Scaling Up!
summary
Argues that scaling software is about reducing cognitive load, not bigger production lines: advocates red green refactor (TDD) and shows a fake-engine test that forces modular, loosely coupled code; defines a deployment pipeline as “commit to releasable” with no extra sign-offs; cites a meta-analysis of ~4,000 projects where teams of 20 beat teams of 5 by only one week over nine months but produce ~5× more defects, and contrasts network math 0.85 vs 1.15 for hierarchical vs distributed scaling.
AI-written summary · about
chapters (12)
- 0:00Start
- 0:10Scaling Up!
- 3:26Optimise for Learning
- 4:47Optimise to Manage Complexity
- 6:28What do We Need to Do?
- 7:34Principles
- 12:11What Do We Value In Good Code?
- 16:31What Drives Quality? - Before TDD
- 19:21What Makes Code Testable?
- 25:06What Does "Deployable" Mean?
- 27:21What Does 'Scaling' Mean?
- 28:14Small Teams
related talks
01h 02mTools for simplified large scale systems
38mThe Dark Side of Scale
35mThe elements of scaling
49mHypergrowth scaling made simple
42mLarge-Scale Architecture: The Unreasonable Effectiveness of Simplicity
44mThe Art of Architecting for Scale
59mBig Scrum: Scaling Scrum with Nexus
49mScaling compiled applications