<?xml version="1.0" encoding="utf-8"?><!-- generator="wordpress/2.1" -->
<rss version="2.0" 
	xmlns:content="http://purl.org/rss/1.0/modules/content/">
<channel>
	<title>Comments on: Dave Friend fudging the numbers</title>
	<link>http://www.unodemocrats.com/blog/2005/05/01/dave-friend-fudging-the-numbers/</link>
	<description></description>
	<pubDate>Thu, 21 Aug 2008 22:39:01 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.1</generator>

	<item>
		<title>By: Amanda</title>
		<link>http://www.unodemocrats.com/blog/2005/05/01/dave-friend-fudging-the-numbers/#comment-1636</link>
		<author>Amanda</author>
		<pubDate>Wed, 04 May 2005 02:50:58 +0000</pubDate>
		<guid>http://www.unodemocrats.com/blog/2005/05/01/dave-friend-fudging-the-numbers/#comment-1636</guid>
					<description>What a challenge chief!  You must be pretty sure to bet $250 on it.</description>
		<content:encoded><![CDATA[<p>What a challenge chief!  You must be pretty sure to bet $250 on it.</p>
]]></content:encoded>
				</item>
	<item>
		<title>By: Brian</title>
		<link>http://www.unodemocrats.com/blog/2005/05/01/dave-friend-fudging-the-numbers/#comment-1913</link>
		<author>Brian</author>
		<pubDate>Tue, 17 May 2005 16:13:35 +0000</pubDate>
		<guid>http://www.unodemocrats.com/blog/2005/05/01/dave-friend-fudging-the-numbers/#comment-1913</guid>
					<description>Actually, your account of a "hit' is a little deceptive these days too, with the common use of the HTTP/1.1 protocol versus HTTP/1.0.  In 1.0 you would be right, for each item (e.g. image) on a webpage, a separate request or 'hit' would be made by the client browser (e.g. Firefox) and then be logged by the webserver.  With 1.1 however, a single TCP connection is made and all items on the website are downloaded in 1 request or 'hit'.  This makes pages load faster too, since the time delay in setting up another connection for each new request is not wasted.

Still, a physical person will likely make more than visit to just the main page of the website, they will likely click on other links, read the blog, post a comment, etc... and all of these each add up to another hit to the website.  So again, it's not representative of the "number of users" (which Friend never said it was, but rather 'hits') - but really, there is no good measure for that...

Webservers can also count the different IP addresses that visit the website, but many ISPs give out 'fake' IP addresses to many dialup users which are simply hidden behind 1 real IP.  So, if every user in this scenario visited the website, the server would only record 1 unique IP address - which is also not accurate of the number of unique "visitors" to visit the site either.

So- while you consider Friend's statement about "hits" deceptive (i.e. "fudging the numbers"), I consider your response to be slightly as well.  He used the proper termonology, referring to the numbers only as "hits". It is very good of you though to explain what a "hit", as most people are not computer geeks like us who understand the intricacies of the Internet. =)

When given a statistic or number, always question what it REALLY means. 

[1] http://getfirefox.com &lt;a href="http://getfirefox.com" rel="nofollow"&gt;IE Sucks&lt;/a&gt;</description>
		<content:encoded><![CDATA[<p>Actually, your account of a &#8220;hit&#8217; is a little deceptive these days too, with the common use of the HTTP/1.1 protocol versus HTTP/1.0.  In 1.0 you would be right, for each item (e.g. image) on a webpage, a separate request or &#8216;hit&#8217; would be made by the client browser (e.g. Firefox) and then be logged by the webserver.  With 1.1 however, a single TCP connection is made and all items on the website are downloaded in 1 request or &#8216;hit&#8217;.  This makes pages load faster too, since the time delay in setting up another connection for each new request is not wasted.</p>
<p>Still, a physical person will likely make more than visit to just the main page of the website, they will likely click on other links, read the blog, post a comment, etc&#8230; and all of these each add up to another hit to the website.  So again, it&#8217;s not representative of the &#8220;number of users&#8221; (which Friend never said it was, but rather &#8216;hits&#8217;) - but really, there is no good measure for that&#8230;</p>
<p>Webservers can also count the different IP addresses that visit the website, but many ISPs give out &#8216;fake&#8217; IP addresses to many dialup users which are simply hidden behind 1 real IP.  So, if every user in this scenario visited the website, the server would only record 1 unique IP address - which is also not accurate of the number of unique &#8220;visitors&#8221; to visit the site either.</p>
<p>So- while you consider Friend&#8217;s statement about &#8220;hits&#8221; deceptive (i.e. &#8220;fudging the numbers&#8221;), I consider your response to be slightly as well.  He used the proper termonology, referring to the numbers only as &#8220;hits&#8221;. It is very good of you though to explain what a &#8220;hit&#8221;, as most people are not computer geeks like us who understand the intricacies of the Internet. =)</p>
<p>When given a statistic or number, always question what it REALLY means. </p>
<p>[1] <a href="http://getfirefox.com" rel="nofollow">http://getfirefox.com</a> <a href="http://getfirefox.com" rel="nofollow">IE Sucks</a></p>
]]></content:encoded>
				</item>
</channel>
</rss>
