Ajax from A to X
summary
Measured experiments improve page responsiveness for AJAX search results by reducing “time from click to first snippet.” Techniques include predicting the URL and pre-sending the request to cut median latency by ~100 ms, fast DOM insertion of the first two snippets to deliver the largest improvement (up to ~350–400 ms at the 95th percentile), and experimenting with streaming (multipart → custom chunked JSON) which gave smaller gains. Implementation relied on profiling, server-side chunk logging, canary A/B tests, minimal client changes, and automated tests to validate before rollout.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 1:05No universal solution
- 2:29Ajax in search scenario
- 4:15Ajax adoption history
- 6:24Optimizing page load
- 7:48Ajax metrics problems
- 9:16Choosing the right metric
- 10:58Measurement accuracy matters
- 12:43First optimization idea
- 13:47Optimizing first snippet rendering
- 14:29Optimizing request send time
- 16:57Fast send experiment
- 19:22Fast insert experiment
- 21:05Progressive download experiment
- 26:43Network-based animation control
- 28:51Optimization results summary
- 31:19Optimization tips
- 31:40Q&A
related talks
48mJavaScript in Browser Performance
58mJavaScript for accessible web applications
28mJavaScript SEO (The "Right" Way)
46m10 secrets to improve Javascript Performance
18mThe need for speed - Single-page apps, Optimistic UIs, Reactivity
38mSecret Shortcuts of Loading Web Performance
16mNeed for Speed: How Modern JavaScript Frameworks Are Solving Web Performance
26mOptimize Your JavaScript App's Performance