Init
Experiments in programmatic art emphasizing interplay of iterative processes and randomness.
These images are the products of hand-coded, non-AI-assisted algorithms. Despite the lack of AI in this project, I still consider these pieces to be in some sense collaborative with machines. To create involves a relationship with the medium of creation, however unidirectional that relationship may seem. We do not typically think of the relationship between a painter and brush to be a collaborative one, but the advent of generative models has perhaps called this assumption into question. How does this type of collaboration differ from that of a human-AI interaction? Does the degree to which it is a collaboration differ at all? Why have we been drawn to this term, artificial intelligence, to describe generative, stochastic processes like neural networks and their products? When considering generative models, we might consider to what end they are entrained.
These images began simply as an exploration of randomness. Drawing rectangles, specific introductions of randomness, such as placement and color, is in some sense free of habits and preferences. It is agnostic, coming from a place of very little supposition, very little assumption. It is an attempt to begin axiomatically, minimizing the domain of the axiom as much as possible. How free are we from that which is assumed? How might we expand or contract these freedoms in time?
What is curve? What is round? Does it include jagged little edges and horns? Mutilated sine waves care little for these notions. Colors are again random. Placement is given a domain of composition. Each coordinate is potentially drawn upon, but origins are preferred in a cascading spiral. The distribution of this spiral is apparently evident, but our discomfort with randomness may lead us to wonder about things like sample size.
A small decision is made here. Solids and symmetry.
If hominid preference had not returned above, it does here. What kind of brush would make these? Some give a sense of direction in the brush stroke, but these are merely random. What does random mean here? Is there any tendency in randomness? We might know if we could observe, or at least observe attributes, of sufficient… oh, there’s that sample size thing again.
What if we delete small parts of an image over some distribution? They become scratchy, attenuated.
Strong commitments to composition and uniformity. What happens when our textured shapes cross a boundary? What is the effect of inversion, noise?
A reaction to the unfriendliness of patterned gradients? Large, clear shapes and complimentary colors. The same motivation for two opposite images reminds us of the enormity of our task. This too is comforting; enormity is familiar.
The images that follow are beautiful to me. They strike a balance between the will of the creator and that of the created, the influence of chaos versus that of control. The lone mass, fearful of integration. The bold sun, present, conspicuous in its absence, and never out of context. Why do we risk these interactions? What is left behind?
Many were made like this. Each is colorful with discrete shape placement. This one is noted for its sense of movement. Its color choice is suggestive of bloom, obscured in likeness and centrality.
The number of elements up for variation we might call degrees of freedom. As these increase, so does difficulty in discerning a sense of direction of these elements, particularly when viewed from a 2-dimensional plane. We might impose an illusion of 3-dimensionality or an actual 3-dimensionality to explore. This may account for the added principality of one such element, perhaps more. How might we continue to generalize our perspective into multidimensionality?
A strong preference for noise, even at the expense of colorful insults.
An experiment of 3-dimensional placement of polyhedra, with a strong preference for 2-dimensional, near-orthogonal perspective.
Future explorations may include other volumes, experimenting with vertex placement, number, and distribution. What might noise do to a surface? What might volumes of noise do when intersecting with solid volumes? Scratchiness and attenuation need also be considered.