…loading, please wait
summary
Argues for flipping to a data-first architecture where a router fetches before rendering, allowing fetches to run in a shared worker/service worker/edge and classify requests as primary or secondary — dubbed server-side data (SSD) no render. Describes building and forking occlusion libraries (smoke and mirrors → vertical-collection) that still hurt perceived performance, cites vertical-collection adoption at ~20% of Ember source, and warns about apps struggling with hundreds of thousands of records and leaf-level complexity.
AI-written summary · about
chapters (15)
- 0:00Introduction and thanks
- 1:27Missteps in web app evolution
- 2:29Rendering wars story time
- 5:17Performance paradox and leaf complexity
- 7:45Pagination and waterfall problems
- 10:17Data fetching and rendering flaws
- 12:49Server side data concept
- 14:14Rendering wrong perspective
- 15:17Abstraction vs leaf complexity
- 19:09Derived state at the leaves
- 23:00Root complexity and legacy
- 24:06Meta programming explained
- 27:15Feature development backwards
- 30:02System abstraction benefits
- 31:04Q&A and closing
related talks
20mA framework for building data driven interactive front ends
20mOpen Source Discoverability: the journey of unearthing an abundance of hidden data
33mTemplate-first Data Visualizations
51mFront End Data
30mFlux for Data Visualization: A Preliminary Manifesto
26mA duck for your dashboard: Performant data apps in the browser with DuckDB
38mThinking DSLs for Massive Visualization
52mAngular Loves Data