A brief history and mishistory of modularity
summary
Argues that modularity’s value is practical not ideological: npm shows thousands of packages with ~224 million downloads a day, the left-pad incident exposes fragile transitive deps, and browserify can spend ~400ms walking the module tree, making many tiny modules hurt runtime. Recommends decomposing around likely-to-change decisions, isolating hard parts, preferring disposable bits over premature reuse, and measuring actual costs (downloads, dependency depth, build and runtime time) rather than prescribing smallness as a virtue.
AI-written summary · about
chapters (5)
- 0:00Start
- 3:38how do you decide what goes in a module?
- 8:25npm is the LARGEST MODULE REPOSITORY IN THE WORLD.
- 16:36what is information?
- 25:06how to computer?
related talks
39mA brief history and mishistory of modularity
29mA Brief History of Modularity
19mModularity Matters
01mMark Richards at DDD Europe
41mThe future is modules not frameworks. JS modules Architecture demo.
36mUnbundling the JavaScript module bundler
48mMicroservices? Monoliths? Modules!
35mKnowing Your TCO (Total Cost of Ownership)
