Re: Squeeze Frozen, NOT
On Tue, 2 Mar 2010 16:18:28 -0500 (EST), Freeman wrote:
> With the same amount of resources (the same number of DDs), but less
> time of course you have to cut somewhere. One possibility would be
> reducing the amount of new features, as in not package the absolutely
> latest version of softwares, but focus on versions that are already
> known to be stable.
Doesn't that defeat the primary purpose of "timed freezes?" Wasn't the
whole idea to keep the stable release more current? My point was that
there is a trade-off between the three variables of manpower, elapsed
time, and release goals (including project quality). Historically,
Debian has regarded quality most important. "It won't be released
until it's ready". With quality and manpower both fixed, time is the
free-floating variable. Now it appears that they want to make
time a fixed variable. And since manpower is not easily increased
with a pool of donated labor, that leaves quality as the free-floating
variable. If they really are serious about holding quality to the
high historical standards, then timed freezes become meaningless.
Regardless of when you freeze, it won't ship until its ready. What
that really means is that the average time between freeze and release
will increase. But the length of time between releases will stay about
the same. Actually, it will increase, unless new manpower can be brought
on, because the size of the distribution continues to increase.