ANNOUNCE: A *second* Python 2.1 release candidate!

[Tim Peters]

[Guido]
> We found and fixed a rare but serious bug in the dictionary code, and
> fixed enough small nits to warrant a second release candidate for
> Python 2.1 -- the final release is still planned for Tuesday, April
> 17.

[Don Tuttle]
> Perhaps I'm dense, but how do you expect any signifgant field
> testing to take place in just 4 days, especially when Easter,
> Passover, and US taxes are going on at the same time?

There won't be massive testing done on the release candidate for (among
others) the reasons you gave.  But the planned 2.1 release date was announced
6 months ago (on 16-Oct-2000), and has been on public display in PEP 226
since then.  2.1c2 is the sixth(!) 2.1 pre-release we've made, starting in
January:  if massive testing hasn't occurred over that stretch of time, it
never will (like all other things Pythonic, prerelease testing is a volunteer
activity, and the mere passage of time doesn't significantly increase the
number of test volunteers).

With the exception of an excruciatingly subtle and rare pre-existing bug
discovered in 2.1c1 and fixed this weekend (significant? yes! there goes your
theory <wink>), 2.1 bugs reported for some time now have been shallow and
mostly platform-specific, and mostly on minority platforms.  In the absence
of any reason to believe people who've put off testing since January are
going to change their minds now, the point of the quick prerelease is mostly
to ensure that no *new* platform-specific glitch snuck in with the last round
of minor tweaks.

BTW, I expect 2.1 is the best-tested release in Python's history, thanks to
the many people who downloaded one of the prereleases and gave it a shot.  If
you're one of them, thanks!  If not, I don't buy that it's because you've
been waiting since January to get Easter, Passover and taxes out of the way
<wink>.

although-i-have-i-just-don't-buy-it-for-you-ly y'rs  - tim