<?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 for d3orn.ch</title>
	<atom:link href="http://www.d3orn.ch/comments/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.d3orn.ch</link>
	<description></description>
	<lastBuildDate>Mon, 13 Sep 2010 21:11:23 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.0.4</generator>
	<item>
		<title>Comment on Fooling Around by zombiecalypse</title>
		<link>http://www.d3orn.ch/2010/fooling-around/comment-page-1/#comment-1944</link>
		<dc:creator>zombiecalypse</dc:creator>
		<pubDate>Mon, 13 Sep 2010 21:11:23 +0000</pubDate>
		<guid isPermaLink="false">http://www.d3orn.ch/?p=1027#comment-1944</guid>
		<description>&quot;I don&#039;t care how it works as long as it works&quot; doesn&#039;t mean that you shouldn&#039;t learn the idioms and paradigms of the language as long as you can work without them. It is more about the stuff behind the scenes. 

A good example is Python: Every object is an instance of a class, and classes are objects, and you can use metaclasses that initialize a class object. But you don&#039;t need that to define 99.99% of the classes you&#039;ll ever encounter.
So you don&#039;t have to care, how class creation works.
At first you don&#039;t need to know, that the for-loop takes an iterable and not simply a sequence or what an iterable even is (hint: it is not a sequence). 
So until you run into a problem, that uses iterables other than lists, don&#039;t care  how for-loops really work, but be open to find out and read some example code, if you&#039;re facing difficulties.

But yes, you won&#039;t master a language just by fooling around. But that&#039;s not the point: You will most certainly run into problems which can not be solved by brute force and so you take a look at a reference, which will then slowly introduce you to the more difficult parts. And your natural curiosity will lead your way from there. It took me quite long to find a problem, that had an application for metaclasses, but I wanted to find out, how I could work with it. But that&#039;s a different topic.

One important part, that wasn&#039;t directly mentioned in this article is, that you have to know the paradigms and the fortes of the language you are learning before. So you&#039;d be prepared to write object oriented if you fool around with Java or ready to wrap your head around higher order functions, if you try Clojure - and be prepared for hell, if you use Haskell. So compiling C with a C++ compiler doesn&#039;t use the forte of C++, which, if anything, are classes.

Maybe you have a point, that it should be mentioned, that you are supposed to find out, what the canonical way to solve a problem in your language is - defining an datatype in Haskell, creating smart objects in C++,... - to prevent porting the style of your old language to your new one.</description>
		<content:encoded><![CDATA[<p>&#8220;I don&#8217;t care how it works as long as it works&#8221; doesn&#8217;t mean that you shouldn&#8217;t learn the idioms and paradigms of the language as long as you can work without them. It is more about the stuff behind the scenes. </p>
<p>A good example is Python: Every object is an instance of a class, and classes are objects, and you can use metaclasses that initialize a class object. But you don&#8217;t need that to define 99.99% of the classes you&#8217;ll ever encounter.<br />
So you don&#8217;t have to care, how class creation works.<br />
At first you don&#8217;t need to know, that the for-loop takes an iterable and not simply a sequence or what an iterable even is (hint: it is not a sequence).<br />
So until you run into a problem, that uses iterables other than lists, don&#8217;t care  how for-loops really work, but be open to find out and read some example code, if you&#8217;re facing difficulties.</p>
<p>But yes, you won&#8217;t master a language just by fooling around. But that&#8217;s not the point: You will most certainly run into problems which can not be solved by brute force and so you take a look at a reference, which will then slowly introduce you to the more difficult parts. And your natural curiosity will lead your way from there. It took me quite long to find a problem, that had an application for metaclasses, but I wanted to find out, how I could work with it. But that&#8217;s a different topic.</p>
<p>One important part, that wasn&#8217;t directly mentioned in this article is, that you have to know the paradigms and the fortes of the language you are learning before. So you&#8217;d be prepared to write object oriented if you fool around with Java or ready to wrap your head around higher order functions, if you try Clojure &#8211; and be prepared for hell, if you use Haskell. So compiling C with a C++ compiler doesn&#8217;t use the forte of C++, which, if anything, are classes.</p>
<p>Maybe you have a point, that it should be mentioned, that you are supposed to find out, what the canonical way to solve a problem in your language is &#8211; defining an datatype in Haskell, creating smart objects in C++,&#8230; &#8211; to prevent porting the style of your old language to your new one.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Fooling Around by Jonas</title>
		<link>http://www.d3orn.ch/2010/fooling-around/comment-page-1/#comment-1939</link>
		<dc:creator>Jonas</dc:creator>
		<pubDate>Thu, 09 Sep 2010 18:44:12 +0000</pubDate>
		<guid isPermaLink="false">http://www.d3orn.ch/?p=1027#comment-1939</guid>
		<description>I can&#039;t agree with you in this point: &quot;I don’t care how it works as long as it works!&quot;

If you learn a new language, you shoudn&#039;t try to understand everything, that&#039;s right. But you should understand the principles of the language, or at least the principles of the paradigms, the language supports. Of course, you can write a Java code without object oriented design. But that&#039;s most likely waste of time and you&#039;ll never learn what objects really are! You can write a C code, compile it with the C++ compiler, and say that you&#039;re learning a new language -  but that&#039;s just absurd. You could learn a lot of languages like that, but you&#039;ll never master one, I think.

With your concept, you can learn the syntax of a language, that&#039;s true. But the syntax is mostly the smallest and most easy part to learn! (As long as you aren&#039;t programming in an esoteric language of course...)

You&#039;d like to go for a beer sometime?</description>
		<content:encoded><![CDATA[<p>I can&#8217;t agree with you in this point: &#8220;I don’t care how it works as long as it works!&#8221;</p>
<p>If you learn a new language, you shoudn&#8217;t try to understand everything, that&#8217;s right. But you should understand the principles of the language, or at least the principles of the paradigms, the language supports. Of course, you can write a Java code without object oriented design. But that&#8217;s most likely waste of time and you&#8217;ll never learn what objects really are! You can write a C code, compile it with the C++ compiler, and say that you&#8217;re learning a new language &#8211;  but that&#8217;s just absurd. You could learn a lot of languages like that, but you&#8217;ll never master one, I think.</p>
<p>With your concept, you can learn the syntax of a language, that&#8217;s true. But the syntax is mostly the smallest and most easy part to learn! (As long as you aren&#8217;t programming in an esoteric language of course&#8230;)</p>
<p>You&#8217;d like to go for a beer sometime?</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Some Thoughts by Autarkis</title>
		<link>http://www.d3orn.ch/2010/some-thoughts/comment-page-1/#comment-1930</link>
		<dc:creator>Autarkis</dc:creator>
		<pubDate>Sat, 28 Aug 2010 08:03:22 +0000</pubDate>
		<guid isPermaLink="false">http://www.swissmtg.ch/blog/?p=985#comment-1930</guid>
		<description>More but smaller = good. Looking forward to it.</description>
		<content:encoded><![CDATA[<p>More but smaller = good. Looking forward to it.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Some Thoughts by d3orn</title>
		<link>http://www.d3orn.ch/2010/some-thoughts/comment-page-1/#comment-1929</link>
		<dc:creator>d3orn</dc:creator>
		<pubDate>Fri, 27 Aug 2010 23:11:08 +0000</pubDate>
		<guid isPermaLink="false">http://www.swissmtg.ch/blog/?p=985#comment-1929</guid>
		<description>Thx Adrian for the nice comment I always like some motivation^^
Yeah we had a nice little conversation with David today and I think we should def. meet with you next week when you&#039;re back. Niko is away for two weeks.
We dicided not to make SCodeBubbles as our bachelor&#039;s but instead it will be a free time and probably Google Summer of Code project (if possible) 
But like I said let&#039;s meet next week or so. 
I also love regular readers and I&#039;m very happy about the increasing numbers of daily readers.
Oh I already fall for Smalltalk and Pharo it&#039;s so nice to work with it. 
cheers</description>
		<content:encoded><![CDATA[<p>Thx Adrian for the nice comment I always like some motivation^^<br />
Yeah we had a nice little conversation with David today and I think we should def. meet with you next week when you&#8217;re back. Niko is away for two weeks.<br />
We dicided not to make SCodeBubbles as our bachelor&#8217;s but instead it will be a free time and probably Google Summer of Code project (if possible)<br />
But like I said let&#8217;s meet next week or so.<br />
I also love regular readers and I&#8217;m very happy about the increasing numbers of daily readers.<br />
Oh I already fall for Smalltalk and Pharo it&#8217;s so nice to work with it.<br />
cheers</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Some Thoughts by Adrian</title>
		<link>http://www.d3orn.ch/2010/some-thoughts/comment-page-1/#comment-1928</link>
		<dc:creator>Adrian</dc:creator>
		<pubDate>Fri, 27 Aug 2010 21:54:57 +0000</pubDate>
		<guid isPermaLink="false">http://www.swissmtg.ch/blog/?p=985#comment-1928</guid>
		<description>I love your blog and happy to hear to you are going to increase the frequency of new posts. Shorter posts are a good thing too, they are quicker to read and more likely to be of use for readers. As a regular reader of your blog, I am happy to read articles of your&#039;s even if they are one or two paragraphs. What matters to me is that they are refreshing and novel content! As a random reader of your blog, I am typically brought here by a search engine and clearly prefer short articles that help to solve my current problem or task that hand. My personal rule for blog posts is that everything I&#039;d email as answer to a friend&#039;s question possibly qualifies as a good blog post. So I&#039;d be more than happy to see your blog going as you work on SCodeBubbles, certainly you&#039;ll solve many Smalltalk problems that are worth sharing.</description>
		<content:encoded><![CDATA[<p>I love your blog and happy to hear to you are going to increase the frequency of new posts. Shorter posts are a good thing too, they are quicker to read and more likely to be of use for readers. As a regular reader of your blog, I am happy to read articles of your&#8217;s even if they are one or two paragraphs. What matters to me is that they are refreshing and novel content! As a random reader of your blog, I am typically brought here by a search engine and clearly prefer short articles that help to solve my current problem or task that hand. My personal rule for blog posts is that everything I&#8217;d email as answer to a friend&#8217;s question possibly qualifies as a good blog post. So I&#8217;d be more than happy to see your blog going as you work on SCodeBubbles, certainly you&#8217;ll solve many Smalltalk problems that are worth sharing.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on How to Learn a New Language by d3orn</title>
		<link>http://www.d3orn.ch/2010/how-to-learn-a-new-language/comment-page-1/#comment-1925</link>
		<dc:creator>d3orn</dc:creator>
		<pubDate>Sun, 22 Aug 2010 18:08:24 +0000</pubDate>
		<guid isPermaLink="false">http://www.swissmtg.ch/blog/?p=921#comment-1925</guid>
		<description>Great article! 
There is no doubt that fooling around is one of the best ways to learn something but I like to read a whole book first and while reading I code the examples. After I&#039;m done reading I like to go the same way you do.
Learning by doing is really one of the best ways to learn ^^</description>
		<content:encoded><![CDATA[<p>Great article!<br />
There is no doubt that fooling around is one of the best ways to learn something but I like to read a whole book first and while reading I code the examples. After I&#8217;m done reading I like to go the same way you do.<br />
Learning by doing is really one of the best ways to learn ^^</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Evolutionary Ideas! by holomorphic</title>
		<link>http://www.d3orn.ch/2010/evolutionary-ideas/comment-page-1/#comment-1586</link>
		<dc:creator>holomorphic</dc:creator>
		<pubDate>Wed, 16 Jun 2010 18:28:26 +0000</pubDate>
		<guid isPermaLink="false">http://www.swissmtg.ch/blog/?p=718#comment-1586</guid>
		<description>Genetic algorithms are mostly popular because of their name. 

Calling it &#039;stochastic beam search with a combination function&#039; (the crossover) is not sexy enough, I guess :)

(Russel,Norvig:AIMA FTW)</description>
		<content:encoded><![CDATA[<p>Genetic algorithms are mostly popular because of their name. </p>
<p>Calling it &#8216;stochastic beam search with a combination function&#8217; (the crossover) is not sexy enough, I guess <img src='http://www.d3orn.ch/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /> </p>
<p>(Russel,Norvig:AIMA FTW)</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Need YOUR help! by zombiecalypse</title>
		<link>http://www.d3orn.ch/2010/need-your-help/comment-page-1/#comment-1579</link>
		<dc:creator>zombiecalypse</dc:creator>
		<pubDate>Mon, 14 Jun 2010 11:22:36 +0000</pubDate>
		<guid isPermaLink="false">http://www.swissmtg.ch/blog/?p=661#comment-1579</guid>
		<description>monochrome is my favorite, because it&#039;s not as monochrome as Piano Black and has a rather sexy web-typographic look</description>
		<content:encoded><![CDATA[<p>monochrome is my favorite, because it&#8217;s not as monochrome as Piano Black and has a rather sexy web-typographic look</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Maturaarbeit 2.0 by Dominique Rahm</title>
		<link>http://www.d3orn.ch/2010/maturaarbeit-2-0/comment-page-1/#comment-1373</link>
		<dc:creator>Dominique Rahm</dc:creator>
		<pubDate>Thu, 22 Apr 2010 15:53:47 +0000</pubDate>
		<guid isPermaLink="false">http://www.swissmtg.ch/blog/?p=526#comment-1373</guid>
		<description>jop thx for the inputs, im my current version the vectorstuff and the missing constructor is allready implemented. 
Yeah class design is the hardest part of a programm^^ especially responsibility-driven design</description>
		<content:encoded><![CDATA[<p>jop thx for the inputs, im my current version the vectorstuff and the missing constructor is allready implemented.<br />
Yeah class design is the hardest part of a programm^^ especially responsibility-driven design</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Maturaarbeit 2.0 by Jonas</title>
		<link>http://www.d3orn.ch/2010/maturaarbeit-2-0/comment-page-1/#comment-1372</link>
		<dc:creator>Jonas</dc:creator>
		<pubDate>Thu, 22 Apr 2010 14:17:29 +0000</pubDate>
		<guid isPermaLink="false">http://www.swissmtg.ch/blog/?p=526#comment-1372</guid>
		<description>Just some inputs:
gravitationConstant could be static. It&#039;s the same for all instances.
Where has the constructor gone? Or how do you set e.g. the weight?
Wouldn&#039;t mass be a better name than weight?
You could use an extra point (or vector) object instead of just two coords (look at java.awt.Point). This would give you the possibility to simply add some methods for translations, vector operations or even a third dimension. Getting the X coord could look like this: planet.getPosition().getX();
I think velocity is more occurate than speed.
Velocity (for you Speed^^) is a vector quantity and should be represented as such (as I said, an extra vector object would be great [the body _has a_ velocity]).
Last but not least: What about the sun (or the moon)? I think an object called Body or the like would do a better job. The sun _is a_ Body with velocity 0, but never a Planet! You even said it by yourself: &quot;have one Object for all the Bodies&quot;!  

Think again about your class design... it&#039;s a hard job!</description>
		<content:encoded><![CDATA[<p>Just some inputs:<br />
gravitationConstant could be static. It&#8217;s the same for all instances.<br />
Where has the constructor gone? Or how do you set e.g. the weight?<br />
Wouldn&#8217;t mass be a better name than weight?<br />
You could use an extra point (or vector) object instead of just two coords (look at java.awt.Point). This would give you the possibility to simply add some methods for translations, vector operations or even a third dimension. Getting the X coord could look like this: planet.getPosition().getX();<br />
I think velocity is more occurate than speed.<br />
Velocity (for you Speed^^) is a vector quantity and should be represented as such (as I said, an extra vector object would be great [the body _has a_ velocity]).<br />
Last but not least: What about the sun (or the moon)? I think an object called Body or the like would do a better job. The sun _is a_ Body with velocity 0, but never a Planet! You even said it by yourself: &#8220;have one Object for all the Bodies&#8221;!  </p>
<p>Think again about your class design&#8230; it&#8217;s a hard job!</p>
]]></content:encoded>
	</item>
</channel>
</rss>

