Teach Your Bundler The Users' Habits
summary
Describes a webpack-based pipeline that uses Google Analytics navigation graphs, a static parser and a clustering step to produce ml plugin driven chunking and a cluster chunks plugin plus runtime prefetcher. Claims a Markov chain model with a 0.5 probability threshold triggers prefetches, demoed on a Brownie Points clone where merging main and settings yields instant load and FAQ is prefetched, reducing perceived wait under poor network. Includes an NPM/webpack integration and adaptive behavior by network speed and user clusters.
AI-written summary · about
chapters (8)
- 0:00Intro
- 4:56Picking Split Points
- 6:36Navigation Graph
- 11:28ML Plugin
- 12:44Google Analytics Package
- 14:38Webpack Plugin
- 15:00Cluster Checks Plugin
- 17:41Future Plans
related talks
36mUnbundling the JavaScript module bundler
34mHow to extend webpack with your own plugins
28mWebpack for a better development workflow
17mBundling your Front-End with webpack
44mPredictive Prefetching
32mWhat's New in Webpack 4
01h 42mEverything is a plugin! Mastering webpack from the inside out
20mWebpack: Your New BFF