Paper short abstract:

Modularity is a central design strategy in modern computing, allowing for both the management of technical change and the vertical desintegration of markets. This paper discusses the implications of Simondon’s concept of concretization for popular accounts of modularity (e.g., Lessig, Zittrain, Wu).

Paper long abstract:

The principle of hierarchical decomposition was first articulated by Simon in his 1962 paper, "The Architecture of Complexity." Simon proposed that many complex phenomena, both natural and artificial, could be described and analyzed as "composed of interrelated sub-systems, each of the latter, in turn, hierarchic in structure until we reach some lowest level of elementary subsystem." In its eventual development as a theory and practice of modularity, Simon's theory of complex systems will have an enormous impact on the design of computing systems: (1) for engineers, modularity will gradually emerge as the primary design strategy to manage complexity and technological change; (2) for managers, modularity will develop as a particular market structure with the power (at least in theory) to 'autonomously' generate innovation and an optimally competitive marketplace. It is these two features that, in the words of Lessig, Zittrain, and Wu, make the Internet uniquely generative.

Simondon's concept of concretization describes the evolution of technical objects as moving from a state of artificial abstraction to one where they "cease to be at war with themselves." This evolution is fueled by the incompatibilities between the abstract ensemble of relations, towards a gradual augmentation of functional synergies between structures. As Adrian Mackenzie states, for Simondon, "movement from abstract to concrete constitutes a form of change or becoming specific to technological action." In this paper, I explore the application of concretization to the historical evolution of networked computing systems and what it reveals about the potential and limits of modularity theory.