<?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: Low defect density implies climate code less, not more, reliable</title>
	<atom:link href="http://shape-of-code.coding-guidelines.com/2012/12/24/low-defect-density-implies-climate-code-less-not-more-reliable/feed/" rel="self" type="application/rss+xml" />
	<link>http://shape-of-code.coding-guidelines.com/2012/12/24/low-defect-density-implies-climate-code-less-not-more-reliable/</link>
	<description></description>
	<lastBuildDate>Wed, 01 May 2013 08:59:49 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.4.1</generator>
	<item>
		<title>By: Derek Jones</title>
		<link>http://shape-of-code.coding-guidelines.com/2012/12/24/low-defect-density-implies-climate-code-less-not-more-reliable/comment-page-1/#comment-19663</link>
		<dc:creator>Derek Jones</dc:creator>
		<pubDate>Tue, 25 Dec 2012 11:26:21 +0000</pubDate>
		<guid isPermaLink="false">http://shape-of-code.coding-guidelines.com/?p=1223#comment-19663</guid>
		<description>&lt;a href=&quot;#comment-19662&quot; rel=&quot;nofollow&quot;&gt;@jasmine gray &lt;/a&gt; 
That would certainly show that the software worked for one dataset, but would the code work correctly when the data involves very different weather (e.g., a drought rather than a flood or vice versa).  The only reliable way is to break tings up into small parts for which it is possible to figure out the expected behavior and then run lots of tests on the parts, i.e., oversized unit testing</description>
		<content:encoded><![CDATA[<p><a href="#comment-19662" rel="nofollow">@jasmine gray </a><br />
That would certainly show that the software worked for one dataset, but would the code work correctly when the data involves very different weather (e.g., a drought rather than a flood or vice versa).  The only reliable way is to break tings up into small parts for which it is possible to figure out the expected behavior and then run lots of tests on the parts, i.e., oversized unit testing</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: jasmine gray</title>
		<link>http://shape-of-code.coding-guidelines.com/2012/12/24/low-defect-density-implies-climate-code-less-not-more-reliable/comment-page-1/#comment-19662</link>
		<dc:creator>jasmine gray</dc:creator>
		<pubDate>Tue, 25 Dec 2012 09:06:14 +0000</pubDate>
		<guid isPermaLink="false">http://shape-of-code.coding-guidelines.com/?p=1223#comment-19662</guid>
		<description>Of course, the real test of whether this software is reliable is whether it can actually predict changes in climate. E.g., if you plug in the data from 20 years ago, will it predict what the state of the system is today? If it can&#039;t do that, then no matter how faithfully the code implements the algorithm, the algorithm is useless.</description>
		<content:encoded><![CDATA[<p>Of course, the real test of whether this software is reliable is whether it can actually predict changes in climate. E.g., if you plug in the data from 20 years ago, will it predict what the state of the system is today? If it can&#8217;t do that, then no matter how faithfully the code implements the algorithm, the algorithm is useless.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
