rethinking patch management with GIT / topgit

[Enrico Weigelt]

* Manoj Srivastava <srivasta at acm.org> schrieb:

>         I hope you never publish those branches, then. For private
>  branches, rebasing is just fine. My branches are not private.

Actually, I *do* publish them (not everything yet). But I always
tell my downstreams to rebase and *never* merge. As long as you
follow that rule, everything's fine.

BTW: I also *always* rebase downstream branches before merging them 
into upstream, so there'll be no conflicts. This all happens in a 
context of lots of other QM sanctions (eg. *never* checkin auto-
generated files, run through a dozen of sysroot'ed crosscompilers,
etc, etc).

> > Yes, I push it to those who are naturally responsible.
> > The term "distro" comes from distribution, not feature development.
> > These are two fundamentally different issues.
> 
>         If you think that distributions do not develop, innovate, or
>  are full participants in the free software community, you definition of
>  distrbution does not match mine.

Please differenciate between people and roles. If you're in the 
role of a packager, your job is to create an automatically 
installable package for your distro, out of your upstream - 
nothing more, nothing less. Of course, you - as a person - can
(and should) also take part in core development. But that's a
completely different role, which implies different constraints
and workflows.


cu
-- 
---------------------------------------------------------------------
 Enrico Weigelt    ==   metux IT service - http://www.metux.de/
---------------------------------------------------------------------
 Please visit the OpenSource QM Taskforce:
 	http://wiki.metux.de/public/OpenSource_QM_Taskforce
 Patches / Fixes for a lot dozens of packages in dozens of versions:
	http://patches.metux.de/
---------------------------------------------------------------------