To what extent, and to what end can we effectively and intentionally design the processes and organization of a small firm, in particular, a virtual organization with its members distributed in multiple locations worldwide? What are the considerations for articulating the values and even “organizational style” for the firm, as a designed whole? For a loosely-organized consulting firm with a distributed membership, achieving growth presents significant challenges, such as the extent of formal process used, or the coordination of business development. The research served to define these commitments of the enterprise, as well as providing an opportunity for the members of a knowledge-creating business to participate in the redesign of the firm. In these respects, the intent of this research was not only to study the firm, but to design it.

Purpose

The study investigated the organizational design for a rapidly evolving consulting firm, with the intended result of developing a design plan and recommendations for action and research. In advance of the research, two areas of focus were considered fundamental to the design of an organizational architecture: Organizational communications and management of knowledge assets. The organizational architecture itself, and the process for constructing it, was considered a critical focus of the study.  Organizational architecture has received attention in the business research literature in recent years (Sotto, 1997, Bloomfield and Vurdubakis, 1997, von Krogh and Roos 1995), and provides a theoretical approach for organizational design.

Three methodological problems are investigated and elaborated in this research, relevant to organizational theory, the effective design of business processes, and to the emergence of knowledge issues in information technology. These are:

1. The evolution of Interactive Management methodologies in their application to developing requirements for organizational systems, and for constructing the organizational ontology.

2. The adaptation of this design process to communications systems and for knowledge management systems.

3. The process of representing the structure and processes of a knowledge-based enterprise as an organizational ontology.

This research was designed to support organizational design for supporting knowledge management. The application of Interactive Management (IM) methodology (Warfield, 1990, 1994) was adapted, by extending the CogniScope process into organizational research. This research approach builds upon a significant base of systems theory, enabling cross-disciplinary applications. Given interest in adapting interdisciplinary methods to business research and knowledge management, this research methodology and case study offer potentially useful findings and theoretical interpretations.

The CogniScope methodology (Christakis and Conaway, 1995, Banathy, 1996), a process used for over 10 years by Interactive Management practitioners in complex systems planning, was adapted to strategic planning and organizational design of the firm. This unique application of the CogniScope methodology begins to address some significant problems emerging in the organizational and technology literatures:

• The unavailability of reliable methods for consensus-based strategic, long-range, or visionary planning.
  

• The unavailability of specific organizational planning methods useful for non-traditional enterprises, such as small consulting firms, distributed or “virtual” organizations, and technology start-ups.
  

• Establishing appropriate directives for business processes in these firms, as few procedures exist from which to derive and share knowledge for applicable work practices.
  

• Developing methodologies for capturing and representing organizational knowledge to serve as guidance for planning and design of the organization, its products and services.

The CogniScope methodology addresses the methodology problems, by providing a research-validated process for generating collective knowledge and gaining consensus on the directives for of an organization and its service portfolio. An organizational ontology defined in this study provides a starting point for constituting enterprise knowledge as a shared representation, designed as an explicit framework with requirements that serve the strategies, visions, and values of all stakeholders. As in its use in the knowledge sciences, the ontology becomes the foundation for development of the organization and the architecture that supports its public and private representations.

The Research Context

Ontological design for organizational effectiveness has its genesis in the work of Flores (1982), who formulated a design theory for organizational communication based on a phenomenology of work practice as conversation networks. Flores drew from hermeneutics (Gadamer, 1976, Habermas, 1973) and Speech Act Theory (Searle, 1975) in developing the model for ontological design, later extended by Winograd (Winograd and Flores, 1986). These influences remain significant in the current research, and have furthermore enabled a re-evaluation of methodological assumptions in the CogniScope process for Interactive Management.

Interactive Management (IM) methods (Warfield, 1990) have traditionally been applied to large-scale social systems design problems, ranging from the complexity of natural resources management (Christakis, 1985) to defense technology planning, with hundreds of documented and published cases (Warfield, 1994). The full process of IM as described by Warfield and Cardenas (1994) provides a structure for facilitating participants in group design to integrate pluralistic views among all members in the service of establishing designs and plans for complex system problems that cannot be resolved through conventional design processes.

This framework for communications and knowledge management is presented in the form of an ontology for organizational design, as it establishes distinctions of the CWA business context as a designed, purposeful foundation for future committed action. Not as a plan, which would merely extend a roadmap of desirable goals and activities, but as an ontological basis for alignment of goals, coordination of action among dispersed collaborators, and social organization as an organization. This approach draws from Flores and Winograd (1986):

“The most important designing is ontological. It constitutes an intervention in the background of our heritage, growing out of our already-existent ways of being in the world, and deeply affecting the kinds of beings that we are. … The ontological designer needs to begin with an orientation to the kinds of questioning that will unconceal the ontology of the work, rather than posing technical questions about what the equipment can do.” (p. 163, emphasis mine).

An organizational ontology becomes generated from the research question “what are the organization’s foundations for values, action, and choice?” and the design question “what should constitute our shared future?” By focusing on these questions, we are designing the future business from goals, philosophies, intentions, and values shared by all the stakeholders. This process can be considered to use Interactive Management process to redesign IM practice.

Overview of the Case Study

CWA Ltd. practices Interactive Management for complex client problems, developing collaborative plans and designs for applications such as long-range strategic planning, enterprise process redesign, and new technology product development.

The intention of the communications framework was to design and deploy group-based communications tools enabling effective business and personal relationships within the CWA community of consultants, customers, and partners. In conducting interviews and initial investigations, CWA members discussed and projected their future work plans, involving several critical considerations:

• CWA’s work is of a highly collaborative nature, requiring effective and timely contribution from assigned members during the conduct of a normal engagement.
  

• CWA’s associates remain widely distributed across North America, and work together over distances to accomplish joint projects and to coordinate for client-based work.
  

• ·CWA’s current communications infrastructure consists of individual email mailboxes and a central administrative individual that coordinates individual commitments as necessary. Knowledge and learning are not effectively shared using these facilities, and coordination of project activities remains a significant undertaking.

The communications framework was established to reveal potentials for increasing cooperation and fostering true joint work and knowledge creation throughout the company. 

Research Method

A research and design program was initiated to investigate the space for social and system design, with the intention of developing a set of requirements and guidelines for software applications supporting communications and knowledge management. We applied a hybrid research method, using qualitative approaches (semi-structured interview, content analysis, hermeneutic interpretation) and quantitative (interpretive structural modeling and paired comparisons of candidates, based on the CogniScope) to triangulate investigation in the domain of interest. This research approach enabled open and inductive inquiry (non-biased with respect to theoretical outcome) to elicit salient drivers of the envisioned business of CWA. It also enabled us to follow the inductive process with deductive theory generation, to articulate a theory for organization of communication and knowledge processes grounded in observed activity and followed by a theoretical foundation, presented here for further evaluation.

Action Research

Action research presents an effective methodology for conducting research and evaluating theoretical proposals in organizations. Through iterative development of theory and testing it in practice, action research applies theory in the practice environment. Checkland (1991) and others (Avison, Lau, Myers, and Nielsen, 1999) have identified the need to declare an explicit methodology to establish validity in action research, and to distinguish it from consultancy in its practice. The CogniScope method serves this need with an explicit methodology and hundreds of documented applications (Warfield, 1994).

Use of the CogniScope Process

Banathy (1996) documents the CogniScope system as the first design inquiry process that “applies a thoroughly tested cognitive technology that integrates software as ‘groupware’ in the work of a designing community (p. 149).” The CogniSystem software differs from decision support systems in its design for a specific, scientifically-grounded process. Christakis and Conaway (1995) describe the CogniScope process as a disciplined inquiry enabling a community of stakeholders to construct a design for a complex social system, and an action plan based on the design. The process integrates the pluralistic views of the stakeholders, reaching consensus with a minimum of compromise.

Using CWA’s own tool, the CogniSystem, for this purpose adds value to the renewal of the inquiry process itself, while pursuing the analysis and design of business directives.

Figure 1 - Design research methodology

Figure 1 outlines the four stage research process adopted for the design plan:

1.       Understanding the Design Space. To investigate the space for design, we conducted background research into business processes, interviewed key stakeholders, and reviewed case studies.

2.       Inductive, qualitative inquiry. Research questions generated assertions defining organizational requirements. Statements were documented in the CogniSystem software, using ISM to structure the inquiry, and not to evaluate the model.

3.       Deductive, theoretical inquiry. By deriving categories inductively, we established a valid set of directives. We compared directives, and derived categories supporting the object dimensions of relationship and assets.

4.       Synthesis of design plan. By evaluating categories and conducting an influence analysis, we determined key drivers for the business ontology. From the final set of influences and directives, we synthesized principles and guidelines.

Ozbekan (1969) was perhaps the first to establish the distinctions in the type of planning and design pursued by this inquiry. He described the problem of idealization, the definition of the “ideal vision,” as the ontological challenge faced by design researchers. The corollary problem faced with each instance of this type of research was epistemological, described as “designing the designing system.” For the organizational ontology, the idealization problem was establishing what the organization should become, its ideal state reflecting stated mission, goals, and values. Establishing ontological validity requires the explication of the epistemology, the methodology used to construct the model.

Establishing validity criteria requires reference to assumptions underlying the inquiry method, since these methods are not canonical. Their adequacy must be explained with respect to component methods used, control of bias, selection of sample, and the research question. The epistemological challenge is then defining the appropriate “design for designing” that ensures adequacy of the models from which system design and organizational planning will ensue.

Ontological design requires extensive reflection on the assumptions used in design, the object and goals of design, and reflection on the epistemology, including the background or history of the methods and beliefs affecting the design process. In ontological design, the assumptions and beliefs of primary sources are investigated, and their influence is questioned. There is also the assumption, built into the ISM inference engine, that relationships among constituent propositions can be derived from the sequence of prior related decisions on other propositions. The underlying mechanisms of this operation are not revealed or explained to allow participants visibility into the process. And finally, the consensus models are built in a “big bang” fashion; even though all the component propositions are revealed and discussed, the rationale for prior decisions are not typically reevaluated, and relationships that emerge among propositions are not usually opened up for critique. Therefore, the possibility for deep participation at the level of interpretation can be questioned, revealing a positivist bias that following the validated process will achieve results in a predictable fashion.

It could be considered that this embedded causality in the model could propagate throughout a design plan, and that these unreflected biases could diminish the possibility for emancipatory communicative action. This epistemological tradition should therefore be questioned, in the pursuit to adopt methodologies for knowledge creation and use consistent with a social constructivist view shared by researchers and even purported by the ontology. Therefore, the research supporting the development of propositions used in the ontology adopted methods from discourse analysis, argumentation, and the linguistics of communicative action as opposed to analytic methods of decomposition, rule-based inference, or formal methods.

Development of the Ontological Model

Ontology Development in Other Domains

This brief background demonstrates support for the CogniScope as a methodology for ontological construction. What are ontologies and how are they used? Research activity denoting studies as “ontological” was primarily found in artificial intelligence research.

Few references support the case for developing organizational ontology. Uschold and Gruninger (1996) discussed ontologies in software engineering, defining the term as a “an explicit account of a shared understanding,” with utility in understanding of domains of expertise, and in improving organizational communication in managing the development process. Their use of ontology refers to a knowledge/object/data model that establishes definitions of shared components used in software engineering, thus reducing communicative ambiguity. Likewise, Fox, Barbuceanu, and Gruninger (1996) describe the preliminary development of an organization ontology for simulating industrial operations, modeling management and behavior.

Faro and Giordano (1998) support this approach in reusable ontologies for evolutionary design, and Chao, et al (1998) show a similar support in techniques for reusability and knowledge sharing in engineering design for automated systems. Mihoubi, Simonet, and Simonet (1998) present reuse ontologies as taxonomies or shared vocabularies for knowledge system development. They define ontologies as representing “explicit specification of a conceptualization that is expressible as a meta-level viewpoint on a set of possible domain theories for the purpose of modular design, redesign, and reuse of knowledge-intensive system components” (p. 367). This definition can be applied to a knowledge-based organization ontology, as a socially-constructed knowledge system.

In the domain of geographical information systems, Besio, et al (1998) used semantic nets to simulate human conceptual organization in their study of risk maps:

“The management of information is made easier using the same association processes present in human mental organisation (formal ontology). Thus, a semantic net links environmental maps (considered at the same level as `human concepts') and roles (which work in the same way of `interrelation of concepts' in a human mind).”

Semantic networks relate ontologies to specifications of knowledge as a shared cognitive schema, as found in knowledge engineering. Gaines (1988) anticipated these constructs and defined a framework in which ontologies represent a formal system of distinctions, based on abstraction. Gaines suggested that the ontology enabled inference by reasoning against the ontology structure, that the formal distinctions supported inference (and therefore planning) against the system of distinctions. In essence, an organizational ontology enables knowledge-based decision support based on core values and business model propositions.

The CogniScope method enabled the development of ontology for organizational design and strategic “planning” that meets the criteria afforded by these prior studies. It clearly meets Uschold and Gruninger’s criteria (1996) of an “explicit account of a shared understanding,” and defines a shareable taxonomy of concepts significant to defining organizational direction, values, and actions consistent with stewardship of relationships and assets. The representational models generated from the CogniScope method include narratives and discourse models, structured proposition models, and structural diagrams of relationships. Compared to the narrowly-defined ontologies used in knowledge engineering (taxonomies of reusable definitions), the rich representations generated by the CogniScope seem to better meet the broader import of the philosophical usage of the term.

Although the focus on the research was on developing the organizational ontology, the epistemological foundation must be positioned appropriately. An organizational epistemology (von Krogh and Roos, 1995) defines methodologies for interpreting linguistic domains accepted within organizations as valid ways of developing and construing knowledge. As in the CogniScope revisions, von Krogh and Roos propose the use of arguments in conversations, which they define as functional, temporal, value and intertwined. As instantiated by Toulmin’s (1956) theory of argumentation, arguments are supported by warrants. von Krogh and Roos propose three types of warrants: definitional warrants that establish the basis of language, propositional warrants that establish the relationships between cause and effects and paradigmatic warrants that define the overall purpose of the organizations. The organizational epistemology in theory appears similar to the ontological approach. A central difference is the epistemology is process-oriented, and can be developed and even selected as necessary to serve different research domains. Ontologies define a domain’s foundation, and are altered frequently, unless the fundamentals of the domain shift. The paradigmatic warrants proposed by von Krogh and Roos seem similar to ontological constructs. Warrants to the organizational purpose will emerge from an implicit ontology. In this research, the paradigmatic warrants were made visible, and were designed into the organizational ontology.

The Framework for the Organizational Ontology

The purpose of the original communications and knowledge management framework was to design systems for optimal business communications, and to define the support systems for knowledge acquisition, encoding, exchange, and valuation. This framework was designed to establish reusable organizational knowledge, and to support consistent knowledge exchange across the distributed network of associates, partners, and clients.

Growing the virtual enterprise in a coordinated direction was one of the key drivers for conducting this action research - but there is very little business research applicable directly to small, knowledge-based virtual enterprises. A virtual enterprise can be considered a network of commitments managed by decentralized decision making and project-oriented subteams. According to Charan (1991), the network is characterized by a social architecture that differs from organizational structure. ‘Social architecture refers to the operating mechanisms through which key managers make trade-offs and to the flow of information, power, and trust among these managers that shapes how those trade-offs get made.” It is concerned with the “intensity, substance, output, and quality of interactions,” all characteristics of knowledge creation, and with “the frequency and character of dialogue among members on a day-to-day basis,” or the organizational communications. Charan specifies the need to design a new social architecture as a function of senior management (Charan, p. 107).

To ground this research into the enterprise, the ontology definition focused on the following questions:

·         Internal strategy: What are the directives, tools, and practices for effective distributed organizational communication and customer communication?

·         Knowledge strategy: What are the directives, tools, and practices for developing the platform for knowledge creation, reuse, sharing, and integration?

·         External strategy: How should the organization be designed for sustainable competitive advantage, to support the social and business practices necessary in a distributed (virtual) environment?

CogniScope sessions were employed for each inquiry, contributing to the directives for communications and knowledge management. Rather than addressing communications and knowledge management directly as a requirements analysis, our inquiry raised hypotheses that enabled a more fundamental redesign. As we explored the directives for communications, we discovered how this domain is significantly, if not essentially based on relationships. Each type of communication pre-supposes a relationship, and even roles and stakes in the relationship, that create a context. The business value of communication was therefore envisioned as stemming from the stewarding of various relationships. Likewise, knowledge management emerged from the valuation of knowledge as a fundamental business asset, relevant for a knowledge and skill-oriented profession such as Interactive Management. The directives emerging from the inquiry into knowledge value were then aggregated into the parent domain of asset value and asset management. The stewardship orientation established principles and guidelines for design planning, and implementation of two domains described below.

·         Stewardship of Relationships - Fostering the network of organizational commitments, including the social network of individuals and agencies with which the organization has business or relationship commitments. This framework was developed from inquiry into the social space for design.

·         Stewardship of Assets - Primarily fostering knowledge value assets. This framework was developed from an inquiry into the space of value possibilities, which established the grounds for valuing particular types of intellectual assets, knowledge and competencies, and other assets.

Two separate frameworks for design were initially developed, for communication and for knowledge management. These were generated from the inquiry process previously described in the Method section.

Development of the Ontology Using the CogniScope Method

A specific “trigger question” initiated the inquiry into both communications and knowledge management.

“What directives should be pursued relentlessly by the core CWA Associates, by the year 2000, to achieve personal fulfillment and enterprise viability?”

This question generated eleven key propositions. These were supported by documented assumptions and warrants. Assumptions described specific factors affecting function of the statement in practice. Warrants were established as design rationale for the statement with respect to its significance. To inductively generate clusters of related directives, these propositions were associated by pairwise comparison using by the next trigger question.

“Does the directive X have significant attributes in common with directive Y, (within the context of what should be pursued by 2000)?”

Three clusters were grouped, and were named as follows:

1.       Infrastructure for Relationship Building

This cluster was named to represent the communications infrastructure required to build and maintain relationships within the business network.

2.       Opportunities for Experiential Development

This cluster was generated from only two associated statements, which stood apart from the other clusters. However, it represents the emphasis in IM practice of requiring customers and stakeholders to directly experience the method to apprehend its full value.

3.       Acceptance Deployment

This refers to the requirement for IM practitioners to more effectively communicate the findings of IM workshops and to introduce the method into other domains to widen its utilization.

The next step in the method was to investigate the supporting influences within the directives using the CogniSystem’s interpretive structural modeling algorithm. A pairwise comparison evaluated the question:

“Would demonstrable progress on the directive X significantly improve progress on the directive Y by (year) 2000?”

This evaluation resulted in a three level structure, represented as a tree-type network, a directed graph showing the key driver influence and the relationships between each supporting influence. The MICMAC algorithm (also in the CogniSystem) analyzed cross-impact strength of directives in the influence network. Running MICMAC identified four independent drivers, influential assertions demonstrating independence from the others tested.

The action research process revealed knowledge not apparent in the initial inquiry, through iteration of the two trial frameworks. Stewardship was revealed as the activity for each dimension, with members of the organization considered as responsible stewards of these critical dimensions. The purpose of the activity of communication was evaluated as serving our network of relationships, with relationships defined as the object. The purpose of knowledge management was stewardship of value, the activity of stewarding the object of assets.

The CogniScope process used the following steps for relating the two dimensions to each other for the ontology. These steps are consistent with the standard form for Interactive Management workshops for complex system design planning (Warfield and Cardenas, 1994).

1.  Developing a Trigger Question

The first step was to define the appropriate trigger question that would effectively relate the two dimensions to each other without biasing either dimension. The following question was used:

Is there (always at least) one end event in enacting (the directive from A, Stewarding Relationships) which is necessary to enact (at least always) one end event in the directive B (from Stewarding Assets)?

This question assumed directionality from relationships to assets, so that the activities of business relationships were focused on enabling the stewardship of assets. 

2.  Generation of Structures

The sets of propositions from each dimension were compared pairwise using the ISM algorithm (between dimensions only), following the trigger question. This revealed the relationships between stewardship of assets and relationships, based on the following two analyses:

1.       Derivation of the path of influences among the directives shown within the influence structure.

2.       Identification of the most influential drivers in this network, based on 1) depth or level in the influence structure, and 2) the number of other directives dependent on the fulfillment of the lower-level driver.

Figure 3 portrays the actual CogniSystem network derived from these analyses, showing the influence path and the levels of influence in the structure.

3.  Analysis of Structures

The analysis of structures revealed the relationships between the two dimensions based on comparisons using the trigger question. A six levels structure was produced, with multiple paths targeting the same few influential drivers. The network produced by the CogniSystem is shown in Figure 3, showing a small set of drivers influencing the network. These are the primary propositions the organizational ontology should be constructed to support, as fulfillment of these will ensure successful implementation of the related propositions.

Although this report does not permit describing every object and the influences as documented in the CWA final report, some description of the “deep drivers” and their relationships to the network will enable sufficient understanding of the method, if not the ontology. The lowest level of the network (VI) exerts the deepest influence, and the ISM rules define the path of influence starting from this lowest level. For this analysis, which compared and interrelated two dimensions overall, the result was that the two deepest drivers shared equal influence, both from the relationships dimension (B), labeled as 8(B) and 13(B). Both of these drivers exerted equal influence on the next two, which are considered two members of a cyclical relationship (in the CogniSystem these two are combined into a single entity, indicating a “cycle”).

 
Figure 3 - Graphical representation of ISM influence structure

The directed arcs from each entity indicate the influence derived from the comparisons of one proposition to another. In some cases the influence path indicates the strongest influence being between entities of the same dimension (A => A, B => B). Each level in the model establishes a cluster of similar degree of influence with respect to other levels. That is, the influence is measurably strongest at the lowest level (VI), with successively less influence proceeding up to Level I. Each entity has one influence arc to another entity, and it becomes clear with even cursory observation that some entities, although not deep drivers, are key receivers of influence (arcs from multiple entities to one) and are critical to understanding the network.

Conclusions

Designing an organization to support effective knowledge management requires participation of the community of contributing stakeholders. Interactive Management provides an action research method for pluralist engagement of these stakeholders in a democratic approach to inquiry. Our research used the CogniScope approach to IM as a tool for eliciting organizational knowledge specifically for envisioning and planning a redesigned organization.

The research showed the CogniSystem tool was effective for structuring knowledge as propositions, and for evaluating these propositions to identify underlying relationships. These relationships show the deep structure inherent in the organization, which was used to design the organizational ontology. Guidelines for design of business applications were drawn from the ontology’s directives, relating requirements from the four dimensions developed in the inquiry. The proposed applications supported the communications and knowledge management systems based on directives in the ontology.

Due to length constraints, the two end products of the research, the ontology specification and guidelines for design, are not presented here. The key finding presented in this paper was the use of the CogniScope process and IM methodologies as valid techniques for eliciting organizational knowledge for structuring a consensus knowledge model for requirements and design.

Future purposes of organizational ontology are proposed as shared strategic models for planning, action, and requirements definition for knowledge systems. The CogniScope advances group inquiry using a structured reasoning tool that maintains both the process and results of analysis and planning rationale. This approach shows promise as a group research model for other types of knowledge elicitation, for design of interactive knowledge systems as well as dynamic organizational structures.

References

1.       Argyris, C. (1992). Seeking truth and actionable knowledge: How the scientific method inhibits both. In C. Argyris (Ed.), On organizational learning (pp. 286- 294). Cambridge: Blackwell Publishers.

2.       Avison, D., Lau, F., Myers, M., and Nielsen, P.A. (1999). Action research. Communications of the ACM, 42 (1), 94-97.

3.       Banathy, B.H. (1996). Designing social systems in a changing world. New York: Plenum Press.

4.       Besio, M., Ramella, A., Bobbe, A., Colombo, C., Olivieri and Persano, M. (1998). Risk maps: theoretical concepts and techniques. Journal of Hazardous Materials, 61(1-3), 299-304.

5.       Bloomfield, B.P. and Vurdubakis, T. (1997). Visions of organization and organizations of vision: The representational practices of information systems development. Accounting, Organizations and Society, 22 (7), 639-668.

6.       Braa, K. and Vidgen, R. (1997). An information systems research framework for the organizational laboratory. In Kyng and Mathiassen (eds.) Computers and design in context. Cambridge, Mass: MIT Press.

7.       Chao, K-M., Smith, P., Hills, W., Florida-James, B., and Norman, P. (1998). Knowledge sharing and reuse for engineering design integration. Expert Systems with Applications, 14,  399-408.

8.       Charan, R. (1991). How networks reshape organizations - for results. Harvard Business Review, Sept.-Oct., 104- 115.

9.       Checkland, P. (1991). From framework through experience to learning: The essential nature of action research. In Nissen, H-E., Klein, H.K., and Hirschheim, R. (eds.) Information systems research: Contemporary approaches and emergent traditions. Amsterdam: Elsevier.

10.   Christakis, A.N. and Shearer, W. L. (1997). Collaboration through communicative action: Resolving the systems dilemma through the CogniScope system approach. Philadelphia: Christakis, Whitehouse and Associates.

11.   Christakis, A.N. and Conaway, D. (1995). Building high-performance project teams: Learning about the CogniScope. Paoli, Penn: CWA, Ltd.

12.   Faro, A. and Giordano, D. (1998). Concept formation from design cases: Why reusing experience and why not. Knowledge-Based Systems, 11, 437-448.

13.   Fischer, G., Lemke, A.C., McCall, R., and Morch, A.I. (1996). Making argumentation serve design. In Moran and Carroll (eds.) Design rationale. Mahwah, NJ: Lawrence Erlbaum Associates.

14.   Flores, C.F. (1982). Management and communication in the office of the future. Ph.D. dissertation. Berkeley: University of California.

15.   Fox, M.S., Barbuceanu, M., and Gruninger, M. (1996). Organization ontology for enterprise modeling: Preliminary concepts for linking structure and behaviour. Computers in Industry, 29, Issue 1-2, 123-134.

16.   Gaines, B.R. (1988). A conceptual framework for person-computer interaction in complex systems. IEEE Transactions on Systems, Man, and Cybernetics, 18 (4), 532- 541.

17.   Gadamer, H-G. (1976). Man and language, and semantics and hermeneutics. Philosophical hermeneutics. D.E Linge (Ed.) Berkeley, Cal: University of California Press.

18.   Habermas, J. (1973). Theory and practice. Boston: Beacon Press.

19.   Holtzblatt, K. and Beyer, H. (1998). Contextual Design: Defining customer-centered systems. San Francisco: Morgan Kaufmann Publishers.

20.   Mihoubi, H., Simonet, A., and Simonet, M. (1998). Towards a declarative approach for reusing domain ontologies. Information Systems, 23(6),  365-381.

21.   Ozbekan, H. (1969). Towards a general theory of planning. In E. Jantsch (ed.), Perspectives of planning. Paris: OECD Publications.

22.   Searle, J.R. (1979). Expression and Meaning: Studies in the Theory of Speech Acts. Cambridge: Cambridge University Press.

23.   Sotto, R. (1997). The virtual organisation. Accounting, Management and Information Technologies, 7(1), 37-51.

24.   Toulmin, S. (1958). The uses of argument. Cambridge, UK: Cambridge University Press.

25.   Uschold, M. and Gruninger, M. (1996). Ontologies: principles, methods and applications. Knowledge Engineering Review, 11(2), 93-136.

26.   von Krogh, G. and Roos, J. (1995). Organizational epistemology. New York: St. Martin's Press.

27.   Warfield, J.N. and Cardenas, A.R. (1994). A handbook of Interactive Management. Ames, Iowa: Iowa State University Press.

28.   Warfield, J.N. (1990). A science of generic design: Managing complexity through systems design. Ames, Iowa: Iowa State University Press.

29.   Warfield, J.N. (1976). Societal systems. New York: John Wiley and Sons.

30.   Winograd, T. and Flores, C.F. (1986). Understanding computers and cognition. Norwood, NJ: Ablex.

31.   Yasai-Ardekani, M. and Haug, R.S. (1997). Contextual determinants of strategic planning processes. Journal of Management Studies, 5(34), 729-741.