Comments on: Debunking Google’s Internet Optimization Tips http://www.codexon.com/posts/debunking-googles-internet-optimization-tips Fri, 12 Mar 2010 01:46:05 +0000 hourly 1 By: Stefano J. Attardi http://www.codexon.com/posts/debunking-googles-internet-optimization-tips/comment-page-1#comment-185 Stefano J. Attardi Sat, 04 Jul 2009 05:25:59 +0000 http://www.codexon.com/?p=388#comment-185 Content-Length is required when using keep-alives (multiple requests over the same TCP connection). Without it, the browser has no way to know when the current response ends and another one starts. "In order to remain persistent, all messages on the connection MUST have a self-defined message length (i.e., one not defined by closure of the connection)". RFC 2616 section 8.1.2.1. Content-Length is required when using keep-alives (multiple requests over the same TCP connection). Without it, the browser has no way to know when the current response ends and another one starts.

“In order to remain persistent, all messages on the connection MUST have a self-defined message length (i.e., one not defined by closure of the connection)”. RFC 2616 section 8.1.2.1.

]]> By: links for 2009-06-29 | Yostivanich.com http://www.codexon.com/posts/debunking-googles-internet-optimization-tips/comment-page-1#comment-173 links for 2009-06-29 | Yostivanich.com Mon, 29 Jun 2009 14:03:25 +0000 http://www.codexon.com/?p=388#comment-173 [...] Debunking Google’s Internet Optimization Tips – Codexon "It appears that the doubt surrounding Google’s new guide is not unfounded. Several errors and half-truths were found. Google’s use of optional HTML tags could save some bandwidth, but for pages like google.com it doesn’t save that much compared to what could have been saved with the headers." Google wrong, what?? (tags: google performance html programming webdevelopment php) [...] [...] Debunking Google’s Internet Optimization Tips – Codexon "It appears that the doubt surrounding Google’s new guide is not unfounded. Several errors and half-truths were found. Google’s use of optional HTML tags could save some bandwidth, but for pages like google.com it doesn’t save that much compared to what could have been saved with the headers." Google wrong, what?? (tags: google performance html programming webdevelopment php) [...]

]]> By: Google’s “Let’s Make the Web Faster” Movement - Monday By Noon http://www.codexon.com/posts/debunking-googles-internet-optimization-tips/comment-page-1#comment-172 Google’s “Let’s Make the Web Faster” Movement - Monday By Noon Mon, 29 Jun 2009 12:05:05 +0000 http://www.codexon.com/?p=388#comment-172 [...] the information you got may not be bulletproof, even if Google wrote it. Articles attempting to Debunk Google’s Internet Optimization Tips are beginning to crop up, and I think it’s great. Google suggests a number of techniques, [...] [...] the information you got may not be bulletproof, even if Google wrote it. Articles attempting to Debunk Google’s Internet Optimization Tips are beginning to crop up, and I think it’s great. Google suggests a number of techniques, [...]

]]> By: K http://www.codexon.com/posts/debunking-googles-internet-optimization-tips/comment-page-1#comment-170 K Mon, 29 Jun 2009 07:03:07 +0000 http://www.codexon.com/?p=388#comment-170 Response Headers: Cache-Control[private, max-age=0] Date[Mon, 29 Jun 2009 07:00:22 GMT] Expires[-1] Content-Type[text/html; charset=UTF-8] Content-Encoding[gzip] Server[gws] Content-Length[3364] X-Cache[MISS from .] Via[1.0 .:80 (squid)] Connection[keep-alive] I got them for Google. I'm not behind any proxies and my ISP doesn't add any. BTW, the request was HTTP 1.1 Response Headers:
Cache-Control[private, max-age=0]
Date[Mon, 29 Jun 2009 07:00:22 GMT]
Expires[-1]
Content-Type[text/html; charset=UTF-8]
Content-Encoding[gzip]
Server[gws]
Content-Length[3364]
X-Cache[MISS from .]
Via[1.0 .:80 (squid)]
Connection[keep-alive]

I got them for Google. I’m not behind any proxies and my ISP doesn’t add any. BTW, the request was HTTP 1.1

]]> By: | Cafyn http://www.codexon.com/posts/debunking-googles-internet-optimization-tips/comment-page-1#comment-169 | Cafyn Mon, 29 Jun 2009 04:48:23 +0000 http://www.codexon.com/?p=388#comment-169 [...] both articles are good read. Codexon also wrote regarding Google’s odd optimization techniques. [...] [...] both articles are good read. Codexon also wrote regarding Google’s odd optimization techniques. [...]

]]> By: Marlin http://www.codexon.com/posts/debunking-googles-internet-optimization-tips/comment-page-1#comment-163 Marlin Sun, 28 Jun 2009 21:41:53 +0000 http://www.codexon.com/?p=388#comment-163 These are general tips too, every "tip" can be different depending on the case involved. Different scenarios yield different results. Always do your own tests in your own enviornment. http://www.dealpi.com These are general tips too, every “tip” can be different depending on the case involved. Different scenarios yield different results.

Always do your own tests in your own enviornment.

http://www.dealpi.com

]]> By: atomi http://www.codexon.com/posts/debunking-googles-internet-optimization-tips/comment-page-1#comment-161 atomi Sun, 28 Jun 2009 21:21:11 +0000 http://www.codexon.com/?p=388#comment-161 oh lol hey Doug. Fancy seeing you here. BTW I agree with op. Have a great day buddy! oh lol hey Doug. Fancy seeing you here. BTW I agree with op. Have a great day buddy!

]]> By: admin http://www.codexon.com/posts/debunking-googles-internet-optimization-tips/comment-page-1#comment-160 admin Sun, 28 Jun 2009 19:40:17 +0000 http://www.codexon.com/?p=388#comment-160 I actually read that before I wrote this article. Date is actually not "required" for servers that don't have a reliable clock. Following this rule seems pedantic to me. Content-length is unnecessary beyond progress bars and is in fact, neglected in case of streaming because you don't know the potential size of the file. Also if you set the content-length too small, many browsers will cut off the transfer. I actually read that before I wrote this article.

Date is actually not “required” for servers that don’t have a reliable clock. Following this rule seems pedantic to me.

Content-length is unnecessary beyond progress bars and is in fact, neglected in case of streaming because you don’t know the potential size of the file. Also if you set the content-length too small, many browsers will cut off the transfer.

]]> By: admin http://www.codexon.com/posts/debunking-googles-internet-optimization-tips/comment-page-1#comment-159 admin Sun, 28 Jun 2009 19:36:56 +0000 http://www.codexon.com/?p=388#comment-159 According to RFC 2616, I don't see any section where it says the Server header is mandatory. http://www.w3.org/Protocols/rfc2616/rfc2616-sec4.html According to RFC 2616, I don’t see any section where it says the Server header is mandatory.

http://www.w3.org/Protocols/rfc2616/rfc2616-sec4.html

]]> By: admin http://www.codexon.com/posts/debunking-googles-internet-optimization-tips/comment-page-1#comment-158 admin Sun, 28 Jun 2009 19:29:02 +0000 http://www.codexon.com/?p=388#comment-158 I found the problem. You guys are using HTTP1.0 requests instead of 1.1. That probably also explains why you are getting spam errors. You're probably running an old or limited browser. I found the problem. You guys are using HTTP1.0 requests instead of 1.1.

That probably also explains why you are getting spam errors. You’re probably running an old or limited browser.

]]>