using "experimental" (was: Re: GCC 3.0 status?)
> Previously Alan Shutko wrote:
> > That means that stable will probably be stuck with an old and
> > not-very-compliant C++ compiler for a long time. Are we sure we want
> > to live with that?
On Sat, Jun 23, 2001 at 07:57:43PM +0200, Wichert Akkerman wrote:
> Unless we want to delay releasing woody to 2008, yes. Switching to g++-3
> is a *major* effort.
This will prolly be shot down but: Would such a "blue sky" distribution
make much sense? I'm thinking of something like experimental, but with
more attention and full of stuff done The Right Way. Examples?
devfs. g++-3. lvm. whiz-bang new filesystems.
Most importantly: traceroute in the "right" place. :)
...things that obviously are a bad idea for today's and tomorrow's distros,
but things we should be thinking about for the far future that will be a
PITA if not well considered. Matured wild changes could be merged into
sid at strategic points, and a sane number of bug reports filed for the
remaining newly broken packages.
Of course I'm reluctant to impose on developers any more than they're
currently burdened, so I'm not sure how to go about it. I'm just feeling
out whether it's worth figuring out. What's really bad about it? Is it
even feasable. Does it already exist in the form of experimental?
- chad
--
Chad Miller <cmiller@surfsouth.com> | Policeman: You're playing God!
unix brujo, shutterbug, bookworm | Dr. Hfuhruhurr: *Somebody* has to!
URL: http://web.chad.org/home/ from ``The Man with Two Brains'' (1993)
Reply to: