~/talk

FaCSSt—CSS and Performance

Talkat Render Conference 2018 · on Friday, 2018-03-23· [css, web-performance]

CSS parsing and delivery actually decide perceived speed: a 1 MB stylesheet costs about 200 ms parse on a Moto G4, embedding images with base64 blows up critical CSS and delays first paint (measured as roughly 700 ms desktop / 7 s mobile in one experiment), and using @import serializes requests — flattening imports cuts start-render from 2.5s to 0.5s in a case study. Mixins gzip better than extends, asset subdomains add DNS/TCP/TLS overhead for critical resources, and loading order matters because CSS blocks inline script execution.

AI-written summary · about

  1. 0:00Start
  2. 3:31Larger files take longer to download...
  3. 6:09Base64 is an anti-pattern
  4. 6:30An Experiment
  5. 8:34@import
  6. 11:26Selector Performance
  7. 12:47Key Selector
  8. 18:52Alphabetic CSS
  9. 20:51Gzip loves repetition and consistency
  10. 21:01Mixin vs. @extend
  11. 22:52That was a very non-scientific demo
  12. 24:22Asset Domains
  13. 26:45Inline scripts block on CSSOM construction
  14. 28:23Load the JS with the async attribute
~/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.