From: Lloyd Linklater <lloyd@2live4.com>
Date: 2007-12-14T21:30:17+09:00
Subject: Re: Newbie Question: What is a class for?

All the other guys said lots of cool and useful technical stuff.  Let me 
use history to give some of the ideology.  In the procedural days, 
programs became unmanagable when they got over 1/4 million lines or so, 
though 1/2 million line programs were not unheard of.

Well, the notion of how to bunch things into blocks did for software 
what the chip did for hardware.  The software version of a chip is the 
class.  You make a class that has some characteristics foundational to 
its descendants.  You program in the things that cover all the basics. 
Then, you add things as you get increasingly specific.

e.g.  You want a class for dogs and cats.  Well, they are fairly 
different so what do they have in common?

class mammal
# stuff
end

class four_legged < mammal
# stuff
end

class cat < four_legged
# stuff
end

class dog < four_legged
# stuff
end

Note that if you add things in the base class, mammal in this case, it 
trickles down to all descendants.  This is VERY handy in many ways.

By coding this in pieces, you can more easily keep it all in your head 
and, therefore, write far more complex programs with far more 
simplicity.  There are, of course, things that lend itself to art rather 
than mere technique in class design, but writing classes so that they 
are easily used is a good thing to know.  Consider Ruby's approach.  You 
can do something.to_s all over the place.  That takes planning.  Not TOO 
much planning, but some.  That means that it had to be thought through 
and included in all descendants.

So, classes makes life simpler when there is, in actuality, a good deal 
more complexity behind the scenes.


p.s.
There are plenty of tutorials on this.  The first one ask.com gave me is 
http://java.sun.com/docs/books/tutorial/java/concepts/index.html
-- 
Posted via http://www.ruby-forum.com/.