Hi all,

If you are interested, I did write this article a few years ago...

Aesthetic selection and the stochastic basis of art, design and interactive evolutionary computation


And this one also...
Chance and complexity: stochastic and generative processes in art and creativity

There might be (I hope!) something of interest and relevance in those articles for you :-) Random processes and serendipity are vital to creativity. Not to be underestimated in my opinion.

Best wishes,

On 11 Jul 2018, at 3:54 am, lonce.wyse <cnmwll@nus.edu.sg> wrote:

Often randomness functions as a stand-in for diverse and complex mechanisms that we need to break symmetry, introduce "innovation", force the modeled parts to deal with complexity, etc., standing in for processes that are not themselves the focus of our work.

- lonce

On 11/7/18 8:18 AM, Jim Andrews wrote:
Burroughs used the random to avoid making the choices he normally would; he talked about using it to cut the loops of one's conditioned behaviour. Perhaps he hoped that the results would "stimulate the unconscious" of the reader or listener.

Speaking of different distributions, in http://vispo.com/aleph3/an.html?d=Jim%20Andrews I wanted the location of the 'brush strokes' to be random--but I wanted them to happen more frequently in the center of the screen. That is, I wanted them to randomly occur anywhere on the screen, but happen more frequently at the center than at the left or right. The solution I came up with was to generate three random x-axis values that could be anything from 0 to the width of the screen, and then calculate the average of those three values. That gave me the x value of a brush stroke. The y value was simply a random number from 0 to the height of the screen.

That's one way to generate a non-uniform random distribution: take the average of several random values.

For instance,

x1 = Math.random();
x2 = Math.random();
x3 = (x1 + x2)/2;

Both x1 and x2 are random nums between 0 and 1. So too is x3. But x3 has a different distribution than x1 or x2. x3 has values around 0.5 more frequently than it has values around any other number.

x1 = Math.random();
x2 = Math.random();
x3 = Math.random();
x4 = (x1 + x2 + x3)/3;

And in the above, x4 is yet more tightly clustered around 0.5


On 7/10/2018 3:44 PM, Philip Galanter wrote:
I randomize the initial gene pool in evolutionary systems. If things are working properly it doesn’t really matter much what the initial genes are, as long as there is some diversity in the gene pool. But spreading them out in the parameter space can help find the maxima quicker/better.

In my generative art class I warn students that randomization is the first and most obvious generative method. It’s a good starting point, but at this point in history probably not a good stopping point.

I suggest that there should be a reason for using it. Cage used randomization to emphasize a zen-like sense that all sounds are of equal value, and it’s an illusion of our human condition to hold some higher than others. Burroughs used randomization to stimulate the unconscious. Ellsworth Kelly used randomization as a result of his interest as to how decay (entropy) can create complex form from what used to be highly ordered (and thus less complex) form.

So my two challenges would be this. Why use randomness? And why use uniform distributions rather than, say, normal distributions like those typically found in “random” aspects of nature?


On Jul 10, 2018, at 8:40 PM, John Clavin via eu-gene <eu-gene@we.lurk.org> wrote:

Randomness is a touch of seasoning artistically added to the design of an algorithm.

John Clavin

Sent from my iPhone

On Jul 9, 2018, at 11:38 PM, Jim Andrews <jim@vispo.com> wrote:


What do you use the random for in your art?


The main thing I use it for is exploration of a compositional space. I think of a compositional space as the set of all possible screenshots you could take of a program in action. The compositional space is the set of all possible compositions your program can create. The compositional space involves various parameters/variables that have different values from frame to frame of the program's animation. Such as the positions of the objects in the art, or the particular set of objects in the art, if there's a large set of objects to draw from and you don't have them all on the screen at once. And their sizes, opacities, rotations, possibly their colors, backgrounds, borders, and so on. Giving these properties of the compositional space random values (constrained within some suitable range) is a good way to explore the compositional range of the space.

As opposed to giving the parameters values that are incremental by small amounts. Because, generally, there are enough possible values that the number of possibilities is astronomical, and giving the parameters small incremental values would result in exploring only a small part of the space. Like the difference between exploring the total space of the universe via the usual travel (inch by inch) or by appearing, from moment to moment, in random parts of the universe. Now you're near the beginning of the universe. Now you're someplace 4 billion years into the thing. Now you're near the end. You get a better sense of the range of the thing, maybe, than the inch by inch approach.

Another example; this is what I've been working on recently: consider the compositional space defined by the bill bissett dirtee konkreet brush at http://vispo.com/aleph3/an.html?d=bissett%20dirty%20concrete . This uses 300+ images (440 Mb) of the concrete poetry of the amazing poet bill bissett in a remix/synthesis. There are lots of possibilities. Each time you view this animation, it's different from the previous time cuz of the use of the random. Whenever it draws a circle, the position of the circle is random, the size of the circle is random, the bill bissett poems chosen to fill the circle are random, the size and position of the bissett poems within the circle is random, and so on. There's a lot of random choices going on. This explores the compositional space more interestingly than if each of those parameters was incremented by a small number each frame of the animation. Cuz, in that case, the changes from frame to frame would be too small.

Mind you, if the values were incremented by big numbers relatively prime to the number of values possible for each parameter, that'd be a different story. It would both move through the space quickly and cycle through all possibilities (the product of the primes in the prime decomposition of the number of elements of each parameter) before repeating, so would look fairly (but not quite) random.

Years ago I called a compositional space a combinatorium when I was developing the stir fry texts ( http://vispo.com/StirFryTexts ). I thought of a combinatorium as the set of all possible texts (or screenshots) that can be created from a particular stir fry text. Once you get into programmed art, you eventually start thinking about compositional spaces or combinatoriums or whatever you want to call the set of all possible texts or screenshots that your program can generate. Or, if it's more narrative-oriented, maybe you're thinking about the set of all possible paths through some kind of story space, or whatever.

Still, I don't think the random is the main source of surprise in my work, though it hopefully does provide an element of surprise. The surprise is when we come across good ones. I don't make my art so that everything it generates is good. I make it so that the possibility hopefully exists of some strong work being generated. My philosophy is that if you want that possibility, you've also got to let shit happen. If everything that the system can generate is more or less equally good, then nothing is exceptionally good.
Other artists may put things together differently. But in my work, most of the generated instances aren't as well-formed as a few of them—at least as I see it; others may see it differently. That is, they may have different favorites or not have favorites at all and be able to argue convincingly for their point of view.

In most of my generative visual works, I take screenshots of what I take to be good ones and present those screenshots in slidvids such as http://vispo.com/aleph3/images/bill_bissett/slidvid15 in which you see a slideshow of 200 screenshots of Aleph Null chewing on a considerable portion of bill bissett.

Similarly, contrast http://vispo.com/aleph3/an.html?d=Jim%20Andrews with http://vispo.com/aleph3/images/jim_andrews/slidvid . The former is Aleph Null chewing on 230 photos of Trump, his cabinet, his advisors, and a few Republicans. The latter is a slideshow of 430 screenshots from that dynamic animation. The dynamic animation does a lot of becoming; it doesn't look particularly face-like, often. But sometimes a real gargoyle pops out, momentarily, of the stream. Those are the sort of screenshots I'm interested in. The compelling ones (evil, ridiculous or whatever). They aren't morphs. Morphs carefully map noses to noses and eyes to eyes etc. In these ones, the position of the eyes varies between photos. Same with the other features. So that sometimes the dynamic animation doesn't look face-like. But the largeness of the range of position of features is what makes for some unique, compelling results, occasionally.

The experience of looking at the screenshots is different from the experience of viewing the dynamic animation. The dynamic, never-quite-the-same-twice animation has a raw, protean energy of swift becoming. The experience of viewing the slideshows, on the other hand, is one of slowly viewing a series of completed compositions.

The dynamic animation is dynamic largely through the use of the random to determine the values of many parameters from frame to frame of the animation. There's nothing random about the slideshows, however.

Still, I think the main source of surprise in my work is not from randomness.  It's surprising to discover you can import your own images into Aleph Null. More generally, it's surprising that Aleph Null is both a tool and an art work in itself. And it's surprising that there are amazing visual poets who each have a nib or two in Aleph Null. It's surprising that this sort of thing is available on the net. These realizations are more surprising than when something good pops out of the stream of the dynamic animation.
eu-gene mailing list -- eu-gene@we.lurk.org
To unsubscribe send an email to eu-gene-leave@we.lurk.org
eu-gene mailing list -- eu-gene@we.lurk.org
To unsubscribe send an email to eu-gene-leave@we.lurk.org
eu-gene mailing list -- eu-gene@we.lurk.org
To unsubscribe send an email to eu-gene-leave@we.lurk.org
eu-gene mailing list -- eu-gene@we.lurk.org
To unsubscribe send an email to eu-gene-leave@we.lurk.org

Important: This email is confidential and may be privileged. If you are not the intended recipient, please delete it and notify us immediately; you should not copy or use it for any purpose, nor disclose its contents to any other person. Thank you.
eu-gene mailing list -- eu-gene@we.lurk.org
To unsubscribe send an email to eu-gene-leave@we.lurk.org