Psychotherapy for Code
summary
Gives concrete techniques and data: version-control hotspots and change-frequency graphs that reveal a power-law distribution across ~250–300 codebases, a surprising claim that most work concentrates in a tiny fraction of files. Shows a real 93,000-line test suite in Roslyn as evidence, visualizes change-coupling across microservices, re-evaluates Brooks law with timelines, and demonstrates refactor steps using chunking to lower working-memory load. Ends with tooling, repo links, and a book for hands-on followup.
AI-written summary · about
chapters (15)
- 0:00Start
- 6:14Process Loss via Coordination
- 8:18The Mythical Man-Month
- 11:54Measuring Brooks's Law
- 13:08Version-Control: Telling the Story of a Codebase
- 22:00A Brief Detour
- 24:53How Do We Define Beauty?
- 27:57The Arrow Pattern
- 29:26Working Memory: A Cognitive Bottleneck
- 31:41Chunks: Stretch Your Working Memory
- 34:08Refactoring the React Hotspot beginwork
- 38:15C# Nullable Reference Types: In Context
- 40:14Organizational factors are some of the best predictors of defects
- 43:34Analysing Microservice Architectures
- 45:48Microservice Dependencies: The Impact of Change
related talks
40mTreat Your Code as a Crime Scene
49mTreat Your Code as a Crime Scene
39m7 Secrets of Maintainable Codebases
47mTreat Your Code as a Crime Scene
31mMaking sense of frontend code with forensic techniques
56mCode & Homicide: What Software Devs (..) from Offender Profiling
46mMeet the Social Side of Your Architecture
38mCode reading