Incoming from Chris Metzler:
But this assumption is wrong. The purpose of the existence of testing
and unstable is *not* to give users choices. It may also be true that
their existence gives users choices; but that's not what they're
fundamentally for. The purpose of their existence is to facilitate the
development process that produces stable releases. Users may decide to
I was around when Ian Murdock first introduced Debian. Back then, we
had SLS and Slackware, the latter having been produced because the
same un-fixed problems tended to be reproduced in subsequent issues of
SLS. Debian's raison d'etre was stability in response to the lack of
it in existing distributions.
I still think that's what Debian should be striving for. I don't see
any point in catering to bleeding-edge-itis in Debian. If the user
wants/needs newer software than stable provides, the Debian system can
accomodate that through the installation of backports or even
/usr/local. Debian has proven itself robust enough to support the
creation of dependent distributions like Libranet and Knoppix. If the
user demands bleeding edge, that's where they should be looking.
No change is necessary. If the user thinks stable is obsolete, it
should be up to them to deal with that, and that means they should
learn to add what they want onto stable, or go elsewhere. testing and
unstable are for those who know what they're doing and are willing and
able to understand the consequences, in the spirit of wanting to help
Debian produce a future stable distribution. Debian should not be
bothering to cater to bleeding-edge-itis in a misguided attempt to
open up Debian to more users. Leave that to the Libranets and
Knoppixes.