Movement in workplace environments - configurational or programmed?

In countless case studies space syntax research has found that the configuration of a spatial system offers a powerful explanation to movement flows. However, this relationship is restricted for complex buildings where movement cannot be assumed as random since there may also be a programme that requires specific actions and interactions. A distinction has to be made here according to the nature of the organisation occupying a building: a strong programme building where the interaction and co-presence of people is highly controlled may not allow movement flows to follow configuration. In contrast, a weak programme building with an all-play-all interface might be expected to experience more randomised movement patterns increasing the significance of configuration as determining factor. Though being useful, these assumptions lack the power to fully explain real life movement flows in workplace environments for two reasons: firstly, most workplace environments follow neither purely strong nor simply weak programmes, they lie in-between the two poles and comprise aspects of both systems. Secondly, configuration considered as the crucial cause of movement in an office may even be limited for weak programmes due to the effects exerted by everyday attractors such as the coffee machine, the watercooler or the photocopier, toilets or the building entrances. This paper explores different strategies for explaining observed movement patterns, among them axial and segment analysis. It aims at an in-depth analysis of strong and weak programme aspects in order to find ways of understanding office movement patterns. The data used stems from two case studies representing those ‘in-between’ settings: a university school and a research organisation hosting theoretical physicists. The results suggest that movement in these workplaces may be reflected best by a metric analysis, as opposed to urban movement that follows angularity patterns. Distances seem to matter most in small and well known spaces. Moreover, it can be shown that flows of people can only be explained through configuration whenever it is possible to exclude attractor driven movement. On this basis a new approach is suggested that combines configuration based integration measures with attractor based ones in order to predict actual movement flows in offices.