[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Slink release - what's left?

Chris Lawrence wrote:
> > Adam, just a general comment from my side (I realize that you are
> > not responsible for this, but you asked): Personally, I'm screaming
> > mad that I find new updates to base packages almost every day on
> > master even though we're in a pretty prolonged freeze now. Meaning I
> I think Adam was speaking in general terms (i.e. so we're ready for
> making potato stuff without suddenly running around like headless
> chickens a month before release).

Sure, I understand that, but I don't see the point in it: as long as
keep breaking all kinds of things three months into freeze, we still
around like headless chicken immediately before the real deep freeze.

BTW, the term 'code freeze' became a running gag in Linux anyway.
avoid it.
> As far as boot-floppies for m68k (and other platforms, for that
> matter) goes, I don't see any need to keep them in daily sync with the
> installed packages on master.  IMHO as long as the base system is

Ok, not quite daily. Upgrading libc hasn't broken boot-floppies in a 
while, the last Real Bad Thing was the hwclock upgrade. 
If you don't build it often enough, you'll miss such breakage until
before release. Otherwise, you waste a lot of time fixing bugs that
scheduled to be replaced by new bugs soon anyway. 

> functioning, the user's first dselect run should take care of anything
> that has been superseded.  People installing from frozen can deal with
> it until it becomes Really Frozen (tm) at the end of the week; then we
> can build new disk images from the archives and make sure they work
> with a week to spare until we're Really Really Frozen And We *Really*
> Mean It This Time (tm).

Just wake me as soon as it goes into real deep, deep freeze :-) And
looking for someone to do boot-floppies next time around. Should be a
easier on the m68k part; I'll be glad to help out creating the
and such in case I still have access to a Mac at that time.


Reply to: