Re: Frozen distribution?

[Seth Arnold <sarnold@willamette.edu>]

* Julian Gilbey <J.D.Gilbey@qmw.ac.uk> [010218 05:00]:
> Why suddenly change the model like this?  Would the following not be
> better and perhaps less confusing, still using a four-tier setup:
> 
> stable   frozen   testing   unstable
> 
> Initially, frozen is set to be the same as testing, and we can allow a
> two-week period for testing to catch up to unstable as things are
> built for multiple arches etc.  Then during the freeze, uploads to
> frozen get installed into testing, if there are no problems, they get
> moved into frozen.  Then stable and unstable continue as before.  I
> think that it saves people having to start using (and abusing)
> experimental.

While this is certainly simple enough to describe (a major virtue) I
don't know which distribution I would want to run. I don't know enough
about frozen and testing to determine which would a) help me the most to
run and b) help Debian the most for me to run, nor whether I would have
to make frequent changes to my /etc/apt/sources.list file.

Does anyone have statistics from the servers to give rough estimates of
how many people run frozen during the freeze? I am inclined to guess a
small number of people (<5% of debian users?) actually run the frozen
distribution for any length of time. By splitting frozen into frozen and
testing, are there enough users to get adequate coverage for both frozen
and testing?

Perhaps a one-line description for each distribution of what sort of
users would want to run that distributions would go a long way towards
convincing me that a fourth state is necessary.

-- 
Earthlink: The #1 provider of unsolicited bulk email to the Internet.