Re: Move proftpd to contrib
I was thinking more that, if we are going to remove a buggy package because of
the bugs, we should still provide it, since there are some people that are
looking for that package. Maybe a section of main called "buggy" if it's still
included for completeness?
Dave Bristel
On Sat, 18 Sep 1999, Robert Stone wrote:
> Date: Sat, 18 Sep 1999 15:49:37 -0700
> From: Robert Stone <talby@trap.mtview.ca.us>
> To: David Bristel <targon@targonia.com>
> Cc: debian-devel@lists.debian.org
> Subject: Re: Move proftpd to contrib
>
> On Fri, Sep 17, 1999 at 07:52:24AM -0700, David Bristel wrote:
> > Or a new section for packages removed from main due to bugs, but possibly
> > still desired by some people? It's safer to have a clear message that
> > "Debian considers these packages to contain too many bugs for inclusion in
> > the main distribution, but we are aware that there are some who want to
> > use these packages anyway." Something like this would eliminate any blame
> > if people use those buggy packages, and then have their systems crash or
> > go unstable, or get hacked. Any opinions?
> >
> I would fear it would come across like were pointing fingers at
> bad software developers in the community, as though we were putting a
> package on probation for being too buggy. I don't think our goal is to
> seperate good software from bad software.
> It might be within our scope to publish bugs per code lines per
> year statistics or other hard number observations to make that decision
> easier for others, and possibly avoid dependencies on software that has
> too high a ratio of bugs to code lines or some other weighted but
> objective comparision. A good bug vector would also give credit to
> software more widely deployed (1 in every X persons sees a bug in package
> Y every Z months).
> Our goal is in a general sense to make free sofware easier to
> install, use, and maintain. If that software has problems, it's not
> our place to single it out. At most it might be worthwhile to help
> identify where more developer effort needs to go, but if we don't have
> the resources to devote that effort, it could be harmful to point fingers.
>
> -Robert
>
Reply to: