Organizing Business Knowledge The Mit Process Handbook [Electronic resources] نسخه متنی

2.3 Applying a Coordination Perspective

2.3.1 Approaches to Analyzing Coordination in Different Kinds of Systems

Any scientific theory (indeed, any statement about the world) must neglect some things, in order to focus on others. For example, Kling (1980) describes how different perspectives (e.g., rational, structural, and political) on the use of information systems in organizations each illuminate aspects of reality neglected by the others. In some situations, one or another of these perspectives may be most important, and all of them are involved to some degree in any real situation. In applying coordination theory to any particular system, it may therefore be necessary to consider many other factors as well.

For instance, in designing a new computer system to help people coordinate their work, ''details''about screen layout and response time may sometimes be as important as the basic functionality of the system, and the reputation of the manager who introduces the system in a particular organization may have more effect on the motivation of people to use it in that organization than any incentive structures designed into the system. Similarly, in designing a distributed processing computer system, the failure rates for different kinds of communications media and processors may be the primary design consideration, overwhelming any other considerations about how tasks are allocated among processors.

Parametric Analysis versus Baseline Analysis There are at least two ways an interdisciplinary theory can help deal with differences like these among systems: (1) parametric analysis and (2) baseline analysis.

PARAMETRIC ANALYSIS In parametric analysis the abstract theories include parameters which may be different for different kinds of systems. For instance, the principles of aerodynamics apply to both birds and airplanes, even though parameters such as size, weight, and energy expenditure are very different in the two kinds of systems. Similarly abstract models of coordination may include parameters for things like incentives, cognitive capacities, and communication costs which are very different in human, computational, and biological systems. Examples of models that have been applied to more than one kind of system in this way are summarized later in this section.

BASELINE ANALYSIS In baseline analysis one theory is used as a baseline for comparison to the actual behavior of a system, and deviations from the baseline are then explained with other theories. For example, in behavioral decision theory (e.g., Kahneman and Tversky 1973), mathematical decision theory is used to analyze the ways people actually make decisions. In the cases where people depart from the prescriptions of the normative mathematical theory, new theories are developed to explain the differences. Even though the original mathematical theory does not completely explain people's actual behavior, the anomalies explained by the new theories could not even have been recognized without a baseline theory for comparison. This suggests that an important part of coordination theory will be behavioral coordination theory in which careful observations of actual coordination in human systems are used to develop, test, and augment abstract models of coordination.

Identifying the Components of Coordination in a Situation In order to analyze a situation in terms of coordination, it is sometimes important to explicitly identify the components of coordination in that situation. According to our definition of coordination above, coordination means ''managing dependencies between activities.''Therefore, since activities must, in some sense, be performed by actors,''the definition implies that all instances of coordination include actors performing activities that are interdependent.^[3] It is also often useful to identify evaluation criteria for judging how well the dependencies are being ''managed.''For example, we can often identify some overall ''goals''of the activity (e.g., producing automobiles or printing a report) and other dimensions for evaluating how well those goals are being met (e.g., minimizing time or costs). Some coordination processes may be faster or more accurate than others, for instance, and the costs of more coordination are by no means always worthwhile.

It is important to realize that there is no single right way to identify these components of coordination in a situation. For instance, we may sometimes analyze everything that happens in a manufacturing division as one activity, while at other times, we may want to analyze each station on an assembly line as a separate activity. As another example, when we talk about muscular coordination, we implicitly regard different parts of the same person's body as separate actors performing separate activities.

CONFLICTING GOALS One important case of identifying evaluation criteria occurs when there are conflicting goals in a situation. In analyzing coordination in human organizations, it is often useful to simply ask people what their goals are and evaluate their behavior in terms of these criteria. However, some amount of goal conflict is nearly always present (e.g., Ciborra 1987; Williamson 1985; Schelling 1960), and people may be unable or unwilling to accurately report their goals, anyway. To understand these situations, it is often useful to both try to identify the conflicting goals and also to analyze the behavior of the system in terms of some overall evaluation criteria. For instance, different groups in a company may compete for resources and people, but this very competition may contribute to the company's overall ability to produce useful products (e.g., Kidder 1981).

Another important example of conflicting goals occurs in market transactions: as we saw above, all participants in a market might have the goal of maximizing their own individual benefits, but we, as observers, can evaluate the market as a coordination mechanism in terms of how well it satisfies overall criteria such as maximizing consumer utilities (e.g., Debreu 1959) or ''fairly''distributing economic resources.

Preview of Examples In the remainder of this section we describe examples of how concepts about coordination have been applied in three different areas: (1) understanding the new possibilities for human organizations and markets provided by information technology, (2) designing cooperative work tools, and (3) designing distributed and parallel computer systems. The early examples use very general notions of coordination; the later ones are more explicit in their identification of specific components of coordination.

This list is not intended to be a comprehensive list of all ways that theories of coordination could be applied. In fact most of the work we describe here did not explicitly use the term ''coordination theory.''We have chosen examples, however, to illustrate the wide range of applications for interdisciplinary theories about coordination.

2.3.2 Understanding the Effects of Information Technology on Organizations and Markets

Managers, organization theorists, and others have long been interested in how the widespread use of information technology (IT) may change the ways human organizations and markets will be structured (e.g., Leavitt and Whisler 1958; Simon 1976). One of the most important contributions of coordination theory may be to help understand these possibilities better.

To illustrate how the explicit study of coordination might help with this endeavor, we begin with a very general argument that does not depend on any of the detailed analyses of coordination we have seen so far in this chapter.^[4] Instead, this argument starts with the simple observation that coordination is itself an activity that has costs. Even though there are many other forces that may affect the way coordination is performed in organizations and markets (e.g., global competition, national culture, government regulation, and interest rates), one important factor is clearly its cost, and that is the focus of this argument. In particular, it seems quite plausible to assume that information technology is likely to significantly reduce the costs of certain kinds of coordination (e.g., Crawford 1982).

Now, using some elementary ideas from microeconomics about substitution and elasticity of demand, we can make some simple predictions about the possible effects of reducing coordination costs. It is useful to illustrate these effects by analogy with similar changes in the costs of transportation induced by the introduction of trains and automobiles:

A first-order effect of reducing transportation costs with trains and automobiles was simply some substitution of the new transportation technologies for the old: people began to ride on trains more and in horse-drawn carriages less.

A second-order effect of reducing transportation costs was to increase the amount of transportation used: people began to travel more when this could be done more cheaply and conveniently in trains than on foot.

Finally, a third-order effect was to allow the creation of more ''transportation-intensive''structures: people eventually began to live in distant suburbs and use shopping malls—both examples of new structures that depended on the widespread availability of cheap and convenient transportation.

Similarly we can expect several effects from using new information technologies to reduce the costs of coordination:

A first-order effect of reducing coordination costs with information technology may be to substitute information technology for some human coordination. For instance, many banks and insurance companies have substituted automated systems for large numbers of human clerks in their back offces. It has also long been commonplace to predict that computers will lead to the demise of middle management because the communication tasks performed by middle managers could be performed less expensively by computers (e.g., Leavitt and Whisler 1958). This prediction was not fulfilled for several decades after it was made, but many people believe that it finally began to happen with large numbers of middle management layoffs in the 1980s and 1990s.

A second-order effect of reducing coordination costs may be to increase the overall amount of coordination used. In some cases this may overwhelm the first order effect. For instance, in one case we studied, a computer conferencing system was used to help remove a layer of middle managers (see Crowston, Malone, and Lin 1987). Several years later, however, almost the same number of new positions (for different people at the same grade level) had been created for staff specialists in the corporate staff group, many of whom were helping to develop new computer systems. One interpretation of this outcome is that the managerial resources no longer needed for simple communication tasks could now be applied to more complex analysis tasks that would not previously have been undertaken.

A third-order effect of reducing coordination costs may be to encourage a shift toward the use of more ''coordination-intensive''structures. In other words, coordination structures that were previously too ''expensive''will now become more feasible and desirable. For example, as noted above, information technology can facilitate what some observers (e.g., Mintzberg 1979; Toffler 1970) have called adhocracies. Adhocracies are very flexible organizations, including many shifting project teams and highly decentralized networks of communication among relatively autonomous entrepreneurial groups. One of the disadvantages of adhocracies is that they require large amounts of unplanned communication and coordination throughout an organization. However, technologies such as electronic mail and computer conferencing can help reduce the costs of this communication, and advanced information sharing tools (e.g., Malone et al. 1987; Lotus 1989) may help make this communication more effective at much larger scales.

What might these new coordination-intensive structure be like? Let us consider recent work on two specific questions about the effects of information technology on organizations and markets: (1) How will IT affect the size of organizations? and (2) How will IT affect the degree of centralization of decision-making in organizations? This work does not focus explicitly on any specific dependencies. Instead, it compares two pairs of general coordination mechanisms that can manage many such dependencies: (1) market transactions versus internal decision-making with firms and (2) centralized versus decentralized managerial decisions.

Firm Size Malone, Yates, and Benjamin (1987) have used ideas from transaction cost theory to systematically analyze how information technology will affect firm size and, more generally, the use of markets as a coordination structure. They conclude that by reducing the costs of coordination, information technology may lead to an overall shift toward smaller firms and proportionately more use of markets—rather than internal decisions within firms—to coordinate economic activity.

This argument has two parts. First, since market transactions often have higher coordination costs than internal coordination (Williamson 1985; Malone, Yates, and Benjamin 1987), an overall reduction in the ''unit costs''of coordination should lead to markets becoming more desirable in situations where internal transactions were previously favored. This, in turn, should lead to less vertical integration and smaller firms.

For example, after the introduction of computerized airline reservation systems, the proportion of reservations made through travel agents (rather than by calling the airline directly) went from 35 to 70 percent. Thus the function of selling reservations was ''disintegrated''from the airlines and moved to a separate firm—the travel agents. Econometric analyses of the overall US economy in the period 1975 to 1985 are also consistent with these predictions: the use of information technology appears to be correlated with decreases in both firm size and vertical integration (Brynjolfsson et al. 1994).

If we extrapolate this trend to a possible long-run extreme, it leads us to speculate that we might see increasing use of ''firms''containing only one person. For instance, Malone and Rockart (1991) suggest that there may someday be electronic marketplaces of ''intellectual mercenaries''in which it is possible to electronically assemble ''overnight armies''of thousands of people who work for a few hours or days to solve a particular problem and then disband. Flexible arrangements like this might appeal especially to people who had a strong desire for autonomy—the freedom to choose their own hours and working situations.

Centralization of Decision-Making Gurbaxani and Whang (1991) have used ideas from agency theory to systematically analyze the effects on centralization of the reductions in coordination costs enabled by IT. They conclude that IT can lead to either centralization or decentralization, depending on how it is used. While this conclusion may not be surprising, the structure of their analysis helps us understand the factors involved more clearly: (1) When IT primarily reduces decision information costs, it leads to more centralization. For instance, the Otis elevator company used IT to centralize the reporting and dispatching functions of their customer service system, instead of having these functions distributed to numerous remote field offces (Stoddard 1986). (2) On the other hand, when IT primarily reduces agency costs, it leads to more decentralization. As used here, agency costs are the costs of employees not acting in the interests of the firm. For instance, when one insurance company developed a system that more effectively monitored their salespeople's overall performance, they were able to decentralize to the salespeople many of the decisions that had previously been made centrally (Bruns and McFarlan 1987). Overall, this bidirectional trend for IT and centralization is consistent with empirical studies of this question (Attewell and Rule 1984).

An alternative approach to this question is provided by (Danziger et al. 1982). In a sense this work can be considered a kind of ''behavioral coordination theory.''In studies of computerization decisions in forty-two local governments in the United States, they found that changes in centralization of power were not best explained any of the formal factors one might have expected. Instead, they found that since people who already have power influence computerization decisions, the new uses of computers tend to reinforce the existing power structure, increasing the power of those who already have it.

2.3.3 Designing Cooperative Work Tools

There has recently been a great deal of interest in designing computer tools to help people work together more effectively (e.g., Greif 1988; Johansen 1988; Ellis et al. 1991; Peterson 1986; Tatar 1988, 1990; additional references in table 2.3). Using terms such as ''computer-supported cooperative work''and ''groupware''these systems perform functions such as helping people collaborate on writing the same document, managing projects, keeping track of tasks, and finding, sorting, and prioritizing electronic messages. Other systems in this category help people display and manipulate information more effectively in face-to-face meetings and represent and share the rationales for group decisions.

In this section we will describe how ideas about coordination have been helpful in suggesting new systems, classifying systems, and analyzing how these systems are used.

Using Coordination Concepts from Other Disciplines to Suggest Design Ideas One way of generating new design ideas for cooperative work tools is to look to other disciplines that deal with coordination. For instance, even though the following authors did not explicitly use the term ''coordination theory,''they each used ideas about coordination from other disciplines to help develop cooperative work tools.

Using ideas from linguistics and philosophy about speech acts. Winograd and Flores (Flores et al. 1988; Winograd 1987; Winograd and Flores 1986) have developed a theoretical perspective for analyzing group action based heavily on ideas from linguistics (e.g., Searle 1975). This perspective emphasizes different kinds of speech acts, such as requests and commitments. For example, Winograd and Flores analyzed a generic ''conversation for action''in terms of the possible states and transitions involved when one actor performs a task at the request of another. An actor may respond to a request, for instance, by (1) promising to fulfill the request, (2) declining the request, (3) reporting that the request has already been completed, or (4) simply acknowledging that the request has been received. The analysis of this conversation type (and several others) provided a primary basis for designing the Coordinator, a computer-based cooperative work tool. For example, the Coordinator helps people make and keep track of requests and commitments to each other. It thus supports what we might call the ''mutual agreeing''part of the task assignment process.

Using ideas from artificial intelligence and organization theory about blackboards and adhocracies. Malone (1990) describes how ideas from artificial intelligence and organization theory combined to suggest a new tool for routing information within organizations. In the ''blackboard architecture,''program modules interact by searching a global blackboard for their inputs and posting their outputs on the same blackboard (Nii 1986; Erman et al. 1980). This provides very flexible patterns of communication between different program modules: any module can communicate with any other module, even when this interaction is not explicitly anticipated by the program designer. In adhocracies, as we saw above, just this kind of unplanned, highly decentralized communication is essential for rapidly responding to new situations (Mintzberg 1979; Toffler 1970). Stimulated, in part, by this need for an ''organizational blackboard,''Malone and colleagues designed the Information Lens system (Malone et al. 1987). A central component of this system is an ''anyone server''that lets people specify rules about what kinds of electronic messages they are interested in seeing. The system then uses these rules to route all nonprivate electronic messages to everyone in the organization who might want to see them. (To help people deal with large numbers of messages, another part of the system uses a different set of rules to sort and prioritize the messages people receive.)

Using ideas from philosophy and rhetoric about decision-making. Two cooperative work tools, gIBIS (Conklin and Begeman 1988) and Sibyl (Lee 1990), are designed to help groups of people make decisions more effectively. To do this, they explicitly represent the arguments (and counterarguments) for different alternatives a group might choose. Both these systems are based on ideas from philosophy and rhetoric about the logical structure of decision-making. For example, the basic elements in the gIBIS system (issues, positions, and arguments) are taken from a philosophical analysis of argumentation by Rittel (1970). The constructs for representing arguments in Sibyl are based on the work of philosophers like Toulmin (1958) and Rescher (1977).

Using ideas from computer science about parallel processes. Holt (1988) describes a theoretical language used for designing coordination tools that is based, in part, on ideas about Petri nets, a formalism used in computer science to represent process flows in distributed or parallel systems (Peterson 1981, 1977). This language is part of a larger theoretical framework called ''coordination mechanics''and has been used to design a ''coordination environment''to help people work together on computer networks.

SUMMARY OF EXAMPLES Clearly, ideas about coordination from other disciplines do not guarantee our developing useful cooperative work tools. However, we feel that considering these examples within the common framework of coordination theory provides two benefits: (1) it suggests that no one of these perspectives is the complete story, and (2) it suggests that we should look to previous work in various disciplines for more insights about coordination that could lead to new cooperative work tools.

A Taxonomy of Cooperative Work Tools As shown in table 2.3, the framework we have suggested for coordination provides a natural way of classifying existing cooperative work systems according to the coordination processes they support. Some of these systems primarily emphasize a single coordination-related process. For instance, electronic mail systems primarily support the message transport part of communication, and meeting scheduling tools primarily support the synchronization process (i.e., arranging for several people to attend a meeting at the same time). There is a sense, of course, in which each of these systems also support other processes (e.g., a simple electronic mail system can be used to assign tasks), but we have categorized the systems here according to the processes they explicitly emphasize.

Some of the systems also explicitly support several processes. For example, the Information Lens system supports both the communication routing process (by rules that distribute messages to interested people) and a form of resource allocation process (by helping people prioritize their own activities using rules that sort messages they receive). And the Polymer system helps people decompose goals into tasks and sequence the tasks (e.g., to prepare a monthly report, first gather the project reports and then write a summary paragraph).

One possibility raised by this framework is that it might help identify new opportunities for cooperative work tools. For instance, the Coordinator focuses on supporting one part of the task assignment process (mutual agreement on commitments). However, it does not provide much help for the earlier part of the process involving selecting an actor to perform the task in the first place (see section 2.3). New tools, such as an ''electronic yellow pages''or bidding schemes like those suggested by Turoff (1983) and Malone (1987) might be useful for this purpose.

Another intriguing possibility suggested by this framework is that it might be possible to implement ''primitives''for a number of different coordination-related processes in the same environment, and then let people combine these primitives in various ways to help solve particular coordination problems. This is one of the goals of the Oval system (Malone et al. 1992; Lai et al. 1988).

Analyzing Incentives for Using Cooperative Work Tools Another use for coordination theory in designing cooperative work tools can be to help systematically evaluate proposed or actual systems. For example, Markus and Connolly (1990) systematically analyze how the payoffs to individual users of a cooperative work system depend on how many other people are using the system. They do this by using an economic model from Schelling (1978) to extend Grudin's (1988) insights about the incentives to use cooperative work systems. For instance, on-line calendars and many other cooperative work applications involve ''discretionary databases''which users can view or update as they see fit. For each individual user, however, the benefits of viewing the database can be obtained without contributing anything. Thus it is often in the interests of each individual user to use the database without making the effort required to contribute to it. Unfortunately, the equilibrium state of a system like this is for no one to ever contribute anything!

An interesting empirical illustration of this phenomenon occurred in a study of how one large consulting firm used the Lotus Notes group conferencing system. In this study Orlikowski (1992) found that there were surprising inconsistencies between the intended uses of the system and the actual incentives in the organization. For instance, Orlikowski observed that this organization (like many others) was one in which people were rewarded for being the ''expert''on something—for knowing things that others did not. Should we be surprised, therefore, that many people were reluctant to spend much effort putting the things they knew into a database where everyone else could easily see them?

These observations do not, of course, mean that conferencing systems like this one cannot be useful in organizations. What they do mean, however, is that we must sometimes be sensitive to very subtle issues about things like incentives and organizational culture in order to obtain the full benefits of such systems. For instance, it might be desirable in this organization to include, as part of an employee's performance appraisal, a record of how often their contributions to the Notes database were used by other people in the organization.

2.3.4 Designing Distributed and Parallel Processing Computer Systems

Much recent activity in computer science has involved exploring a variety of distributed and parallel processing computer architectures. In many ways physically connecting the processors to each other is easy compared to the diffculty of coordinating the activities of many different processors working on different aspects of the same problem.

In this section we describe examples of work that has addressed these issues in an explicitly interdisciplinary way, drawing on insights from other disciplines or kinds of systems to design or analyze distributed or parallel computer systems. In particular, we consider examples of (1) analogies with social and biological systems as a source of design ideas and (2) quantitative tools for analyzing alternative designs.

Analogies with Social and Biological Systems as a Source of Design Ideas Competitive bidding markets for resource allocation. One of the basic problems in designing distributed or parallel computer systems is how to assign tasks to processors, and several distributed computer systems have addressed this problem with competitive bidding mechanisms based on analogies with human markets. For example, the Contract Nets protocol (Smith and Davis 1981; Davis and Smith 1983) formalizes a sequence of messages to be exchanged by computer processors sharing tasks in a network. The ''contracts''are arbitrary computational tasks that can potentially be performed by any of a number of processors on the network, the ''clients''are machines at which these tasks originate, and the ''contractors''are machines that might process the tasks (i.e., the servers). The sequence of announcement, bid, and award messages used by this protocol was described above in our analysis of the task assignment process (section 2.3). One of the desirable features of this system is its great degree of decentralization and the flexibility it provides for how both clients and contractors can make their decisions. For instance, clients may select contractors on the basis of estimated completion time or the presence of specialized data; contractors may select tasks to bid on based on the size of the task or how long the task has been waiting.

Using these or similar ideas, a number of other bidding systems have been developed (e.g., Stankovic 1985; Kurose and Simha 1989). For instance, several bidding systems have been developed to allow personal workstations connected by a local area network to share tasks (Malone et al. 1988; Waldspurger et al. 1988). In this way users can take advantage of the unused processing capacity at idle workstations elsewhere on the network. Furthermore the local bidding ''negotiations''can result in globally coherent processor scheduling according to various priorities (e.g., Malone et al. 1988). (For a review of several related systems and an analysis of a variety of bidding algorithms, see Drexler and Miller 1988; Miller and Drexler 1988.)

The notion of competitive bidding markets has also been suggested as a technique for storage management by Miller and Drexler (Miller and Drexler 1988; Drexler and Miller 1988). In their proposal, when object A wishes to maintain a pointer to object B, object A pays ''rent''to the ''landlord''of the space in which object B is stored. These rents are determined by competitive bidding, and when an object fails to pay rent, it is ''evicted''(i.e., garbage collected). Their proposal includes various schemes for how to determine rents, how to pass rents along a chain of references, and how to keep track of the various costs and payments without excessive overhead. They conclude that this proposal is not likely to be practical for small-scale storage management (e.g., garbage collection of individual Lisp cells), but that it may well be useful for sharing large objects in complex networks that cross ''trust boundaries'' (e.g., interorganizational networks). The scheme also appears useful for managing local caching and the migration of objects between different forms of short-term and long-term storage.

''Scientific Communities''for Information Routing and Resource Allocation Another central problem that arises in distributed and parallel processing systems is how and when to route information between processors. For instance, one interesting example of this problem arises in artificial intelligence programs that search a large space of possibilities, whose nature is not well known in advance. It is particularly useful, in this case, for processors to exchange information about intermediate results in such a way that each processor can avoid performing work that is rendered unnecessary by work already done elsewhere.

One solution to this problem is suggested by the Scientific Community Metaphor embodied in the Ether system (Kornfeld and Hewitt 1981; Kornfeld 1982). In this system, there are a number of ''sprites,''each analogous to an individual scientist, that operate in parallel and interact through a global database. Each sprite requires certain conditions to be true in the global database before it is ''triggered.''When a sprite is triggered, it may (1) compute new results that are added to the global database, (2) create new sprites that await conditions that will trigger them, or (3) stifle a collection of sprites whose work is now known to be unnecessary. In one example use of this system, Kornfeld (1982) shows how sharing intermediate results in this way can dramatically improve the time performance of an algorithm (even if it is executed by time-sharing a single processor). He calls this effect ''combinatorial implosion.''

This system also uses the scientific community metaphor to suggest a solution to the resource allocation problem for processors. Each sprite is ''supported''by a ''sponsor,''and without a sponsor, a sprite will not receive any processing time to do its work. For instance, a sponsor may sometimes support both work directed toward proving some proposition and also work directed toward proving the negation of the proposition. Whenever one of these lines of work is successful, support is withdrawn from the other.

Analyzing Stability Properties of Resource Allocation Algorithms Another way of applying coordination concepts is to help evaluate alternative designs of distributed and parallel processing computer systems. For instance, Huberman and his colleagues (Huberman and Hogg 1988; Lumer and Huberman 1990) have applied mathematical techniques like those used in chaos theory to analyze the dynamic behavior of distributed computer networks. In one case they analyze, for example, heavily loaded processors in a network transfer tasks to more lightly loaded processors according to a probabilistic process. When any processor in such a system can exchange tasks with any other processor, the behavior of the system is unstable for large numbers of processors (e.g., more than twenty-one processors in a typical example). However, when the processors are grouped hierarchically into clusters that exchange tasks frequently among themselves and only occasionally with other clusters, the system remains stable for arbitrarily large numbers of processors. This hierarchical arrangement has the disadvantage that it takes a long time to reach stability. In an intriguing analogy with human organizations, however, Huberman and his colleagues find that this disadvantage can be eliminated by having a few ''lateral links''between different clusters in the hierarchy (Lumer and Huberman 1990).

2.3.5 Summary of Applications

As summarized in table 2.4, the examples we have described show how a coordination perspective can help (1) analyze alternative designs and (2) suggest new design ideas. In each case these applications depended upon interdisciplinary use of theories or concepts about coordination.

^[3]See Baligh and Burton (1981), Baligh (1986), Barnard (1964), Malone (1987), Malone and Smith (1988), McGrath (1984), and Mintzberg (1979) for related decompositions of coordination.

^[4]See Malone (1992) and Malone and Rockart (1991) for more detailed versions of the argument in this section.