[dev] [commits] Horde branch mnemo_4_1 updated. 7324877f91d83f8de26d41ba7ca05be559255994

[Michael M Slusarz]

Quoting Jan Schneider <jan at horde.org>:

> Zitat von Michael M Slusarz <slusarz at horde.org>:
>
>> Not to mention that if you actually clone our git repo, what is  
>> ALWAYS the default branch?  master.
>
> And this still makes sense to me, because I think that people expect  
> a usable, mostly stable code if they checkout our source. And this  
> is exactly what master gets them. But whatever we decide to put in  
> master, we should communicate this better.

This argument goes both ways for me.  In the current state of the  
horde repo, master doesn't make much sense because everything is all  
over the place.  Some things are "stable", some are "unstable", some  
are "kinda stable but new features are being developed here" and some  
are "stable and new features are taking place elsewhere".  So this  
portion of the discussion becomes moot if/when we reorganize the repo  
layout.

As for what the correct usage of master is... I understand the  
argument to have master be the "quasi-stable" branch which is current  
release + patches.  However, my personal usage when I checkout the  
actual source via a VCS is to see what up+coming features are in the  
pipeline.  Someone running a stable branch might be interested in  
patches to that stable branch, but they can get these patches from the  
bleeding-edge branch (presumably).  Also, IMHO it is easier to clone a  
repo and then checkout the branch needed; this should be fairly  
straightforward by looking at the branch names.  It can be more  
difficult to do the opposite - clone a repo and then determine which  
is the "true" development branch (i.e. IMP now has a imp_6_1 and  
imp_7_0 branch.  The imp_6_1 is really the bleeding edge branch, but  
you wouldn't be able to know this from the names).

michael

___________________________________
Michael Slusarz [slusarz at horde.org]