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

Re: experimental or unstable

On Sat, Sep 03, 2011 at 09:01:52PM +0200, Alexander Reichle-Schmehl wrote:
> * Tony Houghton <h@realh.co.uk> [110903 17:32]:
> > There's a new release ready; should I switch back to unstable for it?
> Well, ask yourself this question:  Do you think the upcoming release is
> fit to be part of a stable release?  If the answer is "yes" upload to
> unstable.  If the answer is "No, but I'd like to have pacakges ready for
> testing anyway" upload to experimental.

I kind of fail understand the logic here.

If the changes are incremental and the releases are believed to work well,
the only cost of multiple uploads is a bit of bandwidth and a bit of buildd
time -- ie, machine work.  And if you upload to experimental, that cost is
paid anyway.

On the other hand, it is a lot easier to fix problems if the changes come in
smaller pieces.  Plus, bugs introduced in the first piece will be reported
earlier, giving a better chance they'll be fixed.

So, shouldn't multiple uploads be discouraged only if they require
unnecessary transitions or are of questionable quality?

1KB		// Yo momma uses IPv4!

Reply to: