How Not to Be Wrong in Software Engineering
summary
Covers specific heuristics and math applied to software: a rejected recommendation-system project for primary schools that would likely trap weaker students rather than help them; a coin/Bayesian example (five heads shifts prior to ~12%) illustrating bayesian inference for retrospectives and SLA reactions; practical costs of an 85% code coverage gate (SonarQube) that slowed exploratory work; real work on Dutch grid flexibility using Java and customer-side load as the actual built domain.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 14:10Linear thinking pitfalls
- 25:28Bayesian inference explained
- 33:23Expected value and utility
- 41:34Unlikeliness and black swans
- 48:15Priming and decision bias
- 52:05Summary and takeaways
related talks
30mBehavioral Software Engineering
44mBehavioral Software Engineering
01h 00mCognitive biases in software development
01h 02mCognitive biases in software development
29mOn How Your Brain is Conspiring Against You Making Good Software
27m'The Art of Software Development'
21mDerailing irrationality
01h 06mKeynote - THE MISMEASURE OF SOFTWARE