[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: PG: Hackers and Painters



Zooko <zooko@zooko.com> writes at 08:17 15-May-2003 -0400:
> I have read hundreds of articles pro- and con- and had many private
> conversations about these three topics, and I can remember only one or
> two instances of someone claiming serious, sustained experience with
> the practice and then decrying it.

I can remember only one or two instances of someone claiming serious,
sustained experience with the practice of shooting themselves in the
foot and then decrying it.  Usually, they decide pre-emptively that it's
probably not the best idea, or they wimp out the first time they lose a
few toes -- rather than hanging in there to see if it gets better. :)

At one point I was paid good money to be conversant on numerous software
engineering and analysis/design methodologies.  (There are many, many
more than you'd think, but fortunately most of them die off sometime
after the first or second book.)  Some random observations, not
specifically apropos to XP, but to keep in mind:

  1. There's lots of redundancy among methodologies.  The majority are
     thoroughly redundant and inferior on the face of it.

  2. Methodologists are often not to be well-informed about other
     methodologies.  (There was a panel exercise at one conference, in
     which experts were given the exercise of defending *other* experts'
     methods, and only one expert could speak intelligently about the
     others'.)

  3. Methodologists often make a living as consultants on their
     self-branded methodologies.

  4. A large number of people attempting to implement methodologies have
     little understanding of the methodologies.  Very common are
     diagrams and models that are just plain invalid or use only a small
     fraction of the expressiveness of the language.

  5. A lot of the benefit of the methodologies in practice is almost
     incidental.  For example, mostly a team just needed to have *a*
     convention for even minimal documentation or process.

  6. People who believe they've had some success with some methodologies
     develop cargo cult myths and religions around them.

-- 
                                             http://www.neilvandyke.org/