Friday, 29 January 2016

Compositional Methods

Artists and Engineers both build systems from components.

They learn methods of composition that place them at a certain level of abstraction. A musician for example may learn to compose at the piano, a painter at the drawing pad and an engineer at the computer screen.

These distinctions are fading. Many artists, just like engineers, will spend a generous part of their composition time at a computer screen.

This leads me to ask, how formal are those artistic activities spent at the computer screen. Whatever type of software the artist is using, the computer is capturing their processes systematically so that they could, in theory at least, be replayed over and over again.

The process of composition is one of combining components. The nature of the components dictates the level of abstraction that the engineer/artist is using.

Consider painting, for example. The artist is going to combine colour regions by collage. Colour and region are the components. Collage is the method of composition. The choice of colour and the choice of region shape and location are the parameters of the components.

Consider musical composition. The composer is going to juxtapose musical phrases, sequentially and concurrently. They are going to transform those phrases by transition and more complex transformations, such as inversion. The phrases are the components. Juxtaposition and transformation are the methods of composition. Choice of chord and timbre [instrumentation] are the parameters of the components.

There is a commonality to these compositional methods. The maker thinks in terms of items to be combined. The choice of items and means of combination define the level of abstraction. Once that has been chosen, experimentation with varieties of combination leads along a path of discovery culmination in the production of a new object of interest.

In computing, perhaps more so than any other domain, the choice of components and the choice of means of combination are made explicit. Particularly, one chooses a tool and the tool defines the nature of the components it expects. Direct manipulation tools for images, for example, expect raw images to be available in common open formats (jpg, png etc.) and allow for the editing and mixing of these images to produce new components of the same sort. As it is with images, so it is with music and videos.

More advanced tools provide more advanced means of composition. But they also make assumptions about what it is you want to do and how you want to do it. On the one hand this accelerates the completion of common tasks, but it often excludes or makes difficult the exploration of new pastures. We are defined by the tools we choose.

I find that being explicit about the component set that I am working with and the means of composition enables me to choose (or make) the right tool for the research I need to do. For example, in composing images, I know that I want to be able to create stencils and to paste one image on top of another through the stencil. So my components are images and stencils. And my means of composition include pasting through a stencil and creating new stencils from images. So my inquiry, what happens if you make a new stencil by combining existing stencils, is enabled by the clear choice of components and means of composing.

[continues here]