How many of you are Extreme Programmers?

[John Roth]

"Michael Chermside" <mcherm at mcherm.com> wrote in message
news:mailman.1050512847.28150.python-list at python.org...
> Christopher writes:
> > So how many of you guys use XP processes?
>
> Well, I'm not sure if I qualify or not. I really LIKE most of the
points
> that XP makes, but I find many of its proponants overly exuberant. I
> think exhaustive, automated tests are a wonderful idea. I believe in
> refactoring. I think pair programming works very well most of the
time.
> I think super-short release cycles (with code that grows feature-by-
> feature) is a great idea. I think having a customer on-site is
extremely
> helpful.
>
> I DON'T think that it's wise to just start coding without a bit of
design
> up front (although one must keep YAGNI in mind). I don't think that
using
> the code itself as the only documentation (no design docs) works well.
> Most of all, I don't like mandating any particular methodology... the
> needs vary from project to project and team to team.

At the risk of being overly exuberant, most XP practitioners do
some design up front (enough to get a handle on the problem, at least.)
They just don't treat that design as cast in stone. And I've never seen
a project that wouldn't be helped by a few pages of carefully thought
out introduction to the basic concepts of the thinking behind it. With
UML if it is helpful.

> But none of that really matters, because I don't get to set the policy
> where I work. We don't do unit tests (except for the ones that _I_
write)
> because "they would take too long". We deliver functionality in a
single
> big release. We don't do pair programming except for very brief
periods.
> Perhaps someday I can change those things, but for now, I don't get to
> practice much of this, except in my at-home-on-my-own-time projects
> (basically non-existant since my son was born 4 months ago).
>
> -- Michael Chermside
>
>