I've been trying - I really mean it - to read Jurgen Appelo's Management 3.0 book. With lots of hype - much of it self-promoted - I bought the book and started reading. I bought John Goodpasture's book as well, skimmed it, saw what I wanted (EV and Agile), incorporated those ideas in my NDIA Information Systems Summit paper and moved on.
But I have to put Jurgen's book down after a few minutes. It just doesn't resonant with me on several levels. It's VERY and I mean VERY anecdotal. I get a visceral reaction to self referencing narrative. It's my science training, it's my mother's English professor training, it's my formal proposal writer training. Then there is the analogies that are likely very useful for many in the software development world. But my education, training, and avocation gets in the way - in a big way. I was a budding particle physicist. I wasn't a very good physicist, but I could write really good FORTRAN code for extracting signals way below the noise -60db. This type of software replaces or augments Lock In Amplifiers and other types of signal process algorithms.
So what's the problem with the book? Well the use of analogies that are flawed is somewhat OK. But about every 5th page there is a blatant - read bogus - analogy. The first one that caused me to close the book is on page 42. Jurgen says, "A double pendulum is also a simple system. It is easy to make and easy to understand. And yet, it undergoes unpredictable chaotic motion do to a high sensitivity to the initial setup of the pendulum.
First of all it is easy to make. But it is not easy to understand. The equations of motion for the double pendulum consist of four independent equations for two couple second order differential equations.
Here's the Wolfram solution derived from the classical mechanics texts every physics major has on his shelf 30 years later.
- Arnold, V. I. Problem in Mathematical Methods of Classical Mechanics, 2nd ed. New York: Springer-Verlag, p. 109, 1989.
- Wells, D. A. Theory and Problems of Lagrangian Dynamics. New York: McGraw-Hill, pp. 13-14, 24, and 320-321, 1967.
So what's the issue here? Jurgen uses the double pendulum as an example of "unpredictable" behavior. This is simply not true. The behavior is exactly predictable. The Wolfram tool solves those two coupled DiffEQs and even draws a picture. If you start the pendulum in the same place every time, it will follow the same path EVERY TIME. It has to, the double pendulum is subject to the laws of gravity, Newtons laws of the conservation of energy and momentum. That's it. The double pendulum is NOT chaotic, it is predictable. Now insert non-linear friction and its still predictable. Start it off in a slightly different position and the path is different of course. But that's just Newton's laws again. Freshman classical mechanics. OK, maybe is freshman can solve the Lagrangian for the two-body problem.
So now what? What does this to do with anything in the Management 2.0 book? Well a lot actually. The use or mis-use of analogies is at the heart of the book. The terms used to complex, chaos, emergent and all the other words used by the agilest are of course now suspect. This is not an evidence based narrative about Management 3.0. It is a personal journey through the analogies of how Jurgen thinks about complexity, chaos, emergence.
No problem, hell it's his book. But it gets better. The section titled Innovation is the Key to Survival, starts out sorta nice. But when I get to the 3rd paragraph (page 53) it all goes in the ditch. I know a lot about physics. I know some things about biology, our son is a biology guy. So when Jurgen says, "researches say that complexity - that interesting state between order and chaos - is the root of innovation in physics, biology, psychology, and beyond," I'd like to have some references. Because there seems to be a core misunderstanding about how physics works along with evolutionary biology (the current theory of everything in biology). OK, psychology I can see, since it's not really a science like physics is a science. Now these pages are interesting analogies, with some cool pictures, but they don't hang together when I got to the top of page 54 and the wonderfully misinformed quote, "innovation is doomed to fail when launched top-down program of special people assigned with the exclusive and difficult task of inventing something new."
Jurgen needs to talk to the guys at Boeing on the 787 and 777 assembling processes from the top-down steering committee, or the United Launch Alliance chief engineer on the top down processes for building stir-welded tank technology. These and dozens of other high profile, multi-billion $'s.
So here I am on Chapter 4 and I'm pretty much ready to stop. But I won't. I'll continue on. There is value in there. I'll keep looking.
The big up side is on Page 57, which I'm going to use directly on one of my programs. Jurgen claims this is the "best known" model, with no reference - typical. But I know the method, and I'm OK.
It's the model of creative process, the 1926 The Art of Thought, Graham Wallas and Richard Smith.
- Preparation
- Incubation
- Intimation
- Illumination
- Verification
So in the end, the analogies that are used for the "emergent" systems. People are certainly the most complex of systems, but the speculation that this complexity is on other domains, particle physics for example is not founded on any fact. The Lagrangian in the example of the double pendulum is also the Lagrangian used to define the interactions of subatomic particles - Quarks, Lepton, Hadrons, and the assemblages they form up to molecules. Stock markets have no Lagrangian, neither do people. Software development probably doesn't have a Lagrangian in the way an ensemble of particles do.
So without the analogies that are inappropriate and sometimes just wrong and instead provide a foundation for the management of complex processes based on actual principles, the real value of the book would arrive 80 pages sooner.
I'll start using the items above, thanks for that. And I'll keep going looking for those nuggets that I know are in there.
