Re: problems with the concept of unstable -> testing
On 2008-12-16 15:58 +0100, John Goerzen wrote:
> On Mon, Dec 15, 2008 at 10:52:53PM +0100, Bastian Venthur wrote:
>> Another way to see it is that unstable is constantly flowing and we're
>> just forking a stable distribution from it from time to time.
> That sounds like ubuntu. But speaking of them, how come they are able
> to do this so much more frequently than we are?
I'm not involved with them, but I guess the reasons are:
- Reduced number of supported architectures
- Reduced number of supported packages
- No notion of "release-critical bugs" -- release when the deadline
comes, no matter what. In other words, reduced quality standards.
Especially the last point is important and probably shared by other
distros that also have fixed release schedules (e.g. Fedora). For
users, the rule of thumb is "don't touch this in the first six weeks
after the release if you're prudent" because many severe bugs are
discovered and/or fixed _after_ the release, not before it.
Whether Debian's higher quality standards really compensate for the lack
of up-to-dateness is another question, of course.