Keynote: BDD - Bias driven development
summary
Software teams repeatedly make specific, measurable mistakes: building a homegrown ORM instead of using a standard (not-invented-here syndrome), “fixing” performance by touching code without profiling (placebo fixes), and repeatedly restarting Jenkins to hide flaky tests (ostrich bias). One anecdote: a document-repo for 100 users is reshaped into a social network for 100 million — a bicycle turned into a car without redesign. Strong rule: measure before optimizing and choose existing tools when they fit.
AI-written summary · about
chapters (24)
- 0:00Introduction
- 1:53Understanding cognitive biases
- 4:52Fast vs slow thinking
- 5:15Framing effect in development
- 7:52Confirmation bias and overconfidence
- 8:57Bike shedding in meetings
- 10:45Narrative fallacy in product design
- 12:33Bandwagon and placebo effects
- 13:39Hostile and Ikea syndromes
- 14:42Pro-innovation bias
- 15:47Ignaz Semmelweis and data
- 18:17Dunning-Kruger effect
- 19:03Peak of Mr. Stupid
- 19:24Availability heuristic
- 20:08Biased blind spot
- 20:32Engineers vs artists
- 22:19Goal of software engineering
- 23:02Stop gamification and methodology
- 23:47Learning from civil engineering
- 24:54New isn't always better
- 25:35Measure before optimizing
- 26:41Professional toolbox
- 27:50Bias in this talk
- 28:11Conclusion and doubt
related talks
44mBehavioral Software Engineering
39mKeynote - The gordian knot
01h 00mKeynote: What really matters
38mKeynote
41mSimplicity as a Superpower
50mTactics and strategies on software development: how to reach ...
42mKeynote: Software Design in the 21st Century
48mThe Effective Software Architect - Work Smarter, not Harder