The Global Baseline, 2022
summary
Argues for a conservative Global Baseline anchored to a mid‑range device like the Samsung Galaxy A50 (≈4 GB RAM, 4G) and shows measurement using BigQuery, HTTP Archive, OpenSignal, Geekbench and Web Almanac. Recommends logging interaction‑to‑next‑paint and time‑to‑first‑byte, and sets practical budgets: about 150k of HTML and CSS plus ~300–350k JS to aim for ~5s first‑load and ~3s repeat‑view, while triaging by connection type, device memory and CPU.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 1:05Web performance ecosystem challenges
- 3:31Web performance and society
- 6:212016 baseline guess impact
- 9:31Baseline evolution 2016-2021
- 13:08Factors shaping global baseline
- 15:17Browser diffusion and retirement
- 17:23Network data complexity
- 21:20Desktop performance and redlining
- 23:45Device form factors and market
- 27:39Desktop device lifespans
- 31:12iPhone not representative
- 33:18JavaScript payload growth
- 34:21CPU performance 2022
- 37:53Core Web Vitals pass rates
- 41:022022 global baseline: Galaxy A50
- 43:51Slow is exclusionary
- 45:55Applying global baseline
related talks
40mMy Journey: Adding Baseline Profiles to a Production-Based Android App: Success or Failure?
34mReal world connectivity: Beyond the bubble
31mBeyond the bubble
47mWhose web is it, anyway?
28mA Case Study
29mReal world connectivity: Beyond the bubble
42mBuilding quick apps in 2018
41mModern App Performance