When a compile sucks in out-of-date jars from a previous build or a cup of hot chocolate explodes in the microwave I mutter "that's not supposed to happen" and recheck my approach (after some cleanup).
Release engineering continues to teach me much about problem solving and the importance of favoring empiricism over ideologies or theorizing. Theories help simplify a situation, allowing me to "understand," by focusing attention and data into a narrow band. Many times they guide me quickly to resolution. They also act as lenses warping and obscuring reality. I usually approach problem solving by creating tiny tests with which I can easily vet ideas. My gut and sense of "what is supposed to happen" generates ideas and my little tests separate the useful from the misguided.
The Exploration Process
- Build Blows Up
- Scratch chin repeat 'not supposed to happen' like an angry Timothy Olyphant character
- Create smallest, shortest test that repeats the blow up
- Start working through ideas
In this blog, I plan to talk about problem solving in generate and release engineering in particular. There will be many articles about maven its many non-alluring mysteries.
I hope you find these articles of use or amusement.
Peter
No comments:
Post a Comment