<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/2.2" -->
<rss version="2.0" 
	xmlns:content="http://purl.org/rss/1.0/modules/content/">
<channel>
	<title>Comments on: Test Code is as important as Production Code</title>
	<link>http://www.vinktank.com/test-automation/test-code-is-as-important-as-production-code/</link>
	<description>Pragmatic software testing &#038; development by Kristan Vingrys</description>
	<pubDate>Sat, 04 Sep 2010 16:52:30 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.2</generator>

	<item>
		<title>By: Igor Mironov</title>
		<link>http://www.vinktank.com/test-automation/test-code-is-as-important-as-production-code/#comment-14</link>
		<author>Igor Mironov</author>
		<pubDate>Thu, 09 Aug 2007 12:03:37 +0000</pubDate>
		<guid>http://www.vinktank.com/test-automation/test-code-is-as-important-as-production-code/#comment-14</guid>
		<description>At the conference you also made a very good point about the need to test the tests (by injecting failures). Perhaps this should be done as part of the CI process so that there is some guarantee that the test suite is still relevant.

Another approach that someone mentioned was to include test code changes in peer reviews. If nothing else, this should discourage the sticky-tape programming style.</description>
		<content:encoded><![CDATA[<p>At the conference you also made a very good point about the need to test the tests (by injecting failures). Perhaps this should be done as part of the CI process so that there is some guarantee that the test suite is still relevant.</p>
<p>Another approach that someone mentioned was to include test code changes in peer reviews. If nothing else, this should discourage the sticky-tape programming style.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Tomas Varsavsky</title>
		<link>http://www.vinktank.com/test-automation/test-code-is-as-important-as-production-code/#comment-10</link>
		<author>Tomas Varsavsky</author>
		<pubDate>Mon, 30 Jul 2007 23:35:54 +0000</pubDate>
		<guid>http://www.vinktank.com/test-automation/test-code-is-as-important-as-production-code/#comment-10</guid>
		<description>I agree with you Kristan. Unit and integration testing is often the bulk of the code that I write on projects. The sheer volume of tests makes problems with test code a real pain point of the project. Maintenance of tests is arguably harder than production code. As soon as we allow ourselves some leniency on test code quality we're on a slippery slope destined for disaster.

With tools like Selenium RC there is really no excuse to write badly factored acceptance tests (http://tomasvarsavsky.com/2007/07/27/selenium-page-models/). What I struggle with is the role of the tester in the team. Traditionally testers are not developers. Testers have been exposed to code through recording test tools but don't generally have the design skills that are at the core of a good developer. Is it fair to expect testers to write well factored acceptance tests? Are we at a point where we need a new generation of testers that are code savvy? I think this generation of testers is well on it's way, we just need more of them in the market!</description>
		<content:encoded><![CDATA[<p>I agree with you Kristan. Unit and integration testing is often the bulk of the code that I write on projects. The sheer volume of tests makes problems with test code a real pain point of the project. Maintenance of tests is arguably harder than production code. As soon as we allow ourselves some leniency on test code quality we&#8217;re on a slippery slope destined for disaster.</p>
<p>With tools like Selenium RC there is really no excuse to write badly factored acceptance tests (http://tomasvarsavsky.com/2007/07/27/selenium-page-models/). What I struggle with is the role of the tester in the team. Traditionally testers are not developers. Testers have been exposed to code through recording test tools but don&#8217;t generally have the design skills that are at the core of a good developer. Is it fair to expect testers to write well factored acceptance tests? Are we at a point where we need a new generation of testers that are code savvy? I think this generation of testers is well on it&#8217;s way, we just need more of them in the market!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Nirav Thaker</title>
		<link>http://www.vinktank.com/test-automation/test-code-is-as-important-as-production-code/#comment-9</link>
		<author>Nirav Thaker</author>
		<pubDate>Mon, 30 Jul 2007 13:58:51 +0000</pubDate>
		<guid>http://www.vinktank.com/test-automation/test-code-is-as-important-as-production-code/#comment-9</guid>
		<description>I agree testing code is important, but not sure If I should treat it as production code, and if you want to apply costly SCM processes to test code, it will slowdown entire maintanance cycle, wouldn't it? IMO, Test code is better not treated as sensitively in some situation like maintainance but should be open for modification.</description>
		<content:encoded><![CDATA[<p>I agree testing code is important, but not sure If I should treat it as production code, and if you want to apply costly SCM processes to test code, it will slowdown entire maintanance cycle, wouldn&#8217;t it? IMO, Test code is better not treated as sensitively in some situation like maintainance but should be open for modification.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
