how bleeding edge will the next fedora release be?

[Matthew Saltzman]

On Wed, 10 Dec 2003, Preston Crawford wrote:

> On Wed, 2003-12-10 at 20:09, Matthew Saltzman wrote:
> > Other than that, I think it's reasonable for Fedora's main line to run the
> > latest *stable* releases of packages.  Real bleeding-edge stuff belongs in
> > testing and unstable sections of repositories, so they can be tried out by
> > brave volunteers.  Those of us trying to do useful work need at least a
> > reasonable level of stability.  The Fedora main line is designed for
> > "developers and enthusiasts", but it should not be just a toy.
>
> There are some on this list who seem to disagree with you. I think they

But who?  The final arbiters are the development team.  They may have
their own goals, but if they want to take the community's views into
account, it's important for all views in the community to be aired.

> may just be pushing Red Hat products. If this is the party line, maybe
> it's time to start looking again. Please tell me this isn't what Fedora
> is about. It's started out so promising.

I have a hard time believing that Red Hat wants Fedora to be known as a
toy.  They push it as a technology preview distro, but that doesn't mean
they're not paying attention to quality.

>
> Preston

It's worth making a point that someone else made (can't recall who it was)
in another one of these threads.  We use the word "stability" in two ways:

(1) Stability == Quality, i.e., does stuff work as intended without
crashing, and are bug fixes rolled out quickly.  In this area, I think FC1
is at least as great a success as previous RHL releases.  (Of course,
every release has problems that affect some number of users, and every
release regresses for a few users, too.  FC1 is certainly not perfect, but
it doesn't seem any worse overall than previous RHLs.)  I also have to
believe (until proved wrong) that Red Hat and Team Fedora are serious
about making sure future Fedora releases continue the trend.

(2) Stability == Slow movement, i.e., can we rely on not having things
break due to upgrades too frequently.  This is important for "enterprise"
users because of the management issues, and it's important for commercial
developers (hence my desire for attention to RHEL compatibility in
Fedora).  As a technology preview, Fedora emphatically is not concerned
with this view.  But they can move forward in this area at a pace that
does not sacrifice quality.  I would like to encourage them to do so.

-- 
		Matthew Saltzman

Clemson University Math Sciences
mjs AT clemson DOT edu
http://www.math.clemson.edu/~mjs