3 ms·

I agree entirely with this approach. Especially when I consider that we're delivering a 350KiB minified Javascript hit on initial load just for jquery and a lo

• bkrausz 15y ago
  Why are you loading 10x times the jQuery filesize in just plugins? Certainly you have a lot more bloat to trim before you should consider eliminating jQuery entirely.
  • unicron 15y ago
    It's a very complicated application with a huge amount of HTTP endpoints. We're talking over a million lines of code.
• peregrine 15y ago
  I mean if you use the CDN jQuery the likelihood of loading jQuery up is fairly low considering its so ambiguous.
  • apsurd 15y ago
    you want the word ubiquitous https://www.google.com/search?ix=sea&sourceid=chrome&ie=UTF-8&q=define%3A+ubiquitous https://www.google.com/search?ix=sea&sourceid=chrome&#38...
    • peregrine 15y ago
      Thank you for this.
  • scarmig 15y ago
    Is this usually actually true? It makes a ton of obvious sense, but I've seen numbers thrown around that suggest a surprisingly low number of people actually end up taking advantage of the cache.
    • unicron 15y ago
      It's 100% not true. It's also a pain in the arse if you serve everything over SSL like we do.
      • joshuacc 15y ago
        To the best of my knowledge, the only reason SSL could cause a problem is if you are specifying the protocol (http:// http:// vs https:// https://). The best way to handle this is to use a protocol-relative URL like "//ajax.googleapis.com/mylibrary". The browser will then infer the correct protocol.
        • unicron 15y ago
          You've obviously never dealt with ancient IE versions and script loading then...
          • joshuacc 15y ago
            I'm happy to be enlightened if you feel like sharing.
• peregrine 15y ago
  I mean if you use the CDN jQuery the likelihood of loading jQuery up is fairly low considering its so ambiguous.