Design by Numbers

John Maeda has produced an important work. While Design by Numbers falls outside the usual fare of historians, it deserves their attention, since it will provoke them to reconsider how information technologies relate to their teaching and practice. In particular, it will press historians to reconsider the desirability of letting software companies design and produce their tools for them.

Maeda's ambition in this work is straightforward. He wants to persuade graphic artists to become computer programmers. According to the author, the attitude of many artists towards information technologies has been at best ambivalent, and at worst hostile.

Tools such as Adobe Photoshop seem to undercut the very idea of art, the disciplined exploration of an artistic medium. Years of practice developing particular motor skills are traded for the capacity to read software manuals and press buttons. A Japanese artist who has spent years learning to draw parallel lines with a bamboo quill, for example, is not likely to be attracted to software that claims to replicate the same effect in seconds.

Contemporary software packages offers a further drawback according to Maeda. Existing interfaces and tools reinforce current paradigms of digital media design, imposing a second order of constraint on artistic activity that artists should learn to do without.

If computer art is to be defined by the attributes of skill and discipline, the author argues, digital artists will have to assume responsibility for the design and creation of their software. It is only after artists have gone through the process of tool generation — the process of working within the structure and process of computer language — that they will be in a position to appreciate the constraints and potentials peculiar to electromagnetic media.

The main thrust of Maeda's argument is not supported by prose, but rather by code. Readers are invited to participate first hand in the process of form generation by downloading Maeda's programming system (available for free at http://dbn.media.mit.edu), and typing the provided code to produce the forms shown in the accompanying diagrams.

By following the prescribed exercises, and learning his simplified computer language, dbn (an acronym for Design by Numbers), Maeda hopes artists will come to appreciate the surprisingly complex forms that can be generated by mastery of a few simple commands such as "Line", and codification of a few simple processes, such as "Repeat". Using the code shown below, for example, users can generate the accompanying shape.

While Maeda's work has no direct relevance for the historian, its larger message bears consideration. As we find new applications for the computer in our teaching and research, are we prepared to continue accepting the constraints imposed by generic, market-produced software, tools that imperfectly fill our needs?

A concrete example of the relevance of Maeda's message can be found in the field of 3D modeling. It offers a marvelous opportunity for historians to teach students the problematic relationship between primary sources, historical representations, and the historical objects they purport to represent. A fundamental problem in the modeling process, however, is that of measurement. How does one determine the size of a structure that no longer exists? Photogrammetric software packages currently available on the market can retrieve dimensions from contemporary photographs, but not historical ones. The principles for generating a package that could measure old photographs have already been outlined in the photogrammetry literature, but have yet to find a place in a commercial product. If I had the programming expertise, I could in principle create such a tool myself. Since I currently do not have this skill, however, I cannot. There are other areas, such as Tom Taylor's work with computer games to promote critical thinking, where historians face a similar dilemma.

Maeda, therefore, presents a vision of scholarly and artistic practice fully commensurate with Robert Logan's Fifth Language, who argues that programming will be a skill as central to teaching and research as language. As an object-to-think-with, to use Seymour Papert's expression, it succeeds brilliantly.

As an venue for engaging in graphic arts, however, it can be criticized. Maeda's programming system does not allow users to control the size of the image they produce. Nor does it offer a straightforward way to export the image to an application a user might use, such as a web page. Unless the reader knows how to do a screen save, he or she is stuck.

This criticism aside, however, Maeda is to be congratulated for challenging humanists to reconsider the place of computing in their practice. Historians interested in the implications that information technologies present for our discipline will find it an invaluable resource.