Andreas Jaeger wrote:
On Tuesday 06 July 2010 16:56:42 Per Jessen wrote:
Andreas Jaeger wrote:
So, because of my pain points, I'd like to have a good versioning scheme. Currently we have *no scheme*: We use major.minor without any meaning besides marketing. major gets increased whenever somebody feels like it but there's no documented way of increasing it.
I know this perhaps just me being picky with the words, but we _do_ have a versioning scheme, and it's even well-defined - our problem is that we aren't using it.
Personally and probably mostly because of habit, I favour the major.minor system, but I'm not sure it's really very well suited for a project like openSUSE whose core ingrediences come from other projects.
A major.minor scheme makes little sense unless it comes with solid planning. This is the kind of release planning where we would know the rough outline+contents+timeline of 12.0 by the time 11.1 is released. Or something along those lines, I'm sure you get the idea.
So, what you say is we follow with naming the major.minor theme but do not implement it as it's defined. In your opinion we should plan disruptive changes for 12.0 only but not in 12.1.
Yes, that is how I interpret a major.minor versioning scheme. With hindsight, perhaps we could have had the following disruptive/major changes for 12.0 - new syslog daemon, new cron, new KDE, new sysinit, new postfix. (just an example).
so, that sounds to me like we agree completely that what we do does not fit the major.minor scheme we pretend to use.
Either follow it - what seems to be your proposal - or abandon it - my proposal.
One of those two, yes. Either way, we have to stick to the scheme and whatever meaning we attach to it, otherwise it's worthless. -- Per Jessen, Zürich (25.2°C) -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org