Trackin' stable vs. releases

[Rick Moen]

Quoting Alan DuBoff (aland@softorchestra.com):

> On Thursday 09 May 2002 10:20 pm, Rick Moen wrote:
> > You say you don't like testing, but you're _using_ it.
> 
> Currently testing is becoming stable. Step back in time and testing is
> really unstable.

No.  Absolutely not.  You seem to be missing the really essential point
about the quarantining script that populates "testing".

For background, none of this was possible until package pools went in.
At that point, specialised branches for any desired purpose could be
created without taking up any significant disk or bandwidth on the
mirror sites -- including auto-populated ones like "testing".

It's my understanding that, at first, packages auto-populated into 
"testing" if they'd been present after dupload without replacement or
retraction for fourteen days, and compiled without error on all
supported CPU platforms.  Subsequently, a more sophisticated quarantine
algorithm got substituted for that, described here:
http://people.debian.org/~jules/testingfaq.html

So, testing is the cream of "unstable" packages, those that pass
automated quality-control checks.  Testing isn't ever what unstable
"was".

[snip:  you aren't on unstable]

> I was until 2 days ago, and there was only 1 package different in
> woody/testing from sid/unstable, libstdc++ 2.95.2-14 rather than
> libstdc++ 2.95.2-13 (those are the older libs, 2.95.4-8 is more
> current and linked with most packages. 

But there's a vital difference in kind:  Regardless of how _many_
packages are in the delta, the delta is what hasn't passed quality
checks.

> It makes sense to me to change to woody now, it's going to be 
> stable very soon.

Do you mean "it _did_ make sense for me to change to woody _recently_"?
Because I thought we just settled that your sources.list says "woody",
_today_.  

I still don't understand why you think the testing branch is good today,
but won't be good later.  But whatever makes you happy.  (And note my
working hypothesis from a few minutes ago.)

>> I can't even parse semantically the sentence that ends "it won't
>> affect me".  It's grammatical English, but I can't determine what it
>> means.
> 
> Oh, I'll try and make my English composition better for you next time.
> Maybe you could check the spelling, I didn't do that.;-)

See, this is my reward for trying to be considerate and tactful.  What
you wrote _was_ indeed "it won't effect me".  But I didn't want to call 
attention to an obvious solecism:  The verb "effect" means to cause, or
to put into effect.  It seemed obvious that you meant the verb affect,
which means influence.

So, to be polite, I substituted the word you meant to use on the fly,
and wasn't going to call attention to the usage error.  Thus fail many
good intentions.

-- 
Cheers,     "You have acquired a scroll entitled 'irk gleknow mizk'(n).--More--
Rick Moen    This is an IBM Manual scroll.--More__
rick@linuxmafia.com You are permanently confused." -- ADOM (a roguelike game)