* I am still experimenting with images. I am making digital images; I am "generating" them I should say. I am generating digital images starting from the initial conditions of "pure Gaussian white noise", more or less.
* I'm trying to be thorough, methodical, and systematic about it, but for now I'm unable to do the operations I want to do solely through direct numerical computation. I have to use an image editing software, and that upsets me. What I need to be doing is writing Python programs to do the same basic operations. In due time.
* This is what I've got for now:
* The experiment is simple: I'm trying to go for the "most complex beauty". That is, by utilizing simple rules and only simple rules, I am experimenting with the (computational) generative emergence of complex beauty. I'm not sure I said that right. Let me begin again.
* The idea right now is to use the SIMPLEST RULES, and only the simplest imaginable, to GENERATE images. I want those images to a) be BEAUTIFUL and b) I would like there to be at least "some" complexity.
* The idea is not to produce Low-Complexity Art. I want there to be entropy and complexity and NOISE basically. That's the thing. This isn't about Compression. This is about maximizing the user experience, and I think I need a high-enough level of complexity to achieve that.
* Another example:
* I am ONLY using primitive (basic) operations. I generate noise, I pointillize, and I pixelize. That's about it. And I play with colors. I'm not doing anything I couldn't do through numerical computation alone. At least ideally, that's the goal.
* I made some more images. For now, though, I want to recollect the research I was doing recently on Derivatives.
* Basically, what I came to understand is that a derivative is more or less just a rate of change. When I say something is "the first derivative of x with respect to time" I'm making a statement about the rate of change of a variable with respect to time.
* I am basically just saying something about a changing quantity.
* You have an input value and an output value. It is saying there is a functional relation between the two.
* Math and science are basically just The Study of Change. (Signals are also changes in quantities.)
* I'm still obsessed with this record-keeping/documentation thing. Look at it this way. I've been doing a lot of experimenting with image processing. I've been exploring and experimenting with Python for 3 or 4 years now. I'm also playing with some simple software to manipulate images.
* What I want is to document the process. No experiment should ever go without records written down in the laboratory notebook.
* On the subject of notebooks, I've been having trouble installing the IPython Notebook on my Windows computer. It's okay because it will work on the Linux machine, but it reminded me how difficult things are on the Windows platform. It's always configurating this, configurating that. It never ends. Stuff only half-works part of the time.
* With that being said, I'm likely to end up reinventing the wheel since none of the computational resources that I would need exist. I'll have to write my own libraries/modules. That's okay.
* With the digital images, though, I'm following a very precise step-by-step procedure, an "algorithm" except that for now I'm doing it all by hand.
* I keep trying to use only the most basic functions that I know I will be able to perform through numerical computation. I've already done the math, I just need to translate it into Python code.
* Funny enough, the Processing platform (Processing.org) has a new Python Mode. I tried to install that on the Windows machine and she was a no-show. Error somewhere in installation.
* As always, run into a problem, have no one to turn to for guidance. Back to square one.
* I made this a short while ago. As you may know, I've been working on The Refcards-System now for years. I made a little Refcard to go in the collection:
* Note: It's winter in my heart of hearts.
* ANOTHER REFCARD FOR THE REFCARD-SYSTEM: