On Mon, 2010-11-01 at 22:42 -0400, Matthew Monaco wrote: > On 11/01/2010 10:37 PM, Jason Reardon wrote: > > I'm somewhat of a mailing list lurker. I've read before that one thing Arch > > is lacking in are normal users testing packages. So, I wrote a script to > > install packages from testing when someone asked for a sign-off, so that I > > could give my input if needed. But, just for clarification, if I wish to > > contribute in this manner, I need to run a machine solely from the testing > > repos? > > You can get away with excluding groups of packages from testing, but I don't > think there's a good way to automate the process. I don't think there is any > real strong desire to do so either. (If there was, a solution might be to have > groups in testing which define packages in a related rebuild project like > texlive recently, xorg-server upgrade, lib* upgrade...) Additional dev work for no real purpose. As a user with only one main machine, I run [testing] and don't experience much problems. Some basic knowledge on downgrading is of course needed. Just use [testing] in totality and maybe script a way to reverse any particular update (for example, 30 packages updated, only reverse those) if there's problems.
