Eliminating upgrade risk

[Robin Becker]

well in the last two years I've upgraded four times. It is never a
pleasant experience; I'm lucky enough to be able to follow many of the
discussions and thus get some kind of handle on what's likely to go
wrong.

I think it's foolish to expect everyone to be so lucky. Most people who
use python don't use it full time. The C code I wrote in 1983 still
compiles, my python code from 1999 probably has bugs if I use 2.1.

If the advice from the Gods is that Python is a 'research' language then
I'll know better what to do. Please inform us. Skip's posts in the dev
list seem full of promise.
-- 
Robin Becker