Background – University of Copenhagen

CompArt > Background


Development of advanced control systems

Humans have for centuries utilized naturally occurring biological or chemical processes, i.e., causal processes that under certain conditions proceed predictably and reliably without human intervention. The use of fermentation for preparation and conservation of food and drink is a case in point. Furthermore, since the late 18th century, engineers have constructed mechanical control systems to automatically monitor and regulate the behaviour of production processes (spinning, weaving, printing) and entire machine systems. James Watt’s centrifugal speed governor for steam engines from 1788 is a classic example.

The invention of the stored-program computer in the wake of World War II and subsequent inventions such as the microprocessor provided the technical basis for a new kind of control systems in the shape of computer programs or ‘software machines’. While no less ‘mechanical’ than their 19th century predecessors, software-based control systems provide the basis for constructing machinery in ways that are far less expensive and at the same time much more flexible. Advances in computing thus provided the impetus for the rapid development of modern control engineering and of evermore powerful and flexible machine systems.

Mechanical regulation of human interaction

What is truly new, however, is not that we exploit technological advances to construct more sophisticated control systems, but that we now construct mechanical systems, in the shape of computational artefacts, for the regulation of human interaction. This phenomenon is theoretically underdeveloped and empirically understudied. We have no adequate conceptual foundation for the further development of such technologies with a view to their incorporation in organizational and social practices.

In the absence of a conceptual foundation for the design and implementation of computational artefacts for the regulation of interaction, a range of methodologies such as, e.g., ‘Software Engineering’, ‘Systems Development’, and ‘Participatory Design’, have been developed for the purpose of dealing with the practical challenges of developing adequate computational artefacts for particular work practices. While indispensable in practice, such methodologies are obviously inadequate, as evidenced by the endless succession of failed IT projects.