JavaScript in Browser Performance
summary
This talk is from 2009 jsconf.eu. Front end dominates load time: with empty cache backend is often ~9% while browsers underutilize networks (target ~80%). Measured tools like YSlow and PageSpeed and WebPageTest expose that sites download ~250KB of JavaScript but only execute ~26% by onload. Proven fixes include deferred or parallel script loading (script DOM element, XHR+eval, chained async), gzip and server accelerators; experiments show 400ms delays cut searches 0.6–9%, Shopzilla’s speedups lift revenue 7–12%, Netflix cuts bandwidth 43% with compression.
AI-written summary · about
chapters (34)
- 0:00Intro
- 0:55The web has given us a beautiful world
- 1:54Sharing information and experiences
- 3:04Practice multiple disciplines
- 4:06Footnotes
- 4:37About me
- 5:47Turning it upside down
- 8:14The long pole
- 8:42What can we do
- 9:27What makes you feel slow
- 11:12What makes sites feel slow
- 12:34WebpageTest
- 13:57News
- 15:28Progressive Enhancement
- 17:24Why JavaScript
- 17:48Test Case
- 19:51How to Make JavaScript Less Painful
- 21:27Splitting the Initial Payload
- 23:13xhreval
- 23:53Scriptonelement
- 25:30Gmail Mobile
- 27:50Charles Jolly
- 29:31Why Slow and PageSpeed
- 30:54Im the Performance Cop
- 33:26More Charts
- 36:45Wax
- 37:32Strange Loop Networks
- 38:14Programming Environments
- 39:14SpriteMe
- 42:00Browser Scope
- 43:59HTTP Archive Format
- 45:15Webinars
- 46:10Takeaways
- 47:05Questions
related talks
41mMake JavaScript Faster
21mState of Performance
28mWhat we talk about when we say "web performances"
37mSecret Shortcuts of Loading Web Performance
36mPerformance for the long haul
31mThe Future of Front-End Performance
01h 03mSpeed up your Web Application
21mWhy are web applications slow, and what to do about it