• Alex Russell
  • Software Engineer Google
  • Twitter
Alex is an engineer at on the Blink and Chrome teams at Google. From Service Workers to Web Components to service on TC39 (the standards committee for JavaScript) and the W3C Technical Architecture Group, he seeks through his work to improve the web as a platform for building compelling applications. He previously helped create Google Chrome Frame and, in a past life worked on large-scale JavaScript frameworks. He feels your pain.


  • Service Workers: Offline Now(ish)!
  • Time: 11:40am - 12:30pm | Room: Grand A

HTML Service Workers present new opportunities -- and challenges -- for mobile and offline application development on the web by offering the equivalent of a programmable client-side proxy for network traffic from your application.

This talk by the editor of the Service Worker spec explores:

  • The install process and the First Load Problem (TM)
  • The programmable cache model
    • Populating caches
    • Background updates
    • Matching cache entries to Requests
  • Request-by-request network control
    • What requests get set to which SW? What about iframes?
    • Cross-origin vs. Same-origin opportunities
    • What can be cached, where, how much, and for how long
    • Request routing
    • Composing SW-based apps with importScripts() and event handlers
  • The Offline-First application model
    • How apps differ from pages
    • Thinking about shell vs. content; what's in a URL?
  • The application upgrade process
    • wait-for-restart vs. immediate takeover
    • upgrade gotchas, version skew, and schema migrations
  • Performance considerations
    • Will using SWs slow my app down?
    • Network vs. disk performance
    • Request ordering and priority
    • Streaming response content
    • SPDY/HTTP2.0 and Service Workers
    • Memory concerns
    • Interaction with the browser preload-scanner
  • Future Service Worker-based APIs