The purpose of this paper has been to outline what a methodology for analysing and designing information systems would look like if it were focused on action. We have identified the following three important aspects of actions and these form the basis of our intervention.
Actions are multi-scale in nature.
Actions are always performed in an action context; that is, actions are conducted by an actor in time and space, making use of resources.
Actions are dependent on the execution of other actions for instantiation as part of an action system.
In order to act, an actor needs to know what to do next, how to do it and when to do it. Another way of saying this is that in order to perform an action, an actor needs to know that the action is feasible in the world (i.e. that they can perform that action at the time, in the space, and with resources at hand). The actor also needs to know that the action is possible now (i.e. that the action dependencies have been satisfied). Hence, information about the action context and the action dependencies are both necessary and sufficient for the actor to act routinely.
Our approach to information system design makes use of these two points of leverage. We manipulate the action possibility space to ensure that the action context is appropriate to the required action. By representing the possibility for action, we inform the actor about satisfaction of the action dependencies and signal that they can act now. Our approach leads to a different view of what information is and how best to support action. This points the way to future work on a precise definition of what constitutes an information system.
We suggest that the methodology will be useful to organisations by allowing them to increase efficiency and effectiveness, through supporting routine action, which is generally marginalised in conventional IS methodologies. Unlike conventional information systems, as well as increasing temporal efficiency, situated systems also aim to increase human efficiency; in particular, to reduce wasted human effort expended in search of information. They can also increase effectiveness by making incorrect and possibly dangerous actions impossible to perform. Through the designs we have produced in the project case studies (Johnston et al., 2005; Waller et al., 2006), we have already provided evidence that systems designed using the situated analysis and design methodology are likely to be more lightweight than those designed using conventional methodologies.
As well as being of practical significance, the methodology is theoretically important as it articulates for the first time what many experienced practitioners know tacitly: it is necessary to be sensitive to the importance of action. In addition, the situated analysis and design methodology provides theoretical support for the practical application of ubiquitous computing and this is something that we will be investigating further in future research.