On Mon, 2012-02-06 at 18:02 +0100, Jarosław Górny wrote:
Wiadomość napisana w dniu 2012-02-06, o godz. 17:55, przez Reindl
>> in your arguments if you have any.
> why do you not read the arguments?
> * a new user does not know anything about the menu
> * a new user fall into a boot problem after update
If we are considering such a newbie user as you describe, I bet this
user does not know if system update installed a new kernel or not.
(S)he probably does not know what the kernel is.
So, why do you assume, such a user, having grub menu *not* hidden,
will guess, that in case of boot problem (s)he should try to boot
Well, they do. Generally what people do when something is broken and
they don't really know what or how to fix it is twiddle: they look for
something they can poke, and poke it. A boot menu is an excellent
example of a twiddle-able interface, if you always show it: it's right
there, on boot. It's actually just about the only thing the user CAN
twiddle, if booting the default kernel doesn't work - the only bit where
they feel they can influence the process. So, in my experience, that's
what they do: they try a different menu entry. They don't actually
_need_ the knowledge of what the hell they're doing. All they need is
the knowledge that 'this is a menu with different entries and choosing
one of the other ones might make something different happen'.
Obviously this tweaking reflex can lead to disaster in _some_ cases, but
in the case of recovering from a bad kernel, it actually serves people
I agree with the kernel team that a better option would be some kind of
smart recovery from failed boot, though.
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | identi.ca: adamwfedora