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

Re: how to make Debian less fragile (long and philosophical)

It was me, and I never complained that unstable broke my system. 

I looked into why it broke, and realized there was something more 
fundamental going on than just one mistake in the bash commit. There
was a lurking potential for danger on two fronts: 

   -- No possible way to recover a broken system without boot floppies.
      Even if the dynamic libs break through my fault rather than Debians
      (everyone eventually learns why * and "rm -rf" don't mix, usually
      through some very painful mistake)
   -- The package management code is fragile, and the developers are
      having to work around unnecessarily complex situations to keep
      it working. Even with lots of testing, no software is ever 
      bug free, and making it unnecessarily complex is just asking
      for trouble--especially with something as system critical 
      as the package manager.

So, please understand that I have no problem with the notion that 
things in "unstable" are sometimes broken. I have a problem with 
the way things are set up in the current "stable", and the way they
are planned for the next stable release as well. 

That my attention was drawn to this issue due to a bug in unstable 
is just noise. I probably shouldn't have mentioned that, since it
sidetracked too many people into thinking about the wrong thing.


On Tue, Aug 17, 1999 at 08:33:23AM +1000, Craig Sanders wrote:
> also BTW, this whole thread seems to have come about because you (or was
> it someone else who started this thread?? i've lost track) decided to
> upgrade from stable 'slink' to unstable 'potato' and got yourself bitten
> by a temporary instability. tough. that's the way unstable is.
> there is a good reason why unstable is called unstable...it's NOT
> because the programs in there are any more buggy, it's not even because
> the unstable dist as a whole hasn't been tested (although that is a
> factor), it is mostly because unstable changes rapidly and without
> warning.

Reply to: