Small Is Beautiful
summary
Argues that smaller, constrained systems are easier to reason about, maintain and discard, citing concrete counterexamples: a ZX81 chess program that fits in 672 bytes, a boot sector chess in 512 bytes, and the Unix kernel implemented in C at under 10,000 lines versus modern bloat like an iOS app with 18,000 classes and 429 contributors a week. Shows tools used (Z80 assembler, Perl, C), points to failures from zombie code (a $460M trading loss) and to organizational causes (Conway’s law, diseconomies of scale).
AI-written summary · about
chapters (18)
- 0:00Defining legacy systems
- 2:00Sustainability in development
- 4:40Balancing scale and design
- 6:51Human cognition and constraints
- 10:20Constraint-driven creativity
- 14:12Unix as a model of simplicity
- 18:11Working with constraints
- 20:34The human factor in code layout
- 23:32Measuring technical debt
- 26:12Discardability and innovation
- 31:33Diseconomies of scale
- 34:48Organizational structures
- 39:54Avoiding the alignment trap
- 43:01Quadrants of effectiveness
- 49:00The overhead of team size
- 52:19Self-fulfilling complexity
- 55:42Dark matter in legacy systems
- 58:04Conclusion: Detrimental development
related talks
52mWorse Is Better, for Better or for Worse
58mWorse Is Better, for Better or for Worse
01h 02mWorse Is Better, for Better or for Worse
32mKeeping your code strong
26mThe Power of Simplicity: Crafting Durable Software Systems
56mThe Best Simple System for Now
01h 22mSimplicity: not just for beginners
50mSoftware Is Details