content/marketing/marketing_bouncer_faq.html - openoffice-org - Git at Google

 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
     "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">

 <html xmlns="http://www.w3.org/1999/xhtml">
 <head>
   <meta http-equiv="content-type" content="text/html; charset=utf-8" />

   <title>FAQ - Recording downloads via MirrorBrain</title>
   <meta name="keywords" content="Marketing, OpenOffice, OpenOffice.org" />
   <link rel="stylesheet" type="text/css" href="marketing.css" />
 </head>

 <body>
   <div id="MyContent">
     <h1>FAQ - Recording downloads via MirrorBrain</h1>

     <h2>Why do we measure downloads?</h2>

     <p>One of the <a href="http://marketing.openoffice.org/">Marketing
     Project's</a> key statistics is the number of downloads from the
     OpenOffice.org site. This gives us immediate feedback on the success of
     promotional activities, such as <a href=
     "http://marketing.openoffice.org/press_releases.html">press releases</a>,
     as well as trend information over time.</p>

     <h2>What is MirrorBrain?</h2>

     <p>When people request a download from the <a href=
     "http://downloads.openoffice.org/">OpenOffice.org download page</a>, they
     are directed to <a href="http://mirrorbrain.org/">MirrorBrain</a>. MirrorBrain redirects them to their nearest
     mirror site for the download they require, and records the request in a
     log file.</p>

     <h2>Does MirrorBrain provide the full story?</h2>

     <p><strong>No!</strong> - it only records downloads started a single
     point, the <a href="http://downloads.openoffice.org/">OpenOffice.org
     download page</a>. It omits:</p>

     <ul>
       <li>downloads which people make directly from mirrors</li>

       <li>downloads via other mechanisms, such as <a href=
       "http://distribution.openoffice.org/p2p/">peer-to-peer
       networks</a></li>

       <li>downloads from other third-party repositories - including GNU/Linux
       distributions (see next question)</li>
     </ul>

     <p>Note also that MirrorBrain logs when it successfully redirects someone
     to a download site - if the user chooses not to download, or cancels the
     download, then MirrorBrain will not be aware.</p>

     <h2>Why do MirrorBrain figures look low for GNU/Linux?</h2>

     <p>The OpenOffice.org Community provides OpenOffice.org software in the
     two most common GNU/Linux package formats (rpm and deb). However, the
     majority of GNU/Linux users receive software direct from their
     distributor's respositories. This is a significant omission and explains
     the relatively low number of GNU/Linux downloads in our figures. How big
     is the gap? When we released OpenOffice.org 3.0, Ubuntu - possibly the
     largest desktop GNU/Linux variant - <a href=
     "http://www.theregister.co.uk/2008/10/27/shuttleworth_ubuntu_commitment/">
     estimated</a> they had around eight million users.</p>

     <h2>Are MirrorBrain downloads the same as the number of users ?</h2>

     <p>No. Nothing like it.</p>

     <ul>
       <li>MirrorBrain records only a percentage of total downloads (see
       above)</li>

       <li>Downloads are only one distribution method - people buy CD-ROMs, or
       get one from a magazine cover, or borrow one from friends...; people
       buy PCs with OpenOffice.org pre-installed... etc.</li>

       <li>One download or CD-ROM can be installed on thousands of computers -
       or none at all</li>
     </ul>

     <p>Note: the <a href="http://distribution.openoffice.org/">Distribution
     Project</a> exists to make OpenOffice.org available through as many
     channels as possible.</p>

     <h2>Do other people provide statistics?</h2>

     <p>Several of the <a href=
     "http://projects.openoffice.org/native-lang.html">Native-Language
     Projects</a> record download information for their languages.</p>
   </div>
 </body>
 </html>
	<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
	"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">

	<html xmlns="http://www.w3.org/1999/xhtml">
	<head>
	<meta http-equiv="content-type" content="text/html; charset=utf-8" />

	<title>FAQ - Recording downloads via MirrorBrain</title>
	<meta name="keywords" content="Marketing, OpenOffice, OpenOffice.org" />
	<link rel="stylesheet" type="text/css" href="marketing.css" />
	</head>

	<body>
	<div id="MyContent">
	<h1>FAQ - Recording downloads via MirrorBrain</h1>

	<h2>Why do we measure downloads?</h2>

	<p>One of the <a href="http://marketing.openoffice.org/">Marketing
	Project's</a> key statistics is the number of downloads from the
	OpenOffice.org site. This gives us immediate feedback on the success of
	promotional activities, such as <a href=
	"http://marketing.openoffice.org/press_releases.html">press releases</a>,
	as well as trend information over time.</p>

	<h2>What is MirrorBrain?</h2>

	<p>When people request a download from the <a href=
	"http://downloads.openoffice.org/">OpenOffice.org download page</a>, they
	are directed to <a href="http://mirrorbrain.org/">MirrorBrain</a>. MirrorBrain redirects them to their nearest
	mirror site for the download they require, and records the request in a
	log file.</p>

	<h2>Does MirrorBrain provide the full story?</h2>

	<p><strong>No!</strong> - it only records downloads started a single
	point, the <a href="http://downloads.openoffice.org/">OpenOffice.org
	download page</a>. It omits:</p>

	<ul>
	<li>downloads which people make directly from mirrors</li>

	<li>downloads via other mechanisms, such as <a href=
	"http://distribution.openoffice.org/p2p/">peer-to-peer
	networks</a></li>

	<li>downloads from other third-party repositories - including GNU/Linux
	distributions (see next question)</li>
	</ul>

	<p>Note also that MirrorBrain logs when it successfully redirects someone
	to a download site - if the user chooses not to download, or cancels the
	download, then MirrorBrain will not be aware.</p>

	<h2>Why do MirrorBrain figures look low for GNU/Linux?</h2>

	<p>The OpenOffice.org Community provides OpenOffice.org software in the
	two most common GNU/Linux package formats (rpm and deb). However, the
	majority of GNU/Linux users receive software direct from their
	distributor's respositories. This is a significant omission and explains
	the relatively low number of GNU/Linux downloads in our figures. How big
	is the gap? When we released OpenOffice.org 3.0, Ubuntu - possibly the
	largest desktop GNU/Linux variant - <a href=
	"http://www.theregister.co.uk/2008/10/27/shuttleworth_ubuntu_commitment/">
	estimated</a> they had around eight million users.</p>

	<h2>Are MirrorBrain downloads the same as the number of users ?</h2>

	<p>No. Nothing like it.</p>

	<ul>
	<li>MirrorBrain records only a percentage of total downloads (see
	above)</li>

	<li>Downloads are only one distribution method - people buy CD-ROMs, or
	get one from a magazine cover, or borrow one from friends...; people
	buy PCs with OpenOffice.org pre-installed... etc.</li>

	<li>One download or CD-ROM can be installed on thousands of computers -
	or none at all</li>
	</ul>

	<p>Note: the <a href="http://distribution.openoffice.org/">Distribution
	Project</a> exists to make OpenOffice.org available through as many
	channels as possible.</p>

	<h2>Do other people provide statistics?</h2>

	<p>Several of the <a href=
	"http://projects.openoffice.org/native-lang.html">Native-Language
	Projects</a> record download information for their languages.</p>
	</div>
	</body>
	</html>