[Qgis-psc] Backporting?

[Sandro Santilli]

On Thu, Mar 06, 2014 at 11:38:39AM +0100, Juergen E. Fischer wrote:

> On Wed, 05. Mar 2014 at 12:28:12 +0100, Jürgen E. Fischer wrote:
> > But this comes up after each release, because the previous stage isn't tested
> > enough and hence the release branch gets a few early fixes that also find
> > their way into the packages (although there currently is no point release).
> 
> And this IMHO also applies to the potential point releases.   The backports
> also would need testing.  And that's what we - the community - do too less in
> master already.  Personally I don't see that as a dev task in a OS project.

I understand your concern, but judging "safe" from "unsafe"
backports is something that can be learned and will improve.

It will surely happen to introduce regressions in a patch-level
release, but would statistically be way less frequent than that
happening in a feature release, and gets better over time.

--strk; 

 ()  ASCII ribbon campaign  --  Keep it simple !
 /\  http://strk.keybit.net/rants/ascii_mails.txt