<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	>
<channel>
	<title>Comments on: The Metaphysics of Netflix</title>
	<atom:link href="http://www.digitalsolid.com/2006/12/27/the-metaphysics-of-netflix/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.digitalsolid.com/2006/12/27/the-metaphysics-of-netflix/</link>
	<description>Marketing Technology Musings and Tips by Jeff Larche</description>
	<pubDate>Wed, 20 Aug 2008 07:15:54 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.5.1</generator>
		<item>
		<title>By: Jeff Larche</title>
		<link>http://www.digitalsolid.com/2006/12/27/the-metaphysics-of-netflix/#comment-383</link>
		<dc:creator>Jeff Larche</dc:creator>
		<pubDate>Sun, 31 Dec 2006 15:30:54 +0000</pubDate>
		<guid isPermaLink="false">http://www.digitalsolid.com/2006/12/27/the-metaphysics-of-netflix/#comment-383</guid>
		<description>Thanks, Holly. Predicting tastes is going to get better over time as more people log their preferences, but I suspect other things have to help companies like Netflix. If I understand correctly how Pandora Internet Radio's prediction engine works, it has a music ontology built into it, where every song is categoriized behind the scenes, by rhythm, tempo, melody, etc. The thumbs up / thumbs down votes it gets from us help the engine fine-tune our tastes based on these elements, and ultimately help it recommend a new artist to us performing songs with everything we love and nothing we dislike. Amazing stuff.

NOTE TO READERS:
For those of you who long for music recommendations that have nothing to do with databases but everything to do with great musical criticism, check out Holly's wonderful music blog. She was too modest to include it as a link off of her name, above, so here is the URL: http://thesonginmyheadtoday.blogspot.com</description>
		<content:encoded><![CDATA[<p>Thanks, Holly. Predicting tastes is going to get better over time as more people log their preferences, but I suspect other things have to help companies like Netflix. If I understand correctly how Pandora Internet Radio&#8217;s prediction engine works, it has a music ontology built into it, where every song is categoriized behind the scenes, by rhythm, tempo, melody, etc. The thumbs up / thumbs down votes it gets from us help the engine fine-tune our tastes based on these elements, and ultimately help it recommend a new artist to us performing songs with everything we love and nothing we dislike. Amazing stuff.</p>
<p>NOTE TO READERS:<br />
For those of you who long for music recommendations that have nothing to do with databases but everything to do with great musical criticism, check out Holly&#8217;s wonderful music blog. She was too modest to include it as a link off of her name, above, so here is the URL: <a href="http://thesonginmyheadtoday.blogspot.com" rel="nofollow">http://thesonginmyheadtoday.blogspot.com</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Holly Hughes</title>
		<link>http://www.digitalsolid.com/2006/12/27/the-metaphysics-of-netflix/#comment-374</link>
		<dc:creator>Holly Hughes</dc:creator>
		<pubDate>Fri, 29 Dec 2006 05:38:47 +0000</pubDate>
		<guid isPermaLink="false">http://www.digitalsolid.com/2006/12/27/the-metaphysics-of-netflix/#comment-374</guid>
		<description>This is going to be more and more important in the future -- the only reason I still walk out to an independent bookstore or record store rather than just order product on line is because it's vital to have a human being recommend new things to me.  This goes way beyond word-of-mouth marketing; two human beings face to face can permutate an amazing number of taste corrections in just a few minutes to arrive at a match.  And the reason I like to use Netflix is because their recommendations are already fairly accurate. (In contrast to Amazon's, which suck.)  I wonder if they can use the sorcery responsible for Pandora Internet Radio in this regard -- I'm amazed by how quickly Pandora uses my "thumbs-up, thumbs-down" markers to refine what I want to listen to.
In theory I don't want these internet search engines to be as smart as human beings. But in practice I completely dig it when they are.</description>
		<content:encoded><![CDATA[<p>This is going to be more and more important in the future &#8212; the only reason I still walk out to an independent bookstore or record store rather than just order product on line is because it&#8217;s vital to have a human being recommend new things to me.  This goes way beyond word-of-mouth marketing; two human beings face to face can permutate an amazing number of taste corrections in just a few minutes to arrive at a match.  And the reason I like to use Netflix is because their recommendations are already fairly accurate. (In contrast to Amazon&#8217;s, which suck.)  I wonder if they can use the sorcery responsible for Pandora Internet Radio in this regard &#8212; I&#8217;m amazed by how quickly Pandora uses my &#8220;thumbs-up, thumbs-down&#8221; markers to refine what I want to listen to.<br />
In theory I don&#8217;t want these internet search engines to be as smart as human beings. But in practice I completely dig it when they are.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
