Johnson - Design computing : an overview of an emergent field

I met Brian Johnson online. The year was 2005, and I had applied for a tenure-track teaching position at the University of Washington in Seattle. In the back-and-forth communications that are an inevitable part of academic job applications, I came to know Brian, and after the search reached its conclusion he encouraged me to join ACADIA (the Association for Computer-Aided Design in Architecture). My hunt for employment eventually took me to North Dakota State University, where from a vantage point some ten years distant, I can look back on my time in ACADIA and on conversations and collaborations with Brian over the years as critical to my understanding of the issues and questions of design computing.

But what exactly is design computing? In 2014, I invited Brian, Erin Carraher, Ning Gu, JuHyun Lee, and Wassim Jabi to join me in co-editing a special issue of Architectural Science Review on the theme of design computing (issue 58.3). As a group, we worked to define a shared understanding of the rapidly expanding content and the nebulous boundaries of the field. Its disciplinary content is characterized by a range of interests and issues which frequently overlap and cross-pollinate: simulation, visualization, and representation; humancomputer interaction, user experience, and cognition; artificial intelligence; fabrication and prototyping; parametric design and scripting; data processing and intelligent modeling; smart environments, controls, and infrastructures; algorithmic thinking; mobile, ubiquitous, and pervasive computing; and cloud computingjust to name some of the most central. Clearly, design computing, which Brian argues exists at the intersection of the built environment, design, and computing, is a content-rich field about which much has been written but which (as of yet) shows no risk of exhausting itself as a robust field of inquiry.

Its boundaries are indeed nebulous and characteristically difficult to define. At times, design computing shares formal methodologies with computer science and mathematics, seeming to succumb almost wholly to algorithmic approaches and the application of logical reasoning. However, considered at other times or looked at in different ways, design computing forcefully calls the value of logic into question: Intuition, aesthetics, and human experience all have their roles to play. To precisely delineate the boundaries of design computing is a Sisyphean task. In one sense, I see Brians book as an attempt to define the fields core values and approaches through comparison with its kindred disciplines. Often, he makes these comparisons explicit, as when he discusses traditional and digital media; but occasionally the comparisons are latent within his text and emerge only after repeated readings.

Like mathematics, design computing is concerned with patterns, whether those are patterns of thought, patterns of practice, or visual or spatial patterns. In particular, design computing is concerned with the computability of patternsthe how and why of algorithmic approaches. But unlike the rigorous discipline of pure mathematics, at its core design computing is about applicationand it is a unique kind of application which demands constant interpretation and self-critique. Brian addresses this relationship in his first chapter:

In calculus [he writes] we learn that functions have minima and maxima that can be found by differentiating the equation and solving for the derivatives zero values. Where such situations exist, that is, where a mathematical expression describes the value of a design, there is a direct mathematical, or analytic means of arriving at the best solution. We can, for example, select the best depth for a beam carrying a given load. Automating the improvement of a design is hard. There may be no clear mathematical relationship between the changes you can make to the representation and the performance metrics you wish to improve.

Thus, while design computing is abidingly concerned with the patterns and structures of computing, the questions that drive the field emerge from the needs of design. Quantifiability and computability are indeed essential tools, but there remains a dimension of design computing clearly inaccessible to automation and algorithms. Now, again like mathematics, and also like industrial or product design, the discipline of design computing seeks to identify and solve problems of a wide range and scope. But as Brian acknowledges, every problem isnt a design problem, and design doesnt require a computer. Rittel and Webbers wicked problems (Rittel and Webber 1973) hold special significance in design computing, for like acts of design, these problems are open-ended and resistant to single solutions.

Regardless of their specific discipline, designers rely on media: the material of their work. For example, architects are characterizedat least historically soby their reliance on drawings and physical models; graphic designers similarly by paper and film; and product designers by physical prototypes. Designers media are structuredthink about how they conventionally adhere to established rules of representation and the limitations of scale, or how the structure of software influences the direction and scope of inquiry. Design computing recognizes that the structure of media, whether analogue or digital, is often computable and thus inviting of algorithmic analysis. And yet again, there remains a dimension to the field which is not obviously susceptible to a pure-mathematics approach. Brian notes how the fluidity of interaction and range of expression made possible by pen and paper is substantially challenging to register through digital tools. With this comment and in his subsequent discussion, he suggests that design computingperhaps more so than other design disciplinesacknowledges that media and representations are not neutral with respect to their application or their outcomes.