Regressions

[Rowland Penny]

On Wed, 04 Jul 2018 09:41:50 +1200
Andrew Bartlett <abartlet at samba.org> wrote:

> On Tue, 2018-07-03 at 21:50 +0100, Rowland Penny wrote:
> > On Wed, 04 Jul 2018 08:34:38 +1200
> > Andrew Bartlett <abartlet at samba.org> wrote:
> > 
> > > On Tue, 2018-07-03 at 21:30 +0100, Rowland Penny wrote:
> > > > 
> > > > And you think using an old version of waf isn't a regression ?
> > > > The waf version used by Samba is that old,it has whiskers ;-)
> > > > 
> > > 
> > > Correct Rowland,
> > > 
> > > A regression in Samba means a behaviour (that is a bug) which is
> > > worse in the new release compared with the old one.  
> > > 
> > > While undesirable, maintaining our old version of waf into a new
> > > version is not a regression.  
> > > 
> > > The inverse is the case, a new version of waf is considered a
> > > feature, and may address existing bugs that we have lived with
> > > for some time.
> > > 
> > > Andrew Bartlett
> > >  
> > 
> > No, that's your version, my version is it is something that holds
> > you back and stops optimal performance and can be fixed. Samba at
> > the moment is a bit like a Rolls-Royce fitted with a Ford pop
> > engine, it works, but it would work a lot better with the right
> > engine.
> 
> https://en.wikipedia.org/wiki/Software_regression
> 
> English is a flexible language, but your definition is not what the
> Samba Team as a whole has used for the rule around release blockers. 
> 
> https://wiki.samba.org/index.php/Samba_Release_Planning#Blocker_bugs_an
> d_time_rather_than_feature_based_releases
> 
> https://wiki.samba.org/index.php/Blocker_bugs
> 
> Thanks,
> 
> Andrew Bartlett
> 

I personally don't care, your stance is holding Samba back, waf is that
old it should be replaced, end of.

Rowland