on Thu, Aug 30, 2001 at 07:16:24PM +0200, Viktor Rosenfeld (rosenfel@informatik.hu-berlin.de) wrote: > Ross Burton wrote: > > > > On Thu, 2001-08-30 at 17:09, Robert L. Harris wrote: > > > > > > > > > hmm, Ok, give me a list of stability: > > > > > > potato most stable > > > sid (unstable) next most stable > > > woody (testing) "least" stable > > > > > > Problem is manage ment doesn't understand "shortly after" they want > > > an "average # of hours/days" etc. > > > > Wrong order, unstable is less stable than testing. Testing is packages > > being tested for stable. > > I've found unstable to be of better use than testing. The reason is > that even bugfixes need at least 10 days to go into testing, whereas in > unstable they could be included the next day. ...but not security updates, IIRC. These should be available immediately. By keeping sources lines for the current and all higher stability levels, plus security, fixes, your system will be pulling the most current, latest-release, packages. Or so goes my understanding. > I've been badly bitten on some occasions by a testing dist-upgrade > that left my system broken. > > Granted, unstable is also broken sometimes, but I have found that the > errors are much easier to fix than testing errors. And if I can't get > it to work, it'll usually repair itself with the next dist-upgrade. There *is* a bit of a paradox with testing: because of the time-delay, upgrades aren't made as quickly as in unstable, but bugfixes are also delayed. This can be good: often fixing a bug introduces others. But small fixes addressing sufficiently severe bugs should have priority IMO. -- Karsten M. Self <kmself@ix.netcom.com> http://kmself.home.netcom.com/ What part of "Gestalt" don't you understand? There is no K5 cabal http://gestalt-system.sourceforge.net/ http://www.kuro5hin.org Free Dmitry! Boycott Adobe! Repeal the DMCA! http://www.freesklyarov.org Geek for Hire http://kmself.home.netcom.com/resume.html
Attachment:
pgp8mkKqjTJay.pgp
Description: PGP signature