Decision Dials
summary
Argues decisions are decision dials not binary switches, and designing software is constant trade‑off optimization: choose where to be robust versus malleable, minimize local optimization that breaks global goals. Gives concrete evidence: an on‑site builder reengineers plans to add a zero‑entry ramp that works, an org spends about $1,000,000/year fixing bugs, a shared schema had 800 columns, and a performance issue goes from 15 seconds to 33 ms after profiling. Recommends pair coding, profiling, and write a performance test before optimizing, summarised as profile before optimizing.
AI-written summary · about
chapters (24)
- 0:00Welcome
- 1:53Decisions as dials
- 8:32Long-term user focus
- 10:19Toilet paper orientation
- 13:51Optimism vs pessimism
- 16:48Yes or no as novice/expert
- 20:20Trade-offs in decisions
- 23:11Comfort vs cost
- 25:21Spending today versus saving for tomorrow
- 26:47Toyota’s local vs global
- 31:45Long-term philosophy
- 33:32Why programming is part of living
- 34:37Waterfall vs agile pitfalls
- 37:48Design by contract vs capability
- 41:19Language choice and layers
- 43:04Reuse vs coupling trade-offs
- 45:13Over-optimized DB schemas
- 49:07Design principles purpose
- 51:16Imperative vs functional
- 53:24Performance relativity
- 57:23Testing performance needs
- 58:28Essential characteristics
- 1:00:12Documenting the why
- 1:01:40Q&A
related talks
12mHow to make the perfect decision
42mDecision-making for developers
22mDecide Like a Pro: Navigate the Time-Quality-Cost Trifecta
39mThe Art of Technical Decision Making
27mStrategies for making impossible decisions
57mStrategic Decision-Making for Tech Leaders and Developers
01h 06mDecisions, Decisions — Dan North on Trade-offs in Software
26mIndecisions Are Not All Bad