Good idea; I haven't done any latency studies yet, but that's a great next step...
The really interesting part would be to compute latencies with minification/gzip/both either on-the-fly (e.g., mod_pagespeed) or done in advance, and to figure out what the smallest network latency is for which the on-the-fly minification is still a net win.
Comments
Good idea; I haven't done any latency studies yet, but that's a great next step...
The really interesting part would be to compute latencies with minification/gzip/both either on-the-fly (e.g., mod_pagespeed) or done in advance, and to figure out what the smallest network latency is for which the on-the-fly minification is still a net win.
Interesting. Any studies on the total loading times with/without gzip, similar to this one?: http://ejohn.org/blog/library-loading-speed/