How to support innovative knowledge communities by building an activity theory based object oriented analysis and design method
In the master’s thesis, the approach of developmental work research, and conceptual tools developed within it, offered the theoretical framework for constructing and applying the new method and tools. They were introduced for modeling and analyzing the work processes of software projects as well as ways to enhance these processes. (Ahonen et al., 2000, Engeström, Y. et al., 1995).
This paper describes above mentioned software project as an example case where virtual “change laboratory” was implemented and integrated as a part of the project work. Moreover, the tools and methods introduced in the master’s thesis were applied for modeling and analyzing the work itself. These tools included both the applications of virtual “change laboratory” as well as an object-oriented method to visualize, analyze and to develop further the work among the project team. The object-oriented method was based on tools that were familiar to most software designers. Such tools are; for example, use case driven modeling and ontologies. Special effort was put on choosing a proper ontology for the design.
Ontology, as understood in information sciences, was defined to be a way of constructing representations and their associations related to some real world system. However, the criteria for choosing proper ontologies for the software projects has lacked sufficient background theories and methodologies for describing the social and context dependencies of the representations. Dependencies are usually contained into models in ad hoc way. This paper presents a method for analyzing the intermediate artifacts in a context of activity as a complement for existing ontology methodologies. The purpose is to provide a heuristic way to analyze the cultivation process of subject and object in an activity. Moreover, it provides a method to recognize what kind of affects intermediate artifacts have on the system of activity in a certain context.
Consequently, analysis can be seen as a way of creating new knowledge through identifying associations and relations between artifacts. This is very similar to the way that the whole human evolution works, i.e. it can be seen as a process of remediation, replacing the old intermediate factor with a new one that better serves the activity (Virkkunen and Kuutti, 2000). Identification of relations and associations helps to understand the combination of human and inhuman factors. Analysis can therefore provide a better orientation for the future. This method is seen to be applicable especially for evaluating innovative knowledge communities where new knowledge objects and technologies are continuously introduced (Hyötyläinen, T., 2002). Theory background for the described approach has been searched from theories of social networks (Nardi et al., 2002), communities of practice (Lave and Wenger, 1991) and actor networks (Latour, 1999), but unfortunately they have offered very little concrete tools for this purpose.
The purpose of this paper is also to collect the five years of experience of working in a software company that can be said to be a leading innovation in the business. From this experience, we have tried to identify how the activity theory and developmental work research can be further expanded to include ways to analyze the evolution and motivation of such innovative knowledge communities.
Motivation for this kind of study has been raised from eternal problems of software projects that have remained in the field since the 70’s: time schedules cannot be kept, resourcing fails, complexity of the system and lack of competence drives software houses to low-grade solutions. Moreover, transformation to networked environments and e-commerce business models has caused new kinds of problems to software projects. These problems can be conflicts, distractions and divergences in the every day life of the individual knowledge worker. In developmental work research, instead of pre-emptive actions, these problems are seen as a growing platform for new innovations.
When developing new, complex software and information systems, distractions arise from intertwinement of communicative and instrumental work activity. On the background, there are also the influences of prevailing cultures, values, norms and morals – or the lack of any of these. Contradictions apparent in interaction are studied in this paper especially from the before mentioned point of view by connecting them to the total structure of activity system and it’s contradictions.
One very outstanding report named “Chaos” (Standish Group, 1995) concentrated on studying the commercial software industry and it stated that most essential factor for failures were poor requirement management in the projects. Findings of the Chaos report indicated that if organizations know what they are doing, the how part is not that big of a problem. These challenges have tried to be answered by bringing more efficient practices into different areas of software project such as automation of testing, more extensive use of commercial off-the-shelf components and the utilization of commercial conventions such as iterative processes and architecture centric approaches.
Consequently, even more challenges have appeared due to digital convergence that is taken place as a fusion of telecommunication, data communication and consumer electronics. Producing new services based on new information technology means changing the operational models and networking with other organizations. Learning has been seen as the most important surviving factor and knowledge the most important factor of production (Stewart, 2001). Keywords of the successful software projects in the new millennium have become such concepts as innovation and learning organization. Producing these new services require innovative approaches in the knowledge production as knowledge workers are required to network faster than before and without relying that much of the stability of the prior knowledge they poses.
Apparently, none of these above mentioned points seem to be the “magic bullet” that bring the desired changes into the software projects. In a mobile production, the main issue is the transfer from community of practice into networks where knowledge production and collective learning is situated. However, lots of divergence exists in the theoretical approaches for understanding these problems. At the same time, the technological field continues to converge. Question that this paper tries to answer is that what should be the theoretical and practical method that bridges the gap between knowing and doing?
Wenger presents that expertise and competence are mediated through closely working “community of practices” (1998). The concept of community of practice has been used to describe a certain expertise in some area of work or people that represents that expertise. Those people work closely day after day together formally or informally in order to reach some goals related to their common knowledge.
Nonaka and Takeuchi (1995) presented a theory for knowledge creation and cyclic process that transform the explicit information into tacit knowledge. Transformation process has four steps: socialization, externalization, integration and internalization of knowledge. Transformation of knowledge into different forms is bi directional. Tacit knowledge becomes explicit, but explicit knowledge also becomes tacit. Key to innovative learning and knowledge creation in their model can be understood as conversion of tacit knowledge into explicit forms where it can be combined, followed by an internalization process where this combined knowledge becomes a part of learner’s knowledge structure (Tuomi, 1999).
Engeström (1999) criticized Nonaka and Takeuchi’s model for being too deterministic and focusing too much on the knowledge that the community/team already had. Tuomi saw also that their model was based on individuality (Tuomi, 1999). If we consider Nonaka’s and Takeuchi’s model from the representation methodology point of view where knowledge transforms from object – explicit knowledge to subject – tacit knowledge, then it would suggest that the quality of the knowledge degrades per each cycle due to distractions and contradictions caused by subjectivity, or the creation of qualitatively better knowledge would be dependent entirely on the existing learning capabilities of the team members but no development would be possible. However, this is not the case, higher states of learning can occur even among very novice learners. This would lead to an assumption that not all knowledge can be derived back into individual thinking.
Consequently, Cook and Brown (1999) have studied the forms of knowledge in more detailed level. They see that the knowledge of an individual and a group are different from each other, are equal, and one is no subset for the other. They argue that much knowledge is contained into human’s activity and practices that cannot be located into human’s head.
Engeström’s (1999) framework on knowledge creation is based on activity theory, which is grounded to the aspects that activity and learning are highly context related, dialects exist between thinking and knowing, and that activity theory seeks qualitative changes in the behavior of the people.
Karvonen suggest that such ontological baseline should be taken where “knowing subject” is not seen as an observer outside the reality but on the contrary, part of his reality. We are not outside the world “as it is” but we are in the middle of it. Therefore what we know about the world as well as what knowledge we create of it depends on how we are in the world and how we confront the world (Karvonen, 2002)
According to aspects presented above, Nonaka’s and Cook and Brown’s separation of knowledge could be presented as following.
Figure 1. Context of knowledge creation.
Knowledge creation process is reorganization of physical, explicit, as well as mental, tacit knowledge elements in an activity that enables qualitatively better results. Knowledge elements can reside inside or outside the individual, but most important is the functional relation between elements that realizes only in a certain social and cultural context (Hyötyläinen, 2001).
In the Nonaka and Takeuchi's theory, Japanese engineers traveled all around the world and took pictures from western technological artifacts and used the pictures to recreate both the technological artifact and the knowledge for building such an artifact. Based on Latour's concept of intermediate factor, pictures themselves as well as the subjects of pictures can be seen as technological artifacts that have an intermediating affect on the knowledge creation process. However, as Latour noticed, such functionality must be explained with the combination of human and inhuman factors.
Activity theory relies strongly on the concept of intermediate factor. Virkkunen and Kuutti present that human evolution can be seen as a process of remediation, replacing the old intermediate factor with a new one that better serves the activity (Virkkunen and Kuutti, 2000). In order to better understand the meaning of different intermediating factors, they must be analyzed in respect to their purpose and inner relations within an activity. In the context of activity-theory such analysis is done using the model of activity system (Miettinen, 1997).
Figure 2. The structure of activity according to Engeström (1987)
Engeström sees that primary contradiction of activities in capitalistic socio-economic formations is the inner conflict between exchange value and use value. Secondary contradictions are those appearing between the elements of the figure 2.
One characteristic feature of Engeström’s expansive model is that the object of learning is a complete system of activity and the qualitative change of the system. This means that learning process is collective and long-lasting process by its nature. Carrying out qualitative and profound change is not often possible in rapidly changing situations and relatively short projects. In addition to above mentioned, there exist two basic features in my work that Engeström et al. call multicontextuality and boundary-crossing laboratory (Engeström et al., 1995). This kind of multicontextual networking is also called negotiating knot-working (prev).
Engeström et al. have developed two kinds of intervention methods for pushing forward the development in a certain point of the process, and also for supporting the change management of the working community. They call these methods a change laboratory and a boundary-crossing laboratory. Establishing networks requires crossing boundaries.
According to Wenger (1998), crossing boundaries can happen either by personal participation or through reification that integrates the practices. Star (1989) says that crossing boundaries is mediated through artifact, that is, common boundary object. Boundary object can either be a concrete or abstract tool that units the perspectives of different actors.
While the project was proceeding, different kinds of problems were appearing
as distractions, conflicts and contradictions. These problems were documented
into the project’s web pages (figure 3 and 4), which represented the virtual
space. Problems were then further investigated from the activity system
and the complete structure of the network of the activity system point
of view. Instead of pre-emptive actions, the contradictions apparent in
the system of activity were seen as a growing platform of new innovations.
Figure 3. Project’s web page.
Figure 4. Contradictions as well as artifacts of the project were externalized into the virtual space
Following tools were supported in the virtual space:
The system of activity was translated into a language of object-oriented analysis that was familiar to the project workers (figure 5).
For representation semantics, Unified Modeling Language (UML) was used mainly because it is an industry standard and very widely used in the Nokia. Moreover, the semantic gap between natural language and UML is fairly small.
In the context of system of activity, contradictions documented into web pages were analyzed using the object-oriented analysis in order to find new perspectives, intermediate artifacts or boundary objects that would help carrying out qualitative changes. Moreover, object oriented analysis advanced the outlining of the object during the project because the complete network of activity system was always apparent in the analysis. Results of the analysis often showed that the object was actually a systemic composition rather than a single representation of some abstract or concrete artifact.
This has been a success factor for a corporate such as Nokia in the 90’s, which has utilized a model where networking with subcontractors and clients covers all levels of activity (Pulkkinen, 1996). However, stable and long-lasting process models cannot answer the challenges of innovative knowledge communities in today’s turbulent and rapidly changing software industry. Descriptions of activity must take account the requirements of producing innovations and qualitative changes even in short-lived networks of working communities.
This paper suggests such descriptions of activity to be based on object-oriented analysis. Analysis is collective effort that is carried out in the context of change and boundary-crossing laboratory. This is done in order to enable so called expansive cycle, where considering the activity in a wider context can solve contradictions in the system of activity. Externalization of both the descriptions and work products into a virtual space supports rapid networking and innovative working communities.
Because the object-oriented method presented in this paper is heuristic, it is error prone. Method does not directly produce correct or complete models or descriptions of activity. Existing applications of this method are few and development of the method is still a work in progress. However, it is shown to be suitable specially for describing the activity systems of both the software products and processes that produce the software.
References:
|
Ahonen, H., Engeström,
Y. & Virkkunen, J. (2000). Knowledge management – the second generation:
Creating competencies within and between work communities in the competence
laboratory. In: Knowledge Management and Virtual Organizations. Y. Malthora
(ed.). Hershey, PA: Idea Group Publishing.
|
|
Brooks, F. P. (1975)
The Mythical Man-Month: Essays on software engineering, Addison-Wesley,
1975
|
|
Cook, S. D., Brown, J. S. (1999) Bridging Epistemologies:
The Generative Dance Between Organizational Knowledge and Organizational
Knowing, Organizational Science/Vol. 10. No.4. July-August, s. 381-400,
1999
|
|
Engeström, Y. (1987). Learning by Expanding.
An Activity-theoretical Approach to Developmental Research. Väitöskirja.
Helsinki: Orienta-Konsultit. 1987
|
|
Engeström, Y. (1999).
Innovative learning in work teams: analyzing cycles of knowledge creation
in practice. In Engeström, Y., Miettinen, R. & Punamäki R-L
(ed.) Perspectives on activity theory. Cambridge University Press. 1999
|
|
Engeström, Y.,
Engeström, R., Kärkkäinen, M. (1995). Polycontextuality
and boundary crossing in expert cognition: learning and problem solving
in complex work activities. Learning and Instruction, 5.
|
|
Hacking, I. (2002). Historical Ontology. Harvard
University Press, Cambridge Mass. 2002
|
|
Hyötyläinen, E., Hyötyläinen,
T. (1996) Virtuaaliympäristöt fysiikan oppimisympäristöinä.
Esitelmä Matematiikan ja luonnontieteiden opetuksen tutkimusseuran
ry:n syyspäivillä 18.-19.11.1996. In press. University of Oulu,
1996
|
|
Hyötyläinen,
T (2001). Prosessien kehittäminen verkostuneissa ohjelmistoprojekteissa.
Pro Gradu -tutkielma. Tietojenkäsittelyn ja Informaatiotutkimuksen
laitos. Tampereen Yliopisto. 2001.
|
|
Hyötyläinen,
T. (2002). Converging e-learning and IP multimedia in mobile environments.
Contents and uses: Multimedia in education and learning, EUROPRIX SCHOLARS
CONFERENCE 2002: Understanding the Future of European e-Content Industries,
In press. 2002
|
|
Karvonen, E. (2002) Huono tietoteoria nousee huonosta
ontologiasta. Available at http://www.uta.fi/~tierka/episteme.htm
|
|
Latour, B (1994). On technical mediation. Philosophy,
sociology, genealogy. Common Knowledge 3 (1994), 29-64. 1994
|
|
Latour, B. (1993) Ethnography of a 'high -tech'
case: About Aramis. In Lemonnier, B. (ed.) Technological choices. Transformation
in material cultures since the Neolithic. London: Routledge, 372-398. 1993
|
|
Latour, B. (1999). Pandora’s
hope. Cambridge, MA: Harvard University Press.
|
|
Lave, J. & Wenger,
E. (1991). Situated Learning: Legitimate Peripheral Participation.
Cambridge University Press.
|
|
Miettinen, R. (1997) The Concept of Activity in
the Analysis of Heterogeneous Networks in Innovation Process, Center for
Activity Theory and Developmental Work Research, Department of Education,
Helsingin yliopisto, http://communication.ucsd.edu/MCA/Paper/Reijo/Reijo.html
(13.9.2000), 1997
|
|
Nardi, B., Whittaker,
S., Schwartz, H. (2002). NetWORKers and their activity in intensional networks
(Special issue on activity theory and design, guest edited by Bonnie Nardi
and David Redmiles). Journal of Computer-supported Cooperative Work,
11, 1-2.
|
|
Nonaka, I., Takeuchi, H. (1995), The Knowledge-Creating
Company, Oxford University Press, New York, 1995
|
|
Pulkkinen, M. (1996). Miten jättiläistä
horjutetaan? In Tarmo Lemola and Raimo Lovio (ed.) Miksi Nokia, Finland.
Porvoo: WSOY. 1996
|
|
Standish Group, (1995)
The Standish Group, Chaos, 1995
|
|
Star, S. (1989). The
Structure of ill-defined solutions: Boundary objects and heterogeneous
distributed problem solving. In L. Claser & M. N. Huhns (Eds) Readings
in distributed artificial intelligence 3. Meno Park, CA: Morgan Kafman.
1989
|
|
Stewart, T. (2001). The Wealth of Knowledge – Intellectual
Capital and the Twenty-first Century Organization. Published by Doubleday.
New York. 2001
|
|
Tuomi, I. (1999). Corporate Knowledge. Theory and
Practice of intelligent organizations. Helsinki: Metaxis
|
|
Virkkunen, J., Kuutti,
K. (2000) Understanding Organizational Learning by Focusing on "Activity
Systems", Accounting, Management & Information Technology 10, No 4,
291-319.
|
|
Wenger, E. (1998). Communities
of Practice: Learning, Meaning, and Identity. Cambridge: Cambridge
University Press. 1998
|