My first DontZap use case while testing F11 beta

Anders Rayner-Karlsson anders at trudheim.co.uk
Fri Apr 17 20:32:21 UTC 2009


* Lars E. Pettersson <lars at homer.se> [20090417 21:37]:
> On 04/17/2009 08:38 PM, Anders Rayner-Karlsson wrote:
>> While I am not a representative figure of the user population, I had a
>> case today where:
>
> Yes, but a new Linux user will not be able to do what you describe. They  
> will try all things they can come up with, pressing the keyboard blindly  
> etc., and finally press the reset button, or power cycle the computer.

You are not convincing me that Zap should be enabled by default with
that statement. Maybe I am coloured by experience though.

> I can not see that this is a good way to solve a problem where a few (!)  
> persons accidentally have pressed ctrl-alt-backspace in the past.

The counter-point is that it is hard to see a good or viable reason
for having a on-by-default ability to kill your session, your running
applications and lose data, without confirmation or warning.

> Ctrl-alt-backspace *is* useful, and should, IMHO, be *on* by default, as  
> it always has been. If upstream thinks otherwise, Fedora should be able  
> to set it to on, as, as seen in this thread, and apparently in a lot of  
> others, a lot of users are *against* removing this feature.

I am not arguing that it is useful or not. I have use for SysRq's and
find them useful, but I am not arguing for them to be on by
default. In any event, flogging a stinking horse carcass repeatedly
won't get the carcass winning the next Grand National. ;)

>From one perspective, a safe default while still allowing skilled
users to switch on "poweruser" features is a better starting point
from a usability argument.

"Once you know how to start a fire, you can proceed to singe your
eyebrows off with it."

>> C-A-Bs is akin to Alt-SysRq-C if you have SysRq's enabled. The default
>> setting should be "off" so that people by accident can't trigger
>> it. If you need it - you get to enable it yourself. That is a sensible
>> default for a debug function.
>
> How often does people actually accidentally press ctrl-alt-backspace? I  
> have *never* done it.

By same argument, taking a rather tongue in cheek attitude, how often
does countries with nuclear ICBM's launch them by accident?

While you certainly can present an argument based on that you have
_never_ done this by accident, I'd like to point you at
http://en.wikipedia.org/wiki/Argument_from_ignorance for reasons why
this is, IMHO, not a sustainable platform to argue from.

> I do not consider ctrl-alt-backspace a debug function. Sadly programs  
> still hangs X once in a while, and having a keystroke available to press  
> to kill X, when all you need to do is killing X, is good thing to have,  
> and it should, IMHO, be available till we have a X that does not hang as  
> often as it still does. Why having to reboot the machine when you  
> actually only need to restart X? It does not make any sense, at least  
> not to me.

While I disagree with your classification of the C-A-Bs functionality,
lets just agree to disagree on that, I agree that some programs still
cause issues with X (having today encountered it first hand).

As has been explained *extensively* already, you will still have the
ability to do what you desire - but you will need to (trivially)
enable this yourself in future. As also has been posited already, this
has been blown way out of proportion, quite possibly based on
erroneous assumptions.

--
/Anders




More information about the fedora-devel-list mailing list