Notes on Prof. Kostas Terzidis’s idea of Supernatural Computation

Design as search through discreet spaces

A concern that commercial CAD systems prescribe design processes as the deterministic generation of form “out of nothing” leads Professor Kostas Terzidis to propose, as an alternative, algorithmic devices capable of producing -I use his words- “any possible solution”. In this view optimization and search techniques become avenues to navigate the vast solution spaces yielded by combinatorial “design machines”, notions that led to one of the presentation’s boldest claims; that all designs pre-exist and that creativity can be thought of as an objective search for a single ideal. From the many aspects in which this claim seemed worthy of discussion and debate I was particularly interested in its reliance on the absence of the programmer, the person who encodes the rules and designs the algorithms, as an active character in the story. This absence (a deletion, actually) may stem from Kostas’ belief that recursion and optimization, as well as other computational processes, are beyond the domain of human understanding, an idea he explores in his book “Expressive Form”, which results in a somehow mystifying attribution of agency to the computer.

Other things are lost with the deletion of the maker of the algorithmic machines, such as the hard work invested in building them, and the many decisions taken in order to establish their internal coherence. It may be useful to remember that researchers and practitioners have long equated these decisions to a constraint-driven design process. The deletion of the human, I would argue, impoverishes the eloquent simplicity of many of Terzidis’s design machines, many of which result in interesting images and artifacts, as well as in sometimes truly inspired pedagogical tools, surrounding them with a halo of mysticism that seems more geared towards blurring than towards clarifying the computational principles on which they are based.

Professor Terzidis’ bold stance poses many questions for those of us who value the enterprise of de-mystifying computers and programs, recognizing them as active participants of design processes, without hiding our key role in shaping them with our desires, interests, and yes, good old fashioned hard work.

Daniel Cardoso