Designing an ICT solution for a multi-actor environment: the KMP project experimentation

The objective of this article is to show how the design methodology we experimented allows the creation of both actionable and theoretical knowledge and the reinforcement of both types of knowledge. Especially we intend to demonstrate that the design methodology we built enables to originate a generative dance between theoretical and actionable knowledge, and to solve practical problems in cross-disciplinary, heterogeneous communities from a range of organizations (firms and research laboratories). In order to demonstrate this proposal our article will lean on a specific design research conducted three years ago: the Knowledge Management Platform project (KMP project). Its objective was to design an ICT solution in a specific case of a multi-actor environment. More specifically, the KMP project aimed to build a semantic web service of competencies in order to enhance exchange and combination dynamics of knowledge within the Telecom cluster of Sophia Antipolis (Alpes-Maritimes, France) thanks to an interactive mapping of competencies. The argument is organized as follows. First, we explore design research methodologies in management studies and more specifically recent developments of ICT tools building. Indeed, designing an ICT solution requires a specific attention on use. This leads us to dissociate the designer world from user world. Thus, the key success of an innovation depends on the co-evolution and the convergence of design and use around boundary objects developed during this process. The argument then turns to the KMP experimentation. Finally, we expose and discuss the KMP results. The discussion firstly focuses on the emergence and the development of boundary objects. Then, we analyze how knowledge is both produced and memorized in these boundary objects. Pascal A. et Thomas C., “Designing an ICT solution for a multi-actor environment: the KMP experimenation”, The Third Organization Studies Summer Workshop: 'Organization Studies as Applied Science: The Generation and Use of Academic Knowledge about Organizations', 7–9 June 2007, Crete, Greece. 1 The gap between theory and practice is a persistent and difficult problem for academics and practitioners in the field of management. According to Van Aken (2004), there are serious doubts about the actual relevance of management theory as developed by the academic community. More generally, academic research has a serious utilization problem. Van de Ven and Johnson (2006) identify three ways in which the gap between theory and practice can be framed: first it is typically framed as a knowledge transfer problem; a second approach considers theory and practice as distinct but complementary kinds of knowledge; this leads to a third view “that the gap between theory and practice is a knowledge production problem” (ibid: p803). In this perspective, a new figure of “engaged scholarship”, in which researchers and practitioners coproduce knowledge has emerged in the 90’s (Hatchuel 1994; David 2000; Van Haken 2005; Van de Ven and Johson 2006). This new figure, called “mode 2 knowledge production” (Gibbons et al. 1994), is multidisciplinary and aims at solving complex and relevant field problems. Based on the seminal work of Simon (1969; 1996) and following this line of thought, many scholars suggest that “design” is an ideal-typical mode 2 of research (David 2000; Romme 2003; Van Aken 2005). David (2000) suggests that the central concern of management research should be a general problem of design. Design research develops knowledge at the service of action. The term design is chosen to underline the orientation on building knowledge to solve real-world problems. Hatchuel (2001) and Van de Ven and Johson (2006) emphasize that design research requires clear objectives and a careful arbitrage of the identities of roles of participants, rules of engagement and dissemination and use of study findings. The objective of this paper is to propose a methodology to implement a design research in the specific case of the building of an artifact. We argue that the construction of an artifact allows close interaction between scholars and practitioners, an emergent process of arbitrage in which researchers and practitioners engage with one another to coproduce the solutions and both theoretical and practical knowledge in the field of management. In order to demonstrate this proposal our article will lean on a specific design research conducted five years ago: the Knowledge Management Platform project (KMP project). Its objective was to design an ICT solution in a specific case of a multi-actor environment. More specifically, the KMP project aimed to build a semantic web service of competencies in order to enhance exchange and combination dynamics of knowledge within the Telecom cluster of Sophia Antipolis (Alpes-Maritimes, France) thanks to an interactive mapping of Pascal A. et Thomas C., “Designing an ICT solution for a multi-actor environment: the KMP experimenation”, The Third Organization Studies Summer Workshop: 'Organization Studies as Applied Science: The Generation and Use of Academic Knowledge about Organizations', 7–9 June 2007, Crete, Greece. 2 competencies. Indeed, this cluster had to face two main problems due to its specific multitechnological activity (micro-electronics, information technology and telecommunications): on the one hand, a problem of knowledge visibility and, on the other hand, a problem of missing link among widely distributed types of knowledge within the cluster. This project wins hands down from practitioners. At the present time, both practitioners and the PACA region support it and set it up in a pre-industrialization phase. In this perspective, we can say that this project is a successful design research, having produced actionable knowledge. However, this experimentation has shown that production of actionable knowledge is strongly linked with the creation of theoretical knowledge. In this perspective, the objective of this article is to show how the design methodology we experimented allowed the creation of both actionable and theoretical knowledge and the reinforcement of these two types of knowledge. More specifically, we intend to demonstrate that this design methodology enables to originate a generative dance between theoretical knowledge and actionable knowledge and to solve practical problems in cross-disciplinary, heterogeneous communities from a range of organizations (firms and research laboratories). The argument is organized as follows. First, we explore design research methodologies in management studies and more specifically recent developments of ICT tools building. Indeed, designing an ICT solution requires a specific attention on use. The argument then turns to the KMP experimentation. Finally, we expose and discuss the KMP results. This discussion focuses on the emergence and the development of boundary objects, and their role in the creation of both theoretical and actionable knowledge. 1. The design methodology: an iterative process of codesigning Design methodologies are specific modes of engaging in research. Especially, they are characterized by an active intervention process in the system in which the researchers act. Because academic studies propose distinct views on intervention process and address the issue of the model of intervention, we have chosen to construct our design methodology on two parts. Firstly, we focus on design research methodologies in order to identify the main steps for the intervention process. Secondly, we discuss the process of design in the specific case of an ICT solution. Pascal A. et Thomas C., “Designing an ICT solution for a multi-actor environment: the KMP experimenation”, The Third Organization Studies Summer Workshop: 'Organization Studies as Applied Science: The Generation and Use of Academic Knowledge about Organizations', 7–9 June 2007, Crete, Greece. 3 1.1. The process of design methodology: its main steps In this part, we discuss to different but complementary process design approaches which focus on a specific aspect of the intervention process: those of Hatchuel and Molet (1986) which are more practitioner oriented and those of Romme and Endenburg (2006) which are more theoretically oriented. 1.1.1. The intervention process: Hatchuel and Molet (1986) Hatchuel and Molet (1986) have distinguished five phases in the intervention process: (i) The “feeling of discomfort”: this step could also be called the perception of the issue. During this step, researchers and practitioners co-construct the issue. There is a confrontation between the researcher’s theoretical concepts and the empirical issue. (ii) Building a ‘rational myth’: it consists of the transformation of perceptions into concepts and data and of elaborating a theoretical model that allows the intervention. (iii) The experimental phase: intervention and interaction: the two preceding steps enable to insert a new stimulus into the organization. During this experimental step, practitioners’ reactions can induce modifications of both the tool and the organization. (iv) The inductive phase: portraying a set of logics. It consists in creating a learning process both for the researchers and the different practitioners about the organization’s local logics. (v) The change process: knowledge versus implementation. This process induces, thanks to the implementation of the tool, the crossed transformation of the organization and the tool and the development of knowledge on these two levels. It also allows a return, in a constructive way, to a new understanding of the first phase i.e. the perception of the issue. According to the writers’ point of view, intervening in an organization requires five steps that serve as guidelines for the researcher. These steps reveal in fact that the intervention calls for two main phases: firstly, to co-construct the issue with the practitioners with huge references on theory; secondly, to elaborate a tool that will support the intervention process. However, the link between these two main phases is not clearly explained. We propose now to refer to Romme and Endenburg (2006) in order to enrich the comprehension of the link between the concrete research project and the theory and/or the previous studies. Pascal A. et Thomas C., “Designing an ICT solution for a multi-actor environment: the KMP experimenation”, The Third Organization Studies Summer Workshop: 'Organization Studies as Applied Science: The Generation and Use of Academic Knowledge about Organizations', 7–9 June 2007, Crete, Greece. 4 1.1.2. The design process: Romme and Endenburg (2006) Romme and Endenburg (2006) suggest a science-based approach to organization design in five components: (i) Organization science is “the cumulative body of key concepts, theories, and experientially verified relationships useful for explaining organizational process and outcomes” (Romme and Endenburg 2006: 288). As such, these chunks of general knowledge provide the theoretical foundation for construction principles. (ii) Construction principles are a “set of imperative propositions, grounded in the state-of-the-art of organization science” (Romme and Endenburg 2006: 288). They serve as a body of knowledge that will guide the elaboration of design rules and so the specific design process at hand. Consequently, construction principles serve to bridge the descriptive nature of scientific propositions and the prescriptive nature of design rules. Obviously, as the set of construction principles may be borrowed from different subfields, it is important to establish that this set is coherent. (iii) Design rules are a “set of detailed guidelines for designing and realizing organizations, grounded in a related set of construction principles”. In other words, design rules are “elaborate solution-oriented guidelines for the design process” (Romme and Endenburg 2006: 288). Hence, they collectively constitute a scheme for the design and may therefore be developed as a coherent set of related rules. In a closer approach, Van Aken (2004; 2005) proposes the concept of ‘technological rules’, which he defined as “a chunk of a general knowledge linking an intervention or artifact with an expected outcome or performance in a certain field of application” (Van Aken 2005: 23). In his definition, Van Aken specifically uses the term ‘general’ in order to stress the idea that it is not a specific solution or a specific situation, but a general prescription for a class of problems. For that reason, we may say that a technological rule is a “mid-range theory”, whose validity is limited to a certain application domain. In the management field, these rules are not algorithmic but heuristic (Van Aken 2004). (iv) Organization design is the “developing representations of the intended organizational system being (re)designed with help of the design rules” (Romme and Endenburg 2006: 289). Thereby, organization design is resulting from the interaction between design rules, the contingencies of the organization and so of the design situation, and at least the preferences of the people engaging in the organization design effort. We may notice here that organization design can be a specific solution or an artifact. Pascal A. et Thomas C., “Designing an ICT solution for a multi-actor environment: the KMP experimenation”, The Third Organization Studies Summer Workshop: 'Organization Studies as Applied Science: The Generation and Use of Academic Knowledge about Organizations', 7–9 June 2007, Crete, Greece. 5 (v) Implementation and experimentation: during this step, the organization design or the artifact is implemented and the researcher evaluates the processes caused by that design. According to Hatchuel and Molet (1986), Romme and Endenburg (2006) stress the fact that the whole system ought to be regulated through a set of feedbacks. However, because they suggest preceding the design step by specifying what they call “construction principles” and “design rules”, these two authors offer a real complementary view. 1.2. Designing an ICT solution: a uses approach This section expresses in more detail the design process of an ICT solution. By the way, it explores the interaction between technology and social system, i.e. uses. This question is a fundamental one in the IS (Information System) literature. IS studies generally reflect a particular position on this question (Orlikowski and Iacono 2001), oscillating from a technological position (technological determinism, discrete-entity tool view, autonomous technology) to a social one (strategic choice view, web-based ensemble models). We are favoring conception of the link between technologies and organizations that are more middleground approaches as reflected in emergent perspective and socio-technical one. From this point of view, we discuss in this section two main approaches: the structuration theory of technology (Orlikowski 2000) and the actor-network theory (Akrich et al. 2002). 1.2.1. A practice lens for studying the use of technology: Orlikowski (2000) The structurational perspective of technology (Orlikowski 2000; DeSanctis and Poole 1994) is deeply rooted in the works of Giddens (1984). As we have previously seen, this emergent perspective takes a middle-ground position in the IS debate on the link between technologies and organizations. From this point of view, these works give a specific and stimulating definition of technologies and uses. According to Orlikowski (2000), technologies present two distinctive aspects: the technology as artifact and the use of technology. Firstly, technology may be define as a technical artifact, which has been constructed with particular materials and inscribed with developers’ assumptions and knowledge about the world at a point in time. These features are coupled with the spirit of technology, which is defined as “the general intent with regard to values and goals underlying a given set of structural features” (DeSanctis and Poole 1994: 126). Pascal A. et Thomas C., “Designing an ICT solution for a multi-actor environment: the KMP experimenation”, The Third Organization Studies Summer Workshop: 'Organization Studies as Applied Science: The Generation and Use of Academic Knowledge about Organizations', 7–9 June 2007, Crete, Greece. 6 Secondly, technology may be termed a technology-in-practice, which refers to the specific structure routinely enacted by individuals as they use the artifact in recurrent ways in their everyday situated activities. In that sense, technology-in-practice may be defined as the set of rules and resources enacted by the users of technology in their recurrent social practice. This double acceptation of technology has several main consequences (Orlikowski 2000): − While technology may be defined as a set of physical properties developed by designers, it is only when this technology is used in recurrent social practices that it structures human action. − When users choose to use a technology, they are also choosing how to interact with that technology. From this point of view, users may, deliberately or inadvertently, use it in ways not anticipated by the developers. The physical properties inscribed in the artifact don’t predetermine use of technology. Thus, “use of technology is not a choice among a closed set of predefined possibilities, but a situated and recursive process of constitution, which may also and at any time ignore such conventional uses or invent new ones.” − Use of technology is heavily influenced by the users’ understandings of the properties and features of a technology, and these understandings are in turn strongly influenced by the images, descriptions, rhetorics and ideologies enacted by individuals belonging to different contexts. − Finally, technology-in-practice is a kind of structure situated in the structural properties of social systems. Thus, in interacting with a technology, users mobilize many structurals, those borne by the technology itself, and those deriving from different contexts (group, community, organization, company, etc.) in which their usages fit. Therefore, these theoretical insights lead us to dissociate the designers’ world from the users’ world. Effectively, when they develop the artifact, designers enact a set of structural properties that give functionalities. Respectively, in their recurrent use of technology, users may enact technology-in-practice in ways no anticipated by developers. Developing a method for designing an ICT tool with regard to the structurational theory also leaves us to have a specific concern on the interactions between the designers’ world and the users’ one. Furthermore, it underlines that technology-in-practice only occurs in recurrent Pascal A. et Thomas C., “Designing an ICT solution for a multi-actor environment: the KMP experimenation”, The Third Organization Studies Summer Workshop: 'Organization Studies as Applied Science: The Generation and Use of Academic Knowledge about Organizations', 7–9 June 2007, Crete, Greece. 7 social practices: consequently, users must have the technology as soon as possible in the design process. 1.2.2. The users’ world: building a socio-technical network Because they have a specific concern on the role of actors in the design process of an innovation, we mobilize the actor-network theory (ANT) for developing our methodology. ANT has been used to investigate the success or the failure of a number of technological innovations. In their approach, these scholars criticized both the Schumpeterian model and its entrepreneur’s role and the diffusion model of innovation: “the bringing together of market and technology, through which both inventions and the outlets which transform them into innovations are patiently constructed, is more and more a result of a collective activity and no longer the monopoly of an inspired and dedicated individual” (Akrich et al. 2002: 189). The key to innovation is thus “the creation of a powerful enough consortium of actors to carry it through, and when an innovation fails to be taken up this can be considered to reflect on the inability of those involved to construct the necessary network of alliances (the socio-technical network) amongst the other actors” (Tatnall and Gilding 1999: 961). Network is used in a special way to describe shifting alliances of both human and non-human actors. Innovation has a collective dimension in which users intervene, and is continuously transforming depending on actors who participate to the process of conception. The model of translation proposed (Latour 1986) in this approach maintains that the movement of an innovation through time and space is in the hands of the socio-technical network who may modify it, deflect it, betray it, add to it, appropriate it, or let it drop. We think that this approach, also named model of interessement, is relevant for designing an ICT solution. On the one hand, it allows understanding the key to success in innovation: “socio-technical compromises and negotiations are the two key notions which allow this work of mutual adaptation which commands adoption to be understood” (Akrich et al. 2002: 211). In this approach, “to adopt an innovation is to adapt it, and this adaptation generally results in a collective elaboration, the fruit of a growing interest” (Akrich et al. 2002: 209). In other words, the success of an innovation depends on a process of adaptationadoption depending on the socio-technical environment. This success also relies on the construction of a socio-technical network that will defend and support the process of innovation. This is also the active role played by spokespersons: “the fate of innovation, its content but also its chances of success, rest entirely on the choice of the representatives or Pascal A. et Thomas C., “Designing an ICT solution for a multi-actor environment: the KMP experimenation”, The Third Organization Studies Summer Workshop: 'Organization Studies as Applied Science: The Generation and Use of Academic Knowledge about Organizations', 7–9 June 2007, Crete, Greece. 8 spokespersons who will interact, negotiate to give shape to the project and to transform it until a market is built” (Akrich et al. 2002: 217). One the other hand, this approach is fruitful by conceiving innovation as the result of a ‘whirlwind’ model, which allows the multiple socio-technical negotiations. In this schema, “innovation continuously transforms itself according to the trials to which it is submitted i.e. the ‘interessements’ tried out. Each new equilibrium finds itself materialized in the form of a prototype which concretely tests the feasibility of the imagined compromise” (Akrich et al. 213). Here, the first prototype is rarely the good one, several successive passes (loops) are generally essential. 1.3. The coevolution of design and use: an integrative methodology The design methodology oriented uses we propose is finally based on two points: first, we introduce users in the designers’ loops through uses scenarios. Second, we separate designers’ world from users’ world in order to better understand the emergence of uses in real work practices. 1.3.1. Building uses scenarios Since the late 1980s, researchers in human-computer interaction (HCI) have used scenarios as an effective means to discover user needs and better embed the use of artifact in work processes. Thus, uses scenarios can be a means for designers to manage the task-artifact cycle in order to achieve greater usefulness and usability (Carroll et al. 1998). In the HCI field, uses scenarios may be defined as “working design representation of user experiences with and reactions to system functionality in the context of pursuing a task” (Jarke et al. 1998: 159). Scenarios focus on the interaction between a system and its environment. Actually based on the distributed cognition theory, uses scenarios often address a narrow work context: classroom, cockpit, and office ... As we see above, structurational approach stresses the necessity to analyze interactions between actors and artifacts through organizational practices. Here, the relevant context is a wider one: the social system and its structural properties. Thus, we propose to complement interaction uses scenarios (HCI approach) with environmental uses scenarios (structurational approach). However, in order to achieve usability and moreover usefulness, uses scenarios are necessary but not sufficient. As structurational and socio-technical approaches emphasized, Pascal A. et Thomas C., “Designing an ICT solution for a multi-actor environment: the KMP experimenation”, The Third Organization Studies Summer Workshop: 'Organization Studies as Applied Science: The Generation and Use of Academic Knowledge about Organizations', 7–9 June 2007, Crete, Greece. 9 artifact must be placed in their real organizational practices, in other words, in the users’ world. 1.3.2. An integrative uses oriented design methodology According to the structurational approach, we propose to separate the designers’ world from the users’ world. The design process (designers’ world) is composed of iterative loops (Akrich et al. 2002). Its main steps are based on the work of Hatchuel and Molet (1986) and Romme and Endenburg (2006) complemented with uses scenarios. The users’ world represents the socio-technical network that supports the process of innovation. Users-pilots (or lead users) are not the only components of this socio-technical network. As we see above, the success of a design project depends on the art of interessement i.e. to enlarge the socio-technical network. Finally, in our methodology, the project results from the coevolution of these two worlds. We proposed to summarize this methodology as follows: 1: the feeling of discomfort 2: building a theoretical support