Re: [XP] Agile and CMM are contradictory
- I agree with most of what you say, but I can't help thinking of Edison. His
greatest invention was a repeatable process for inventing things. Menlo
Park has been a successful model for something approaching the continuous
invention of software development ever since.
Perhaps a model like CMM but based on industrial R&D would be more useful to
- Eric Hodges writes:
> I agree with most of what you say, but I can't help thinkingEric:
> of Edison. His greatest invention was a repeatable process
> for inventing things. Menlo Park has been a successful
> model for something approaching the continuous invention
> of software development ever since.
Scrum and XP are identical in nature to Edison's invention
process. They are repeatable but only at a high level, because
they always break a project in iterations. However, they
are not repeatable or defined at the low-level. At that level
they are just a set of practices working together.
Contrast this with the CMM requirements to document detailed
Could Edison predict always what the goal and result of an
invention, the time spent doing prototypes, or even the
sources of inspiration?
I doubt it. Yet he could still break the _high-level_ process.
We do the same in agile development i.e. Scrum or XP.
Eric Hodges writes:
> Perhaps a model like CMM but based on industrial R&D wouldThis is exactly what Takeuchi and Nonaka argued -- that
> be more useful to us.
successful R&D companies were doing something different and
that the ones following this model were much more successful.
And that's what we need to do to move the software industry
out of the tar pit:
to make the realization that we need a different paradigm
-- one that better fits our needs,