<?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: JUnitFactory: First Impressions</title>
	<atom:link href="http://blog.technoetic.com/2007/04/08/junitfactory-first-impressions/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.technoetic.com/2007/04/08/junitfactory-first-impressions/</link>
	<description></description>
	<lastBuildDate>Thu, 29 Apr 2010 08:58:57 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.0</generator>
	<item>
		<title>By: Jörg Thönnes</title>
		<link>http://blog.technoetic.com/2007/04/08/junitfactory-first-impressions/comment-page-1/#comment-5849</link>
		<dc:creator>Jörg Thönnes</dc:creator>
		<pubDate>Wed, 18 Jul 2007 06:04:10 +0000</pubDate>
		<guid isPermaLink="false">http://blog.technoetic.com/2007/04/08/junitfactory-first-impressions/#comment-5849</guid>
		<description>Hi Steve,

just found this review. I know Agitar from a very interesting presentation at the JAX-W conference in Munich, Germany in November 2006.

I found the concept appealing and a interesting complementation to the normal way to write unit tests. From this experience I support Albertos statement of &quot;test amplification.&quot;

On the other hand, I see your points of being to dependent on the implementation of the class. Could facades help here?

In summary, I did not try any Agitar products yet, but would love to do so if I find some time. Until then, my view stays a theoretical one...

Cheers, Jörg</description>
		<content:encoded><![CDATA[<p>Hi Steve,</p>
<p>just found this review. I know Agitar from a very interesting presentation at the JAX-W conference in Munich, Germany in November 2006.</p>
<p>I found the concept appealing and a interesting complementation to the normal way to write unit tests. From this experience I support Albertos statement of &#8220;test amplification.&#8221;</p>
<p>On the other hand, I see your points of being to dependent on the implementation of the class. Could facades help here?</p>
<p>In summary, I did not try any Agitar products yet, but would love to do so if I find some time. Until then, my view stays a theoretical one&#8230;</p>
<p>Cheers, Jörg</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Steve Bate</title>
		<link>http://blog.technoetic.com/2007/04/08/junitfactory-first-impressions/comment-page-1/#comment-3704</link>
		<dc:creator>Steve Bate</dc:creator>
		<pubDate>Wed, 11 Apr 2007 00:13:47 +0000</pubDate>
		<guid isPermaLink="false">http://blog.technoetic.com/2007/04/08/junitfactory-first-impressions/#comment-3704</guid>
		<description>Hello Alberto and Kevin,

Thanks for the comments. The way JUnitFactory &quot;studies&quot; the code and creates tests is very interesting. I can understand why it would be a fun product to work on.</description>
		<content:encoded><![CDATA[<p>Hello Alberto and Kevin,</p>
<p>Thanks for the comments. The way JUnitFactory &#8220;studies&#8221; the code and creates tests is very interesting. I can understand why it would be a fun product to work on.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Kevin Lawrence</title>
		<link>http://blog.technoetic.com/2007/04/08/junitfactory-first-impressions/comment-page-1/#comment-3690</link>
		<dc:creator>Kevin Lawrence</dc:creator>
		<pubDate>Tue, 10 Apr 2007 16:08:22 +0000</pubDate>
		<guid isPermaLink="false">http://blog.technoetic.com/2007/04/08/junitfactory-first-impressions/#comment-3690</guid>
		<description>Great review, thanks Steve. Feedback on JUnit Factory is what we need most of all.

The low coverage is probably related to the security exceptions. JUF makes tests that will be safe to run on your client so file deletions and socket connections to random ports are disallowed.

Kevin</description>
		<content:encoded><![CDATA[<p>Great review, thanks Steve. Feedback on JUnit Factory is what we need most of all.</p>
<p>The low coverage is probably related to the security exceptions. JUF makes tests that will be safe to run on your client so file deletions and socket connections to random ports are disallowed.</p>
<p>Kevin</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Alberto Savoia</title>
		<link>http://blog.technoetic.com/2007/04/08/junitfactory-first-impressions/comment-page-1/#comment-3661</link>
		<dc:creator>Alberto Savoia</dc:creator>
		<pubDate>Sun, 08 Apr 2007 19:07:06 +0000</pubDate>
		<guid isPermaLink="false">http://blog.technoetic.com/2007/04/08/junitfactory-first-impressions/#comment-3661</guid>
		<description>Hi Steve,

Thank you for a very well written and objective review.

Making JUnit Factory freely available is a lot of fun and a very useful learning experience (both for us and the users :-)).

One of the challenges for us is to deciding the right set of defaults.  We are trying as hard as we can to not have a bunch of options to detract from the basic simplicity of JUnit Factory.

One of the things we keep finding is that JUF rewards good coding behavior.  I have a slogan at Agitar:
Ugly Code --&gt; Ugly Tests.  

Of course, we cannot yet claim: Beautiful Code --&gt; Beautiful Tests (as in beautiful code *guarantees* beautiful tests), but we are working on it ... one code pattern and test heuristic at a time.

I also keep stressing the importance of looking at JUnit Factory and automated test generations in general as: test *amplification* tools.  We can do *some* magic if all you do is press a button, but the real payoff is when you write, say, 40-50 lines of test helpers (test data helpers, set-up helpers, and assertion helpers) and in return you get 400-500 lines of really good tests.

As far as uploading the entire project goes, that&#039;s the inherited behavior from our commercial product AgitarOne.  The benefit is that you pay a one-time &quot;penalty&quot; the first time you use it on a project but, after that, you only upload incremental changes and it&#039;s very fast.  We are looking into having a more incremental loading approach in the future.

As you reminded your readers, JUnit Factory is still very experimental, so keep experimenting and let us know how we can make it more useful.

Thanks again for taking the time to write a thoughtful review.

Alberto</description>
		<content:encoded><![CDATA[<p>Hi Steve,</p>
<p>Thank you for a very well written and objective review.</p>
<p>Making JUnit Factory freely available is a lot of fun and a very useful learning experience (both for us and the users <img src='http://blog.technoetic.com/wp-includes/images/smilies/icon_smile.gif' alt=':-)' class='wp-smiley' /> ).</p>
<p>One of the challenges for us is to deciding the right set of defaults.  We are trying as hard as we can to not have a bunch of options to detract from the basic simplicity of JUnit Factory.</p>
<p>One of the things we keep finding is that JUF rewards good coding behavior.  I have a slogan at Agitar:<br />
Ugly Code &#8211;&gt; Ugly Tests.  </p>
<p>Of course, we cannot yet claim: Beautiful Code &#8211;&gt; Beautiful Tests (as in beautiful code *guarantees* beautiful tests), but we are working on it &#8230; one code pattern and test heuristic at a time.</p>
<p>I also keep stressing the importance of looking at JUnit Factory and automated test generations in general as: test *amplification* tools.  We can do *some* magic if all you do is press a button, but the real payoff is when you write, say, 40-50 lines of test helpers (test data helpers, set-up helpers, and assertion helpers) and in return you get 400-500 lines of really good tests.</p>
<p>As far as uploading the entire project goes, that&#8217;s the inherited behavior from our commercial product AgitarOne.  The benefit is that you pay a one-time &#8220;penalty&#8221; the first time you use it on a project but, after that, you only upload incremental changes and it&#8217;s very fast.  We are looking into having a more incremental loading approach in the future.</p>
<p>As you reminded your readers, JUnit Factory is still very experimental, so keep experimenting and let us know how we can make it more useful.</p>
<p>Thanks again for taking the time to write a thoughtful review.</p>
<p>Alberto</p>
]]></content:encoded>
	</item>
</channel>
</rss>
