
<?phpxml version="1.0" encoding="utf-8"?>
<rss version="2.0" 
xmlns:content="http://purl.org/rss/1.0/modules/content/"
xmlns:wfw="http://wellformedweb.org/CommentAPI/"
xmlns:dc="http://purl.org/dc/elements/1.1/"
>
<channel>
<title>Sphinn / maharg / Voted News</title>
<link>http://sphinn.com</link>
<description>Sphinn</description>
<pubDate>Mon, 28 Apr 2008 05:55:58 -0700</pubDate>
<language>en</language>
<item>
<title><![CDATA[Google Paper: Better Image Search Though VisualRank / Image Rank]]></title>
<link>http://sphinn.com/story/43011</link>
<comments>http://sphinn.com/story/43011</comments>
<pubDate>Mon, 28 Apr 2008 05:55:58 -0700</pubDate>
<dc:creator></dc:creator>
<category>Google Searching</category>
<guid>http://sphinn.com/story/43011</guid>
<description><![CDATA[So the idea is that somehow, by more closely being algorithmically related to more images, an image is more relevant? That seems odd, does anyone have a better handle on why that makes sense? If I make an image that not only looks like the white house, but also the taj mahal, my image will be more relevant? It seems to me that being well connected should still play a role<br/><br/>5 Vote(s) ]]></description>
</item>

</channel>
</rss>
