability design. You're basing 
this entirely on my argument that control-alt-backspace isn't as big a 
problem as the bandwidth we're wasting discussing it, or are you just 
trying to bait me some more?

How about just adding a package in synaptic that disables all keyboard 
shortcuts unless implicitly enabled? I mean, I've lost data a number of 
times for sticking things with cut or copy then closing the originating 
app and opening something else and...poof...not there. Had it happen 
today as a matter of fact.