Kevin Crowston
Although you will perform with different ingredients for different dishes, the same general processes are repeated over and over again. As you enlarge your repertoire, you will find that the seemingly endless babble of recipes begins to fall rather neatly into groups of theme and variations . . .
—Child, Bertholle, and Beck (1981)
Interdependency and coordination have been perennial topics in organization studies. The two are related because coordination is seen as a response to problems caused by dependencies. For example, Thompson (1967) hypothesized three coordination mechanisms—standardization, plan, and mutual adjustment—used in response to three different patterns of dependencies—pooled, sequential, or reciprocal (pp. 54-55). Most studies, however, describe dependencies and coordination mechanisms only in general terms, without characterizing in detail differences between dependencies, the problems dependencies create or how the proposed coordination mechanisms address those problems (Grant 1996; Medema 1996). This vagueness makes it diffcult or impossible to determine what alternative coordination mechanisms might be useful in a given circumstance. Similarly it is hard to translate from dependencies to specifications of individual activities or to uses of information and communication technologies (ICT) to support a process (e.g., as part of a business process redesign effort; Davenport and Short 1990; Hammer 1990; Harrington 1991; Harrison and Pratt 1993).
For example, consider the process of fixing bugs in a software product, a process that Crowston (1997) analyzed in a search for alternative coordination mechanisms. Using Thompson's theory, we might note that programmers all contribute code to a final product, and thus share a pooled dependency, and that they occasionally rely on each other's work, thus creating a sequential or sometimes reciprocal dependency. Furthermore, we might find, as predicted, that standardization, plans and mutual adjustment are all used. This analysis leaves many questions unanswered however. For example, how else might we organize this process? What dependencies would be left (or created) if instead of dividing the work among specialists, it were performed by generalists? Can we design a process that would reduce or eliminate the sequential dependencies between programmers? If not, what information do sequentially dependent programmers need to exchange and when? Would electronic mail or computer conferencing be useful to support this information exchange?
Although past conceptions of dependency do not directly address these questions, newer perspectives in artificial intelligence offer a more precise notion of dependency that do. In this chapter I draw on both literatures to develop a taxonomy of organizational dependencies and associated coordination mechanisms. The taxonomy suggests that there are four basic dependencies—task-resource, sharing, flow, and common output—each managed by a set of coordination mechanisms. The typology also suggests possible coordination problems caused by the fact that activities and objects in the world can be divided into subcomponents. The central section of this chapter is devoted to explaining these dependencies, their derivation, and the details of each.