Re: [gentoo-user] Re: Is GWN dead?

[Alan McKinnon]

On Monday 14 January 2008, Benno Schulenberg wrote:
> Bo Ørsted Andresen wrote:
> > Furthermore stage 1 is completely unsupported and for a very good
> > reason.
>
> Which good reason, Bo?  You seem to know it, so maybe give a link
> somewhere; don't make us guess or search.

The vast unending stream of completely useless bug reports and requests 
for help from users who had

a) chosen the wrong stage 1 or 2 for their arch
b) set the wrong flags and compile options
c) listened to ricer advice and been left with an unusable system
d) bitch and moan as to why it takes 96 hours to get a bash prompt
e) changed the install commands to "something better" which didn't work 
then consumed too much support time that could have been better spent 
elsewhere, especially since the answer usually turned out to be "don't 
try and be clever, just trash what you already did and do it properly 
with a stage 3"

when all of this was completely avoidable if they had just chosen to 
build from a stage 3 in the first place!

A stage 1 has only one purpose in life - to build a stage 2 and to do it 
in a safe way insulated from any host system.

A stage 2 has only one purpose in life - to provide something that can 
correctly run 'emerge -et system' which produces what you get with a 
stage 3 install (to a degree of course).

So stages 1 and 2 really belong inside catalyst, the more invisible the 
better (as they are just bootstrap mechanisms).

They are still around as catalyst still builds them, if you know where 
to look they are freely downloadable and can be used. But now when the 
user makes a hash of it the community can legitimately tell the user to 
stop wasting their time with unsupported stuff.

-- 
Alan McKinnon
alan dot mckinnon at gmail dot com
-- 
gentoo-user@l.g.o mailing list