31 Jul
2009
31 Jul
'09
2:46 p.m.
On Fri, Jul 31, 2009 at 8:54 AM, Jean-Baptiste
Mestelan<mestelan(a)gmail.com> wrote:
2009/7/30 Ken Restivo <ken(a)restivo.org>
In my opinion it depends on the number of folks running the distro and
the process the distro maintainers use to release new packages. If
maintainers rubber stamp anything that appears on their source code
doorstep as stable then I suspect end-users see problems. On the other
hand, if the distro has a large enough install base, and then within
that install base enough people running the new/testing packages, then
after some period of time without leading edge users reporting
problems the testing code moves to stable and folks running stable
don't see too many problems.
I suspect that Arch is similar to Gentoo in the sense that you choose
what sort of system to run, and then if you run stable you can
probably pick individual testing packages? On my Gentoo machines I run
the system as close to pure stable as I can and then only unmask the
few Linux Audio apps I want to run leading edge on. It reduces the
amount of time I spend building system software and keeps me focused
on audio testing versions..
- Mark
Linux audio stuff moves fast, and works better
with rolling releases,
This was also the point made in a recent thread ('Audio distribution
proposal') ; I bought the idea enough to give ArchLinux +
archaudio.repo a try. And this worked pleasantly well : in a few
hours' time, I could set up a fast and lean system, with good
performance for the main audio apps.
Still, a few updates later, I got to think again about this 'rollling'
model : does the constant upgrading not mean that you are constantly
introducing instability into your system ?