On Wed, Oct 31, 2001 at 10:28:58PM -0500, Russell Hires wrote:
>
> > Everything is as it was before. The VM change seems a little
> >unusual to some people, but if that's your only reason to question
> >the versioning scheme you need to get over it...and quickly.
>
> I think this is bad for the credibility of the kernel hackers (and
> Linus/Linux especially), because it gives the idea that we can't
> trust what they say is good/not-good, stable/unstable. If people are
> (g)rumbling that the 2.4.x kernel isn't stable, or that you can
> crash it under heavy loads, how is Joe Corporate User going to know
> what he should use? Or is he going to want to use it at all, because
> he can't rely on the information that says, "2.4.x is stable"? A
> lot of people don't want to test out kernels. I know I don't. This
> can't be good for Linux.
These days stable is pretty relative. The 'stable' stamp doesn't mean
it is supposed to be bug free or even that it will work in your
environment at all. Windows and MacOS have released major releases
only to be followed closely by a .1 update within weeks of the initial
release. Heck, releasing software has got to be the best way to test
it :)
Joe Corporate User's best practice is to choose a distro that will
keep track of all of the arcane kernel development stuff and update
the distro's kernel packages according to what is _most_ stable.
This archive was generated by hypermail 2.1.3 : Fri Aug 01 2014 - 16:07:07 EDT