> Don't the isapnptools work well enough for now? I'm of the
> opinion that we should shorten the development kernels to about 1/2 the
> time it currently takes. Of course, we could only add 1/2 the features,
> but those features would get to the user more quickly, and the transition
I completely agree on this one... 2.2 should probably have
been split off at 2.1.42 (or thereabouts).
The time between new major releases should be shorter because:
- users get confused with >10 minor.minor numbers (this scares of
loads of corporate people --> less users --> less support from
hardware vendors)
- users end up using development kernels because some things
are supported that aren't in 'old-stable kernels'...
This way they introduce new bugs into their system (bugs that
shouldn't be there, because they aren't in the stable release)
- if a kernel is in devel-mode too long, loads of new features
and new (interacting) bugs are introduced... Introducing and
solving bugs should be done (ideally) one at at time :)
> would potentially be much smoother. And best of all, kernel hackers
> wouldn't feel like they have to backport features to the previous stable
> kernel (making it less than stable).
This is another (very important) issue.
Rik.
----------
Send Linux memory-management wishes to me: I'm currently looking
for something to hack...