PEP Parade

[Donn Cave]

Quoth "A.M. Kuchling" <akuchlin at ute.mems-exchange.org>:
| In article <a6b91d$oau$1 at nntp6.u.washington.edu>, Donn Cave wrote:
| > but putting us in kind of a backwater.  If we sort of collectively
| > agree on 2.1.1 (or whatever) for that role, though, that backwater
| > could get pretty lively.
|
| Seems unlikely; if people can't muster enough interest to participate
| in handling bugs and patches in the current CVS tree, the likelihood
| of taking on the even drier and duller job of backporting fixes seems
| an order of magnitude less probable.  Unless someone can be found to
| pay money for it, of course.

Could be just the opposite, though.  If you're submitting a fix so that
it will turn up in the next release with everyone else's fixes, then
people working with 2.1.x have it pretty good (because the next release
is a minor one, not 2.3 or whatever comes out of the CVS tree next.)

But I really wasn't thinking of fixes so much.  If there are lots of
people using 2.1.2 and it has a serious bug, then I'm sure someone
will come up with a fix sooner or later, but I was thinking more in
terms of 3rd party work, so to speak.  Packages.  Like the phenomenon
Roy was describing, but on a more global scale - you find out there's
a cool module out there, but alas, you'd need to upgrade to use it,
and the upgrade would be more pain than it's worth.  Steve Turnbull's
solution "So you standardize locally", but applied at this scale -
so you standardize globally.  2.1.2 is really a step forward for the
Python world as a whole, since there's a massive 1.5.2 contingent out
there right now.  I don't know, there's been some gnashing of teeth
about Redhat's release schedule vs. Python's, maybe they and other
platform vendors really need a not-so-current release that's still
somewhat active.

	Donn