<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/2.3.3" -->
<rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	>
<channel>
	<title>Comments for Adam Byrtek</title>
	<link>http://www.adambyrtek.net</link>
	<description>Sceptical Point of View</description>
	<pubDate>Thu, 07 Aug 2008 20:37:44 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.3.3</generator>
		<item>
		<title>Comment on Why it&#8217;s good to be lazy by Adam Byrtek - Functional programming in Python</title>
		<link>http://www.adambyrtek.net/2008/04/24/why-its-good-to-be-lazy/#comment-908</link>
		<dc:creator>Adam Byrtek - Functional programming in Python</dc:creator>
		<pubDate>Thu, 24 Apr 2008 06:30:58 +0000</pubDate>
		<guid>http://www.adambyrtek.net/2008/04/24/why-its-good-to-be-lazy/#comment-908</guid>
		<description>[...] Update 2008-04-24: Slides are also available in English. [...]</description>
		<content:encoded><![CDATA[<p>[&#8230;] Update 2008-04-24: Slides are also available in English. [&#8230;]</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Trivial accessors and uniform access by Marcin Kaszyński</title>
		<link>http://www.adambyrtek.net/2008/02/09/trivial-accessors-and-uniform-access/#comment-333</link>
		<dc:creator>Marcin Kaszyński</dc:creator>
		<pubDate>Sun, 30 Mar 2008 18:12:00 +0000</pubDate>
		<guid>http://www.adambyrtek.net/2008/02/09/trivial-accessors-and-uniform-access/#comment-333</guid>
		<description>There is a small bug in the Python version of Money: both getter and setter should use self._amount instead of self.amount.

What are the other problems with this approach?</description>
		<content:encoded><![CDATA[<p>There is a small bug in the Python version of Money: both getter and setter should use self._amount instead of self.amount.</p>
<p>What are the other problems with this approach?</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Benchmarking HTTP performance by Adam Byrtek - Thinner Ruby deployment</title>
		<link>http://www.adambyrtek.net/2008/03/05/benchmarking-http-performance/#comment-117</link>
		<dc:creator>Adam Byrtek - Thinner Ruby deployment</dc:creator>
		<pubDate>Mon, 10 Mar 2008 19:08:07 +0000</pubDate>
		<guid>http://www.adambyrtek.net/2008/03/05/benchmarking-http-performance/#comment-117</guid>
		<description>[...] the post on benchmarking HTTP performance I mentioned that according to my tests a cluster of Mongrels performs about 10-20% worse than the [...]</description>
		<content:encoded><![CDATA[<p>[&#8230;] the post on benchmarking HTTP performance I mentioned that according to my tests a cluster of Mongrels performs about 10-20% worse than the [&#8230;]</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Subversion Scripts for Finder by Adam Byrtek</title>
		<link>http://www.adambyrtek.net/2008/02/24/subversion-scripts-for-finder/#comment-38</link>
		<dc:creator>Adam Byrtek</dc:creator>
		<pubDate>Wed, 27 Feb 2008 22:45:20 +0000</pubDate>
		<guid>http://www.adambyrtek.net/2008/02/24/subversion-scripts-for-finder/#comment-38</guid>
		<description>Thanks for your comments! I considered migrating my personal repositories to Git, but I'm quite happy with my current setup, so it's still on my someday/maybe list. Anyway I still have to work with existing Subversion repositories, and Subversion is much more popular comparing to other version control systems, so it was a reasonable decision to prepare scripts supporting it.</description>
		<content:encoded><![CDATA[<p>Thanks for your comments! I considered migrating my personal repositories to Git, but I&#8217;m quite happy with my current setup, so it&#8217;s still on my someday/maybe list. Anyway I still have to work with existing Subversion repositories, and Subversion is much more popular comparing to other version control systems, so it was a reasonable decision to prepare scripts supporting it.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Functional programming in Python by Rafał Zawadzki</title>
		<link>http://www.adambyrtek.net/2008/01/30/functional-programming-in-python/#comment-37</link>
		<dc:creator>Rafał Zawadzki</dc:creator>
		<pubDate>Wed, 27 Feb 2008 21:59:44 +0000</pubDate>
		<guid>http://www.adambyrtek.net/2008/01/30/functional-programming-in-python/#comment-37</guid>
		<description>Ups - no comments, so - this one will be first ;)</description>
		<content:encoded><![CDATA[<p>Ups - no comments, so - this one will be first ;)</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Subversion Scripts for Finder by Jakub T</title>
		<link>http://www.adambyrtek.net/2008/02/24/subversion-scripts-for-finder/#comment-34</link>
		<dc:creator>Jakub T</dc:creator>
		<pubDate>Tue, 26 Feb 2008 09:49:34 +0000</pubDate>
		<guid>http://www.adambyrtek.net/2008/02/24/subversion-scripts-for-finder/#comment-34</guid>
		<description>Since i started using git i started using version control for everything i do. Even if you don't need dedicated server in subversion for your repository you need to first import and checkout your project that is a big PITA. In git everything is simple you just enter the work directory at the moment you decide that it's important to have it versioned and you issue 'git init'. Starting from that moment everything is versioned. 
What is even more interesting - git is transparent and if you use subversion as your official vcs you can still use git locally to track all your small modifications and branching. Later on, using git extended options you will be able to commit all the changes to your central subversion repository. This makes the transition really smooth.

@Tomek:
I don't know what's the difference for end users if something is written in python/ruby/c....?</description>
		<content:encoded><![CDATA[<p>Since i started using git i started using version control for everything i do. Even if you don&#8217;t need dedicated server in subversion for your repository you need to first import and checkout your project that is a big PITA. In git everything is simple you just enter the work directory at the moment you decide that it&#8217;s important to have it versioned and you issue &#8216;git init&#8217;. Starting from that moment everything is versioned.<br />
What is even more interesting - git is transparent and if you use subversion as your official vcs you can still use git locally to track all your small modifications and branching. Later on, using git extended options you will be able to commit all the changes to your central subversion repository. This makes the transition really smooth.</p>
<p>@Tomek:<br />
I don&#8217;t know what&#8217;s the difference for end users if something is written in python/ruby/c&#8230;.?</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Subversion Scripts for Finder by Adam Byrtek</title>
		<link>http://www.adambyrtek.net/2008/02/24/subversion-scripts-for-finder/#comment-33</link>
		<dc:creator>Adam Byrtek</dc:creator>
		<pubDate>Mon, 25 Feb 2008 20:57:23 +0000</pubDate>
		<guid>http://www.adambyrtek.net/2008/02/24/subversion-scripts-for-finder/#comment-33</guid>
		<description>@Tomek: Subversion also doesn't need a dedicated server, it can work directly from a filesystem or through a ssh tunnel. I don't claim that it is a better option, but certainly more popular, and there are many tools and plugins that support Subversion.</description>
		<content:encoded><![CDATA[<p>@Tomek: Subversion also doesn&#8217;t need a dedicated server, it can work directly from a filesystem or through a ssh tunnel. I don&#8217;t claim that it is a better option, but certainly more popular, and there are many tools and plugins that support Subversion.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Subversion Scripts for Finder by Tomek Paczkowski</title>
		<link>http://www.adambyrtek.net/2008/02/24/subversion-scripts-for-finder/#comment-32</link>
		<dc:creator>Tomek Paczkowski</dc:creator>
		<pubDate>Mon, 25 Feb 2008 20:49:32 +0000</pubDate>
		<guid>http://www.adambyrtek.net/2008/02/24/subversion-scripts-for-finder/#comment-32</guid>
		<description>I would argue about personal use of subversion, for that task I find bazaar better. It's so simple to make it just work for version control. You don't need to create separate repository just for keeping your files and you don't need to have separate server for colaboration or storing files on external machine. Bazaar can work via ssh, ftp and others, including svn repo. It's easier for me to use it, seems more intuitive. Did I mention it's also written in Python?</description>
		<content:encoded><![CDATA[<p>I would argue about personal use of subversion, for that task I find bazaar better. It&#8217;s so simple to make it just work for version control. You don&#8217;t need to create separate repository just for keeping your files and you don&#8217;t need to have separate server for colaboration or storing files on external machine. Bazaar can work via ssh, ftp and others, including svn repo. It&#8217;s easier for me to use it, seems more intuitive. Did I mention it&#8217;s also written in Python?</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Trivial accessors and uniform access by Tomek Paczkowski</title>
		<link>http://www.adambyrtek.net/2008/02/09/trivial-accessors-and-uniform-access/#comment-7</link>
		<dc:creator>Tomek Paczkowski</dc:creator>
		<pubDate>Sat, 09 Feb 2008 19:39:29 +0000</pubDate>
		<guid>http://www.adambyrtek.net/2008/02/09/trivial-accessors-and-uniform-access/#comment-7</guid>
		<description>There is a good article about differences between Java and Python: &lt;a href="http://dirtsimple.org/2004/12/python-is-not-java.html" rel="nofollow"&gt; Python Is Not Java&lt;/a&gt; by Phillip J. Eby. It's surely reading-worth for Java programmers wanting to learn Python.</description>
		<content:encoded><![CDATA[<p>There is a good article about differences between Java and Python: <a href="http://dirtsimple.org/2004/12/python-is-not-java.html" rel="nofollow"> Python Is Not Java</a> by Phillip J. Eby. It&#8217;s surely reading-worth for Java programmers wanting to learn Python.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Trivial accessors and uniform access by Bartosz Bankowski</title>
		<link>http://www.adambyrtek.net/2008/02/09/trivial-accessors-and-uniform-access/#comment-6</link>
		<dc:creator>Bartosz Bankowski</dc:creator>
		<pubDate>Sat, 09 Feb 2008 19:07:05 +0000</pubDate>
		<guid>http://www.adambyrtek.net/2008/02/09/trivial-accessors-and-uniform-access/#comment-6</guid>
		<description>It was an interesting article for me, as I didn't know about the UAP before. However, I don't agree with your assumptions:

&#62; But in Java you have to introduce getters 
&#62; and setters from the very beginning, or it 
&#62; will bite you back in the future. It 
&#62; clearly contradicts with the DRY principle 
&#62; and a preference for evolutionary design, 
&#62; which discourages writing code that is 
&#62; useless right now, but may (or may not) 
&#62; be needed in the future.

You say that I have to create getters and setters from the very beginning, while
this is not true. I never do it. The very first time I create getter or setter is when I need to use it from my code.

Described behavior would not only be against the DRY rule, but also could lead to broken encapsulation.

&#62; The whole problem boils down to the fact 
&#62; that in Java you can’t apply the Uniform 
&#62; Access Principle

Which problem? From my point of view, the problem is not in a language, but in a design.</description>
		<content:encoded><![CDATA[<p>It was an interesting article for me, as I didn&#8217;t know about the UAP before. However, I don&#8217;t agree with your assumptions:</p>
<p>&gt; But in Java you have to introduce getters<br />
&gt; and setters from the very beginning, or it<br />
&gt; will bite you back in the future. It<br />
&gt; clearly contradicts with the DRY principle<br />
&gt; and a preference for evolutionary design,<br />
&gt; which discourages writing code that is<br />
&gt; useless right now, but may (or may not)<br />
&gt; be needed in the future.</p>
<p>You say that I have to create getters and setters from the very beginning, while<br />
this is not true. I never do it. The very first time I create getter or setter is when I need to use it from my code.</p>
<p>Described behavior would not only be against the DRY rule, but also could lead to broken encapsulation.</p>
<p>&gt; The whole problem boils down to the fact<br />
&gt; that in Java you can’t apply the Uniform<br />
&gt; Access Principle</p>
<p>Which problem? From my point of view, the problem is not in a language, but in a design.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
