<?xml version="1.0" encoding="utf-8"?><!-- generator="wordpress/2.0.4" -->
<rss version="2.0" 
	xmlns:content="http://purl.org/rss/1.0/modules/content/">
<channel>
	<title>Comments on: Blogs Are Event Driven - Websites Are Not</title>
	<link>http://blog.whoisireland.com/2005/03/19/blogs-are-event-driven-websites-are-not/</link>
	<description>Search Engines, Domains, Statistics and Analysis</description>
	<pubDate>Wed, 07 Jan 2009 09:47:00 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.0.4</generator>

	<item>
		<title>by: Cloud</title>
		<link>http://blog.whoisireland.com/2005/03/19/blogs-are-event-driven-websites-are-not/#comment-110</link>
		<pubDate>Fri, 15 Apr 2005 11:23:35 +0000</pubDate>
		<guid>http://blog.whoisireland.com/2005/03/19/blogs-are-event-driven-websites-are-not/#comment-110</guid>
					<description>Well POTB is well behind any link analysis techniques but it is a step beyond IrishBlogs.ie.  If more people used Technorati tags in their blogs, perhaps a classification of posts at least would be possible.</description>
		<content:encoded><![CDATA[<p>Well POTB is well behind any link analysis techniques but it is a step beyond IrishBlogs.ie.  If more people used Technorati tags in their blogs, perhaps a classification of posts at least would be possible.
</p>
]]></content:encoded>
				</item>
	<item>
		<title>by: John McCormac</title>
		<link>http://blog.whoisireland.com/2005/03/19/blogs-are-event-driven-websites-are-not/#comment-94</link>
		<pubDate>Tue, 29 Mar 2005 09:26:14 +0000</pubDate>
		<guid>http://blog.whoisireland.com/2005/03/19/blogs-are-event-driven-websites-are-not/#comment-94</guid>
					<description>The problem with trying to classify blogs is that a blog is going to have multiple classifications and a classification system would always be a step behind.

The update algorithm that I've been messing with works by detecting link cascades as a topic spreads between blogs. Apart from having three separate link structures, this is the most effective way of doing things from what I can see. Time to start looking at matrices mathematics again.</description>
		<content:encoded><![CDATA[<p>The problem with trying to classify blogs is that a blog is going to have multiple classifications and a classification system would always be a step behind.</p>
<p>The update algorithm that I&#8217;ve been messing with works by detecting link cascades as a topic spreads between blogs. Apart from having three separate link structures, this is the most effective way of doing things from what I can see. Time to start looking at matrices mathematics again.
</p>
]]></content:encoded>
				</item>
	<item>
		<title>by: ectoraige</title>
		<link>http://blog.whoisireland.com/2005/03/19/blogs-are-event-driven-websites-are-not/#comment-92</link>
		<pubDate>Fri, 25 Mar 2005 12:20:29 +0000</pubDate>
		<guid>http://blog.whoisireland.com/2005/03/19/blogs-are-event-driven-websites-are-not/#comment-92</guid>
					<description>To deal with 'b-logs which are event-driven, one could try to build sets of blogs which share similar interests. They would tend to update at around the same time, and they would link to a lot of the same sites. They would also respond to one another's blog entries.

Instead of then checking each individual blog for updates, you could just check a small percentage of the set members to determine whether you should spider all the members.

Of course, people rarely have a single passion, so there would be a lot of crossover: the challange is to find the right granularity.

Just a thought.</description>
		<content:encoded><![CDATA[<p>To deal with &#8216;b-logs which are event-driven, one could try to build sets of blogs which share similar interests. They would tend to update at around the same time, and they would link to a lot of the same sites. They would also respond to one another&#8217;s blog entries.</p>
<p>Instead of then checking each individual blog for updates, you could just check a small percentage of the set members to determine whether you should spider all the members.</p>
<p>Of course, people rarely have a single passion, so there would be a lot of crossover: the challange is to find the right granularity.</p>
<p>Just a thought.
</p>
]]></content:encoded>
				</item>
</channel>
</rss>
