The Next Generation of Generative Art

2006 / MERZ AKADEMIE / FINAL PROJECT / INSTALLATION / COMPUTATIONAL DESIGN / PROCESSING

Applets

You can try the programms as applets in your browser (requires JAVA):
PARTICLE_01 / PARTICLE_02

Project details

The pictures and applets shown are the practical part of my master's thesis. The installation consists of two computer screens, each displaying a program written by me in Processing, generating unique graphics, relying on common, often used mechanisms like particles and randomization. Picking up the fact, that despite these are often used by beginners in the field of designing with code, they can lead to distinct artworks.
Questioning the obvious assumption, that tools, which offer an easier access to programming by simplifying certain aspects might lead to a lack of variety and depth in the works.

Thesis Abstract

In the last few years much software was developed to make the interactivity and benefits of programming more readily available for a multitude of artists and designers. Programs such as Max/MSP, Flash, Director, vvvv and Processing all quickly gained popularity among their target audience because they allow a simplified approach to the otherwise complex task of processing and visualizing data in different ways or creating interactions between objects.

Each program uses a slightly different or unique approach. Max/MSP and vvvv offer some kind of visual programming environment where the user can create flowchart-like program structures with boxes and patch cords. Flash and Director are authoring tools with a timeline based interface, giving the user the possibility to extend their functionality through their own proprietary scripting languages, while Processing offers a Java-like programming language with a focus on visual functions.
Relying on these tools, the designer or artist can achieve visual results that rival those of custom written programs based on classic programming languages such as Java, or C without typing hundreds of lines of code. However, programming is a complex task which offers in return great flexibilty in the realization of one's projects. Simplifying the process of programming means to hide certain steps in that process from the user, therefore reducing the options and functions and thus limiting his/her possibilities.

This thesis discusses the influences of those programs on the field of generative art. Providing a set of ready to use functions and/or a simplified programming environment assists creative tasks for an audience which would otherwise have no access to it because of lacking or nonexistent programming skills. However, this ready-made shortcut might also hinder the user to experiment with algorithms and processes later on when he tries to achieve more specific goals. This could lead to a flood of works built on these toolsets, sharing similar aesthetics as a consequence of the limited options available to a designer or artist in the creation process. Simplifying the creation, concentrating on the immediate visual result may also degrade the importance of experimentation as an underlying idea of the creative process. This can result in a lack of depth in the works, if the there is no option for the user to evolve his skills beyond the given toolset.

In addition, one has to consider if it is desirable to oversimplify the task of programming. The computer and it‘s software have become a part of our daily life and culture. Designers or artists in the field of New Media, need to be literate in this medium, as it is in the center of their activities. However, literacy and full comprehension is only achivable by being able to read and to write in the said medium. This means being able to use and create software. The latter is only understood by being capable of writing code, and constantly building upon an understanding of the underlying principles.

Within the scope of software-based generative art, this could mean the artist is most likely not fully aware of what he is doing, as an illiteracy in the medium might be enforced by given toolsets.

Using the programming language Processing as an example, an overview is given on the topic of probable similar works, and a critique arises out of the simplification of creating software as well as a possible solution in the form of teaching procedural literacy at design schools.

The practical part of the final project explores the possibilities given to the artist by writing code. Custom programs are built with Processing to generate graphics. They contain often used aspects of works created in the recent period:

• Particles serve as foundation for the generative system, they are a favoured object because they represent the principles of object oriented programming.
• Randomization is used for control of the particle behavior, because it allows the artists to add autonomy to the process of creation.

However, even with these key elements, the artist is confronted with a plethora of decisions. For example:

• How should the particle system behave at all?
• Which parts of the behavior could or should be random?
• What is the general graphical representation?
• Is there an interaction between single particles? What kind of interaction?
• Which colors should be used?

This list ist very rough. Writing code is basically about describing processes in their finest details. Nearly everything comes down to being a decision with an impact on the end result. First, the artist has to be aware of what he wants to do, because it is impossible to describe a workable process if it isn‘t connected to a clear goal. Second, small changes in a variable of the code can result in a completely unintended aesthetic or vice versa. So even experimentation is always linked to the deliberate decision for one specific outcome.

The result of this exploration are two generative artworks which are quite unique in their behavior, aesthetics and audience perception, which is emphasised by the fact that each of them is shown on their own computer screen.