[Development] [Qt-creator] gerrit-speak

[Oswald Buddenhagen]

On Fri, Dec 14, 2012 at 07:49:26AM -0800, Thiago Macieira wrote:
> On sexta-feira, 14 de dezembro de 2012 13.57.16, Oswald Buddenhagen wrote:
> > another reason is that any sane commit policy prescribes atomicity on a
> > commit level, so forced grouping of commits isn't much of an advantage.
> 
> Note that "atomicity" here is ambiguous.
> 
> Ossi was referring to "this change must be complete and testable on its own".
> 
indeed. as it happens, point 8.1 of our commit policy explains it like
that.

> I often read it as "it must be as small as possible" and sometimes my commits 
> are not testable by themselves.
>
this is a self-contradicting interpretation. a minimal fully functional
change is not splittable (i.e., it's atomic). therefore no part of that
change can legally exist on its own (i.e., satisfy the criterion of
being atomic), as otherwise the complete change would not be atomic to
start with.