Design by Numbers (review)

Throughout the history of computer art, there has been a vocal subcommunity interested in the investigation of code and algorithm as an aesthetic focus. Some have considered the exploration of algorithm as a way to penetrate to the essence of image, design and aesthetic experience; some have even felt there are resonances between the algorithms and the core of nature; other computer artists have been interested in code because it is an aspect of their craft largely not appreciated or understood by audiences. John Maeda's book, Design by Numbers, injects new life into those discussions.

Maeda is director of the Aesthetics and Computing Group at MIT's Media Lab. He and his students are famous for the interactive, visually striking systems they develop that are helping to forge new attitudes toward design. The power of these systems lies in the interplay of algorithm, interactivity and ultimate visual manifestation. Design by Numbers is an elegant introduction to Maeda's approach to thinking about code and visuality. To accomplish his goal, Maeda has developed a streamlined programming language called dbn, which is customized toward visual thinkers (dbn is available for free download on the Web). The commands of this language attempt to cut to the essence.

In an updating of the spirit of the Bauhaus school, Maeda systematically introduces a series of concepts, thought experiments and hands-on exercises that explore the interrelationships of design and coding. There are chapters on commands, lines, variables, repeat, dots, nesting and questions (conditionals). Later, more advanced chapters look at time, paint, reaction, touch, network, change and numbers. Each chapter simultaneously explores the core idea of the code and the visual concept. For example, the chapter about lines examines what a line is computationally and what it is visually.

The book invites hands-on experimentation and could serve as an introduction to both programming and design. Each chapter introduces the core idea in programming/visual display and then plays with variations that can be developed. Each code snippet is illustrated by an image of its visual result. All visual displays are restricted to a 100 x 100 pixel box, with a grayscale range for each pixel of 100, providing an elegant coherence to the book. The sequences of grayscale box experiments are visually wonderful to scan.

The advanced chapters tackle more complex issues. For instance, Maeda laments that many artists and designers never seek to penetrate beyond commercial computer applications, such as paint programs. They thus never realize what assumptions and limits underlie these systems. Even with his streamlined language, he leads the reader to explore the essence of image-processing, interactive systems, mice, keyboards and so on. His introduction to networks and distributed Internet drawing is brilliant—he poses a conceptually simple system in which people anywhere can read and write to a limited number of positions on a server and then proceeds to explore shared visual creation systems.

There are some places the book could have gone further: some readers will find the simplified, code-oriented approach overly modernist; the book is short on philosophical discussion about the assumptions of this system and the underlying faith it implies in the ability to create totally controllable systems, and there is also little consideration of the fact that numbers can only represent certain domains poorly—such as emotion or semantics. The discussion on randomness, however, is revealing: Maeda considers noise an unfortunate distraction for elegant systems, while others might consider it the essence of contemporary life. The book would also have been enhanced with some discussion of the long history of algorithmic computer art. Despite these faults, however, I strongly recommend this book.