Re: about volatile.d.o/n
On Fri, 08 Oct 2004, Thomas Bushnell BSG wrote:
> This is what stability is about. What you are calling for is
> abandoning Debian's stability judgment to upstream's, in a situation
> where upstream isn't making any stability promises at all.
I can speak only for myself, but I can guarantee you that this is *NOT* how
I will deal with Cyrus backports, and I am certainly not doing it for
amavisd-new as a co-maintainer, either.
There are packages and packages, and there are upstreams and upstreams.
If that means Cyrus IMAPd and amavisd-new are not good enough for
volatile.d.o, well, then I will keep the backports elsewhere (such as
in backports.org)...
But really, stability MUST be the name of the game for volatile.d.o (which
really suggests to me that we should have volatile/stable and
volatile/unstable, which do *NOT* track sid, but rather work a bit like what
sid -> testing does, but with grace times on the other of a month. And
always based on debian stable). This still has nothing to do with "no new
upstream versions".
> So backport the appropriate changes only, and find programmers who can
> do a good enough job not to screw it up and destabilize it.
Often, upstream IS good enough not to screw up and destabilize things.
Often, trying to backport things in pieces IS going to be much harder and
riskier (in stability terms) than doing a full new upstream version upgrade.
This is by no means the general rule, but those of us who have such an
upstream, know it.
--
"One disk to rule them all, One disk to find them. One disk to bring
them all and in the darkness grind them. In the Land of Redmond
where the shadows lie." -- The Silicon Valley Tarot
Henrique Holschuh
Reply to: