Bug#174348: insufficient choice for boot loader location

[Zefram <zefram@fysh.org>]

reopen 174348
thanks

Eduard Bloch wrote:
>How many users use a such config?

Not many.  But why should you get in their way?

>>                                                   Firstly, that option
>> ought to be labelled accurately, as "the MBR of the first disk" rather
>> than simply "the MBR".

I note that you haven't actually responded to this report of inaccuracy.

>                                          And how should we make sure
>that lots of dumb users choose the correct MBR when they get the choice
>("hey, dude, my Linux is on hdf so I choose hdf").

By making the MBR of the first disk be the default, and putting
explanatory text such as "use this if unsure".  Hey, isn't that what
you're alrady doing?

>That is what this shell is for.

In that case, as I said, some clarity about the way in which the installer
was going to invoke lilo would be nice.  Without knowing that, it's not
feasible to make the small change necessary.

>                                Since the features you request are
>questionable and require lots of changes in the C code in Debian STABLE,
>I am closing this bug report.

I'm not asking for changes in stable.  I'm asking for the installer to be
fixed in some future version.  Don't you have this amazing structure of
`unstable' and `testing' precisely to avoid making destabilising changes
in stable?

>No install process can detect everythink for you.

Which is precisely why the user needs to be able to override the
installer.

-zefram