[conspire] Fwd: wheezy is OUT. tonight. Re: ANNOUNCEMENT: BAD meeting for 2013-05 (and mini-installfests, and ...)

[Rick Moen]

Quoting Tony Godshall (togo at of.net):

> OTOH, if you were pointed at stable, suddenly your apt-get update
> brings in a whole lot of changes

Yep.  But you knew that's what it said on the tin when you decided to
use a release distribution, instead of a rolling one.

> ...and there are places past which apt-get install breaks and you
> really want to do an apt-get dist-upgrade.

Hold on -- what?

I'm sorry, but you pretty much always want to do 'apt-get update &&
apt-get dist-upgrade', and particularly have an incentive to do that if
a smaller package operation like 'apt-get install foo' seems to have a
problem.  So, why are you saying this has anything to do with the
triggering of the release process?  It doesn't.

> Testing, not so much, as it is a rolling version that lags sid only by
> mechanically passing packages that don't break things after a certain
> number of days IIRC.

Yes, it says on the tin that it's a rolling distribution, so you get a
significant inflow of updates any time you do 'apt-get update && apt-get
dist-upgrade' without a sudden flood at release time because, duh, there
is no release time.

> If you track stable, you are presumably averse to uncontrolled
> breakage, and you might be well advised to point by name or by version
> number....

Arrgh.

God.  Dammit.  _Again?_

Do I have to rebut this bit of bad advice _every single time_ Debian has
a release or even when people talk about Debian releases?

I guess I need to finally get around to FAQing it, because I'm really
sick of having to debunk this stupid bullshit idea.