The Myth of Perfect Abstractions
summary
Argues that abstraction is an inverse problem where observed concepts hide interacting components, illustrated by the concrete claim lock is a lock because of the key and several systems modeling examples showing properties arise from interaction not intrinsic design. Recommends practical steps: model with simple heuristics, build minimal domain models, and intentionally leave abstractions incomplete so they can be revised. Useful for developers through architects and CTOs who must change designs when reality diverges.
AI-written summary · about
chapters (4)
related talks
19mFinding Your Abstraction Sweet Spot
36mOn the Spectrum of Abstraction
47mThe Axes of Abstraction
26mThe Power Of Small Abstractions
07mA New Kind of Abstraction
24mPlatform Abstractions an Asset or Liability? - Let’s Understand the Abstraction Debt Trap
29m"Turtles all the way down: Abstractions and when we should use them"
52mThe Philosophy of Architecture