Archive for October, 2010


October 30, 2010

Sound is one of the most intriguing artifacts there is. On one hand, it is definitely real. You can hear sound. On the other hand, modelled on a computer, sound turns into AAA (an artificial artifact). Understanding the laws of real sound, and simulating these laws by incorporating them into the characteristics of artificial sound is a fascinating endeavor that I have always been interested in, but have done little about. I should have all the necessary background in mathematics and computer science to penetrate quickly and deeply the theory of sound and to write computer tools that accompany my deepening understanding of the subject. It seems also to be the right object of study to resharpen my mathematical skills and maybe make me understand mathematical topics that I have only glossed over so far and never gained a really intuitive understanding of.

As always, learning by doing (and experimenting) is the best way to go. I have therefore setup a github project that will contain the book where I will write down the stuff I have understood so far, and accompanying computer code.

Where to begin? This is a large field but there seem to be no comprehensive sources that explain it all, from the physical properties of sound, to the artificial generation of sound, and representing, storing, loading and playing back sound on a computer. Hopefully I will develop such a comprehensive resource. But for now, let’s start with the beginning, which seems to be the physical properties of sound. And who could be a better teacher of physics than Feynman? So my first step is to digest chapter 47 of volume 1 of the Feynman lectures on physics: Sound and the wave equation.

It’s better to burn out than to fade away

October 23, 2010

Thunder Lizards and Pivots, a must-see for every entrepreneur.


October 10, 2010

I had already made big plans for how “this” works in Babel-17, including how there would be two different inheritance mechanisms depending on the use of “this”.
But in combination with definitions, there exists already a way for an object to refer to itself; well, maybe not to itself, but to an identical copy of itself, which in a purely functional language pretty much is the same. For example instead of

  def message = "hello world"
  def double = (this, this)

one could get rid of the “this” keyword by the following reformulation:

  def myobj = object
    def message = "hello world"
    def double = (myobj, myobj)

It seems worthwhile to refrain for now from a special “this” keyword for Babel-17, and to see how far one can take the above reformulation.