Re: removal of submaster

Tobias Hintze <[email protected]> Tue, 14 Feb 2006 14:45:00 +0100
Newsgroups gmane.linux.distributions.rock.devel
Message-ID <[email protected]>
On Fri, Feb 10, 2006 at 02:05:34PM +0100, Andreas V. Meier wrote:
> Hi,
> 
> as discussed on irc, I want to ask to replace submaster.
> 
> Reasons:
>  - Patches get applied only rarely, like half-year-wise. They are updated two or
>  more times while lingering in the web-frontend, sometimes don't apply anymore
>  or just with manual fixes, and more negative implications of aged patches.
>  - devs that use submaster and have applied patches locally regularly get into
>  synchronization troubles when the master trunk applied patches after a long
>  time, due to the fixes that the application procedure made necessary.
> [add your reasons here]
> 
> Replacement:
>  1. experimental subversion branch, that is accessible for (web-)registered
>  users. This almost the same as the submaster webinterface, but much more
>  comfortable to use. 
>  Patches are merged to the stabilization-trunk after a  regression test. 
>  Votes can be stored in subversion properties
>  2. git (blindcoder?)
>  [add other suggestions here]
> 
> Background:
> I am a quite calm and lazy person. I continued using submaster, although it
> imposed usability regressions. It had made many ROCK developers already leave
> the project in favour of a fork. My reason for not switching was mainly the
> uncomfortable explosive ego of the fork's leader. From what I hear, I am not the
> only one that feels that way.
> We should change something, in a way that it pleases the _majority_ of the
> remaining dev's. Otherwise I fear ROCK will continue to peter out and die due
> to lack of vital ressources - developers.
> 

i can only backup this whole discussion and i agree on most of what's
been said.
all regression tests and apply-cycles as we have them now are good to
keep quality up. but they shouldn't be done for trunk but for a
stable-like tree (cause regression testing is a sort of stablization).
so i vote for the idea of an `experimental-branch` instead of submaster
and regression tests within another branch.
indeed i think that the name of this (experimental) branch should be `trunk`.

th
--


> Ciao
> 
> Andreas
> 



> _______________________________________________
> rock-devel mailing list
> [email protected]
> http://www.rocklinux.net/mailman/listinfo/rock-devel


-- 
Tobias Hintze http://hbs-solutions.de
HBS solutions GbR - Network & Information Systems