What Do You Mean?
summary
Argues that software is fundamentally a system of codified knowledge, not just machine instructions, and shows concrete failures when meaning is missing: C++ library semantics around C++ std::sort omit “permutation of inputs” and undecidable preconditions, letting pathological implementations (permutation sort, bogo sort, sleep sort) pass naive tests; contracts and formal logic don’t fully contain meaning; prioritizing by estimated business value is inherently probabilistic; systems change the world and expose hidden assumptions only when contradicted, so designs must admit iterative learning.
AI-written summary · about
chapters (24)
- 0:00The quest for meaning
- 2:55Abstraction and knowledge
- 5:17Software as a system of meaning
- 7:17Codifying knowledge in code
- 9:24The human side of systems
- 11:42Communication and collaboration
- 14:44Architecture and design
- 17:38Iterative learning cycles
- 20:05Planning and improvisation
- 23:25Language and algorithmic logic
- 28:05Formal methods and logic
- 33:20Algorithms vs procedures
- 36:46The reality of assumptions
- 40:15Focus and perception
- 42:17Limits of communication
- 45:17Customer requirements
- 48:49Subsystems and modularity
- 52:28Semiotics and signs
- 55:57Velocity and direction
- 59:25Meaning and consequences
- 1:01:39Subjective reality
- 1:06:15Multiple channels of data
- 1:10:54Business value estimates
- 1:16:03Full stack development
