On Tue, 2007-02-13 at 18:48 +0100, Johannes Wiedersich wrote: > Greg Folkert wrote: > > On Tue, 2007-02-13 at 09:54 +0100, Johannes Wiedersich wrote: > >> Kamaraju Kusumanchi wrote: > >>> Greg Folkert wrote: > >>> > >>>> Also, Kamaraju, I believe I sent an e-mail to your gmail account. An > >>>> update to your howto. > >>> For those who do not know, Greg is talking about > >>> http://www.people.cornell.edu/pages/kk288/debian_choosing_distribution.html > >> Nice page. One small comment: The plot 'mainenance problems' appears > >> twice on my iceweasel. > > > > Look at the two images they are different. > > Agreed. > > > That what the two graphs showed is qualitative. > > I understand. But still with all measures the curves give the impression > that there is practically little difference between testing and unstable > and that stable has about 10 to 20% (qualitative!!!) of the problems of > those. > > My personal experience with 15 boxes of stable and a few boxes of > testing/sid is: > > - there are more problems with sid than with testing > > - once debian is installed, there are next to zero problems with stable > > - on a quantitative plot, the differences are even bigger. On average, > there is about one (security) update per week on my stable systems. > There are about 100 times as many bug fixes in testing and even more in > sid. > > Just look at http://bugs.debian.org/release-critical/ to get an > impression of how many more release critical bugs there are in sid > (approximately the red curve) compared to testing (approximately the > green curve). > > Unfortunately, I don't know how find out how many RC bugs there are in > stable. RC bugs in stable right now is O (zero) RC == Release Critical (or so I've been told) Since Stable is already released... well you draw the conclusion. > A quick scan shows that most of those are about non-free > documentation or other issues that don't directly affect the usability > of the system. > > On a qualitative plot, I would simply put testing a bit below unstable > and stable much closer to the lower axis. Well, you see, there are LOTS of bugs discovered in the first few months after Stable release. Its a proved fact that testing doesn't get tested enough, until it is migrated to stable. Many, many latent bugs are discovered right after release. Things only tremendous amounts of testing/using will bear out. These graphs the Matt Exon made are really a "close approximation" not definitive. -- greg@gregfolkert.net Novell's Directory Services is a competitive product to Microsoft's Active Directory in much the same way that the Saturn V is a competitive product to those dinky little model rockets that kids light off down at the playfield. -- Thane Walkup
Attachment:
signature.asc
Description: This is a digitally signed message part