> On Fri, 25 Apr 1997, Stephane Bortzmeyer wrote:
[...]
> > studied. This is clearly no longer true and I regret it. It seems Linux
> > developers lost the problem of "pure" users (not developers). 2.0.x
> > kernels have introduced new stuff and this is not acceptable (the Apache
> > break in 2.0.14 was specially catastrophic).
>
[...] (I couldn't justify quoting the whole thing <G>)
> Well do I remember Linus deciding to let 2.0 loose on the world because
> it seemed the only way to get width in the wide beta testing. I don't
> know if he was wrong, but the subsequent release flurry (software du
> jour, quoi) alienated an incredible number of the Great Unwashed whose
> support we sorely need to make this thing survive the M$ juggernaut and
> keep growing. (So ok, I'm at best half washed, and that's maybe why I
> feel strongly about it: I have to keep my boss's boxen doing their thing
> and it gets messy when the "stable" kernel releases keep dying under
> them to the montra of "NT, NT, NT!"
Why don't we just go 2.0 (Stable) -> 2.1 (Unstable) --Rename bugless 2.1
to--> 2.2 (Stable) -and Immeditaly begin work on-> 2.3 (Unstable)?
> The real problem, and it's not at all an easy one, is that stabilizing
> isn't nearly as good fun as developing. To keep BillG thinking Linux is
> the real competitor, there have to be reasonably frequent stable
> (really) releases; but to keep the folks who build Linux having the fun
> that makes it worth building, there has to be scope to play and break
> things. The two things got mixed in the 2.0/2.1 series, I think, and
> we're paying.
In part, probabaly, that's because we (or you, anyway, 2.1.12 was current
when I first tried to install Linux (and didn't screw my harddrive in the
process <G>)) simpily said *STOP - no new features beond this point*,
rather then freezing out a new stable line, and begining a new
expermential line at the same time. (Harder to get stuff straight, but
avoids a period of no new stuff.)
-=- James Mastros
Hose spele cheker deid undor the extrme lood.