~/talk

To push, or not to push?! - The future of HTTP/2 server push

Talkat Full Stack Fest 2018 · on Thursday, 2018-09-06· [web-platform, web-performance]
slides:none on file

HTTP/2 server push promises one RTT savings but often overpushes because the server lacks client cache knowledge; real sites see mixed results and a Shopify case reports a 50% faster time-to-text (≈1.2s) from preload of fonts. Practical patterns are async push from an edge or CDN, or using a service worker to avoid repeat pushes. Upcoming fixes include cache digests to let servers know client caches and 103 Early Hints plus priority hints to let browsers fetch and prioritize earlier and smarter.

AI-written summary · about

  1. 0:00Start
  2. 5:05Critical Path
  3. 8:24The Preload Api
  4. 9:02Hidden Critical Sub Resources
  5. 10:22Render Tree
  6. 11:17Preload Api
  7. 14:20Traditional Request Flow
  8. 19:21Recap Server Push
  9. 27:40User Agent Sniffing on the Server
  10. 27:56The Rate of Adoption of Push Is Extremely Low
  11. 29:24Async Push Pattern
  12. 32:25103 Early Hint Status Code
  13. 38:30Hosting Static Assets with S3
~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.