/me found an interesting article about HTTP Compression (gzip), this jgwebber guy really writes some cool articles for Web Developers, so even if it’s an old article I find it highly relevant even today
jgwebber “If you’ve dealt with HTTP much, you’ve probably at least heard that it supports gzip compression. And under some circumstances, this even turns out to be true! You might think that supporting something as simple as decoding gzip-encoded content would be simple and straightforward, and you’d be right (especially given that gzip code has been available freely since, well, about forever).
It seems, however, that people working on most major browsers at various points found ways to make this difficult. It would be one thing (albeit a silly thing) if they simply didn’t support gzip encoding, but it’s another matter entirely that a number of browsers request encoded content which they then proceed to barf on. There are lots of articles out there describing, in varying degrees of detail, the mess that is HTTP compression. IBM has a good one.”
Read the full article:
* as simple as possible, but no simpler: The Insanity of HTTP Compression