[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Debian 2.2 Release.

Jeff Breidenbach wrote:
> Because we have data, this can be an scientific question rather than a
> philosophical one (woohoo!). I suggest the following experiment:
>   Review the list of grave/release-critical bugs in Woody right now
>   vs. the list of grave/release-critical bugs in Potato right
>   now. Take into account the number, nature, and causes of those bugs.
> Which hypothesis (A, B, Both? Neither?) does the data support?

Lies, damn lies, and statistics.

Your little experiment assumes that the same number of people are
testing frozen and unstable, that the only changes being made to unstable that
may produce RC bugs are upgrades of leaf packages, that developers fix RC
bugs equally quickly whether they are in frozen or unstable, and that people
place them same severities on bugs in unstable as they do to similar bugs in
frozen. I expect none of these assumptions are correct.

Anyway, see Buddha Buck's post. We have hard data that slipping updates
into frozen does not work. We've tried it time and time again, and saw
delays every time. 

(Perhaps we don't really have any evidence that *not* fiddling with frozen
leads to a faster release, but I'd like a chance, just once, to actually try
it, without people butting in with their pet updates! We're almost at
release here. SHEESH!)

> >I was talking about applications (see my list : mozilla, freeciv,
> >gnumeric...).  They usually can't break anything but themself.

Such packages still have to be integrated into the system. They still
may not be stable in the environment they are placed in. (And how do you
expect to update gnumeric without updating gnome-libs?) See Nils
Lohner's post.

see shy jo, tremendously annoyed

Reply to: