Keynote: Psychology of Software Development: We’re Only Human After All
summary
Humans keep the same 1.4 kilos of jelly brains but work in a far more complex 2023 pipeline than the 1970 handoff model, so communication and psychology matter: working memory averages seven plus or minus two items, flow breaks badly on interrupts, and availability bias skews risk perception (lightning vs botulism, disease vs accidents examples). Practical fixes shown include information radiators instead of buried wikis, kill-your-darlings refactors to avoid sunk-cost traps, and explicit steps to reduce groupthink and tribal kill your darlings ownership.
AI-written summary · about
chapters (29)
- 0:00Human psychology in software
- 2:131970s dev pipeline
- 8:23Dev environments evolution 1970-2023
- 10:06Computing vs humaning skills
- 13:17Brain info processing & memory
- 18:53Human perception limits demo
- 23:43Card trick & cognitive shortcuts
- 26:30Limited attention in software
- 27:55Memory models impact
- 30:22Working memory constraints
- 33:51Flow state in programming
- 36:18Cost of interruptions
- 38:49Fast vs slow thinking
- 40:55Availability heuristic
- 44:07Ikea effect & code ownership
- 46:55Sunk cost fallacy
- 49:05New tools hidden costs
- 53:18Code review & perfectionism
- 54:22Fundamental attribution error
- 56:52Imposter syndrome
- 59:21Team communication importance
- 1:01:28Bees, languages & communication
- 1:04:18Knowledge in code & docs
- 1:05:21Maslow’s hierarchy
- 1:06:01Groupthink & Bay of Pigs
- 1:08:13Psychological safety
- 1:10:18Social identity theory
- 1:13:13Minimal differences & conflicts
- 1:15:03Embrace individuality
