COORDINATION THEORY - Human-Computer Interaction and Management Information Systems: Foundations - page 117

Malone and Crowston's (1994) first agenda item was how to represent coordination processes.
They note several possibilities, such as flow charts, Petri nets, and state transition diagrams. In more
recent work, Malone et al. (1999) propose a representation technique (the “Process Handbook”) that
represents processes at various levels of abstraction, as described above. A further advantage of this
approach is its ability to suggest new processes by navigating through the dual hierarchies, abstrac-
tion and composition. As well, the work reviewed above on modeling techniques shows that coor-
dination can be incorporated into many modeling techniques.
The second agenda item was to determine what kinds of dependencies there are. Crowston
(1994; 2003) conceptualized dependencies as arising from shared use of resources by multiple tasks,
thus providing a conceptual basis for a typology of dependencies, though the resulting typology was
nearly the same as shown in Table 7.1. Resources in this framework include anything produced or
needed by a task. For example, Faraj and Sproull (2000) introduce the notion of expertise coordina-
tion (“the management of knowledge and skill dependencies”) that “entails knowing where exper-
tise is located, knowing where expertise is needed, and bringing needed expertise to bear.”
The third agenda item was how to classify different coordination processes. The initial
approach to this problem was to list coordination processes by the dependency they addresses.
Malone et al. (1999) take this approach further by proposing a hierarchy of coordination processes
from general to specific. Other authors have proposed different organizations. For example,
Etcheverry, Lopisteguy, and Dagorret (2001a; 2001b) propose a catalog of coordination patterns.
A pattern is defined as “a solution to a problem in a given context,” so the catalog is organized by
coordination contexts. More work could be done to bring together and organize the mechanisms
that have been studied.
The fourth agenda item was about the generality of coordination mechanisms. Most of the
work applying the typology of coordination mechanisms has assumed rather than tested the gen-
erality of the mechanisms. Nevertheless, the list of mechanisms does seem to have been useful in
a variety of settings.
A final agenda item was how to analyze specific coordination practices, e.g., resource alloca-
tion. Malone and Crowston (1994) asked, “Can we characterize an entire 'design space' for solu-
tions to this problem and analyze the major factors that would favor one solution over another in
specific situations?” Most applications of CT are not very explicit about evaluation or factors that
make particular coordination mechanisms more or less desirable. Some work has addressed spe-
cific metrics for coordination. For example, Frozza, and Alvares (2002) offer a list of criteria for
comparing mechanisms: predictivity, adaptability, action control, communication mode, conflicts,
information exchange, agents, applications, and advantages and disadvantages. Albino, Pontran-
dolfo, and Scozzi (2002) develop the notion of coordination load, “a quantitative index that mea-
sures the effort required to properly coordinate a given process,” based on an analysis of the
workflow in the process. The goal of this index is to allow a comparison of alternative coordina-
tion modes. Nevertheless, it is clear that we are far from characterizing the design space for any
of the identified dependencies or coordination mechanisms.
In conclusion, the past ten years have seen considerable progress on the coordination theory
agenda laid out by Malone and Crowston. The basic outlines of the theory are clear. Methods for
capturing and documenting processes and coordination mechanisms and systematic typologies of
dependencies and mechanisms have been developed. There are numerous examples of the appli-
cation of CT in a variety of settings. Taken together, this body of work provides a solid basis for
the application of CT. Challenges for future research include developing testable hypotheses (e.g.,
about the generality of coordination mechanisms) and more structured approaches to evaluate and
choose between alternate coordination processes.