<?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"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: Chainlove alerts</title>
	<atom:link href="http://uphpu.org/2008/10/12/chainlove-alerts/feed/" rel="self" type="application/rss+xml" />
	<link>http://uphpu.org/2008/10/12/chainlove-alerts/</link>
	<description>PHP application development and support</description>
	<lastBuildDate>Tue, 29 Nov 2011 03:17:09 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.0.5</generator>
	<item>
		<title>By: wade</title>
		<link>http://uphpu.org/2008/10/12/chainlove-alerts/comment-page-1/#comment-20349</link>
		<dc:creator>wade</dc:creator>
		<pubDate>Mon, 20 Oct 2008 00:08:21 +0000</pubDate>
		<guid isPermaLink="false">http://uphpu.org/2008/10/12/chainlove-alerts/#comment-20349</guid>
		<description>I hadn&#039;t thought of storing the data anything longer than the cache file. Tossing it in a SQLite database is a cool idea though. It would be very interesting to learn more about their system. For example, they do not always run a product until it is sold out; they appear to run it only as long as it&#039;s hot. As soon as units stop moving at a certain pace, they swap it out with something else. Although I don&#039;t know how automated it is. These products will often appear again later. I wonder if you would be able to start predicting patterns with enough data.</description>
		<content:encoded><![CDATA[<p>I hadn&#8217;t thought of storing the data anything longer than the cache file. Tossing it in a SQLite database is a cool idea though. It would be very interesting to learn more about their system. For example, they do not always run a product until it is sold out; they appear to run it only as long as it&#8217;s hot. As soon as units stop moving at a certain pace, they swap it out with something else. Although I don&#8217;t know how automated it is. These products will often appear again later. I wonder if you would be able to start predicting patterns with enough data.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: thebigdog</title>
		<link>http://uphpu.org/2008/10/12/chainlove-alerts/comment-page-1/#comment-20348</link>
		<dc:creator>thebigdog</dc:creator>
		<pubDate>Sun, 19 Oct 2008 23:38:39 +0000</pubDate>
		<guid isPermaLink="false">http://uphpu.org/2008/10/12/chainlove-alerts/#comment-20348</guid>
		<description>nice job wade. have you thought about storing the data in a sqlite db so you can do some data analysis on it? like how often they actually change the data.</description>
		<content:encoded><![CDATA[<p>nice job wade. have you thought about storing the data in a sqlite db so you can do some data analysis on it? like how often they actually change the data.</p>
]]></content:encoded>
	</item>
</channel>
</rss>

