Chapter Fourteen. Ptolemaic Reasoning

All models are wrong; some models are useful.

GEORGE BOX

Knowledge itself is power.

FRANCIS BACON

A few years ago I gave a talk titled "A Tale of Two Projects." My objective was to provide a glimpse of what an engineering approach to software development looks like. I argued that good software organizations often succeed on complex, high-risk projects because they use effective software development practices, while poor organizations often fail on simple, low-risk projects because they use poor practices. The talk was generally well received, but one person's evaluation said this: "An excellent example of the kind of static, linear, Ptolemaic reasoning that ignores the dynamics and complexities of real software projects."

In case you've forgotten your grammar school science, Ptolemy was an astronomer who lived about 100 A.D. and believed that the sun revolved around the earth. His theory was replaced in 1543 when Copernicus theorized that the earth revolved around the sun. The attendee was saying that the approach I recommended was a sun-moves-around-the-earth approach either that, or that my ideas were 400 years out of date!

Copernicus replaced Ptolemy's theory when he discovered observational data that Ptolemy's theory could not account for. Similarly, data from real-world software projects strongly supports the efficacy of the engineering-oriented development practices I described, this attendee's particular comments notwithstanding.