LFS-6.0 print process

Jeroen Coumans jeroen at linuxfromscratch.org
Thu Jul 8 09:58:14 PDT 2004


Ian Molton said the following on 08-07-2004 18:35:
> On Thu, 08 Jul 2004 16:43:08 +0200
> Jeroen Coumans <jeroen at linuxfromscratch.org> wrote:
> 
> 
>>>testing should be where we distill the best parts of unstable and test / stabilise them, ready to become the next stable.
>>
>>Yes, and rather then add another branch, it would be much easier to work 
>>from stable.
> 
> 
> I dont see why.
> 
> testing as it is currently proposed is FAR nearer to unstable than stable

Good point. I guess that ideally, we'd have a development team which 
works on unstable, and a seperate testing/release team which works on 
testing/stable. The testing team would keep up with unstable to merge 
small fixes (spelling, new minor package versions, etc.), perhaps 
regularly release a new x.x.x release. That way the two wouldn't be so 
far apart, and we'd have much easier and frequent releases. For a new 
major release, the testing branch would merge the larger, more 
structural changes as well, perhaps in a seperate branch.

But it all depends on what people find important and are willing to 
spend time on. If everyone is happy with adjusting unstable to become 
stable through a testing branch, then that's fine with me. I just don't 
want it to become a backdoor through which highly controversial changes 
(such as udev, hotplug) get ported into testing/stable without discussion.

-- 
Groeten/Greetings,
Jeroen Coumans
{faq,website}@linuxfromscratch.org
www.jeroencoumans.nl



More information about the lfs-dev mailing list