Follow-up Comment #1, task #2171 (project savane):
> Currently we increase a minor revision at each revision, minor
> being x in 1.0.x.
> Since we do not make so many release, the 1.0. is not likely
> to change anytime soon.
>
> I think we should use 1.x. In others words, the release would
> be 1.1, 1.2, 1.3.
Hi,
I absolutely agree with this version numbering scheme. I always thought that
increasing the third number wasn't appropriate for the changes that have been
done between new releases.
However:
> There are occasion where a release is made only to fix one bug
> in a release made very recently. It would not make sense to
> increase the minor number. I suggest that, in this case, we
> add to the release a +1, 1.1 with a bugfix would be 1.1+1.
I don't agree here. Why introducing a new separator? Why don't we just stick
with dots? In my understanding, a release containing only bugfixes for
version 1.1 should be called 1.1.1, the next bugfix would then be 1.1.2 and
so on.
This is by far the most common practice to name/number releases in the free
software movement.
Moreover, you're worried about polluting the namespace for distros ("[...]
using - is not really a good idea as distros frequently use that to
distinguish the packaging revision."), which is absolutely correct in my
opinion.
So why don't we just stick with the simplest possibility, i.e. dots? Maybe
the plus sign gives some distros a headache as well ...
_______________________________________________________
Reply to this item at:
<http://gna.org/task/?func=detailitem&item_id=2171>
_______________________________________________
Message sent via/by Gna!
http://gna.org/
_______________________________________________
Savane-dev mailing list
[email protected]
https://mail.gna.org/listinfo/savane-dev
