• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Wireless systems development methodologies: An analysis of practice using Actor Network Theory

Posted on: Thursday, 20 November 2003, 06:00 CST

ABSTRACT

There is a view that the structure of systems development on innovative projects is very different from that of traditional information systems projects (28). This calls for new systems development methodologies tailored to the specific needs of the current development environment. This paper uses the concepts of Actor Network Theory (ANT) to interpret the events that led to the turning around of a failing systems development methodology to success at the mobile application and development center. This paper uses ANT concepts to provide a richer understanding of complexities involved in wireless systems development from a social technical perspective.

Keywords: Mobile Applications Development, Actor Network Theory, Systems Development, Arrow Methodology.

INTRODUCTION

Successful systems development takes on greater significance in view of the increasingly complex wireless applications that need to be developed and integrated. With the development of enabling technologies such as wireless data networks and various types of affordable wireless devices, wireless computing has become widely accepted and applied for both consumer and business initiatives. The structure of systems development on innovative projects is very different from that of the traditional information systems projects (28). This calls for new systems development methodologies tailored to the specific needs of the current development environment. Changes in technologies, economies, organizational structures, and competitive environments are causing fundamental changes in the development of both types of information systems and the process by which they are developed (26). This paper describes the creation of a systems development methodology that was portrayed as a failure and uses ANT to interpret the actions that led to the successful implementation of a new methodology called Arrow methodology. The case studied was at an organization called the Mobile Application Development and Integration Centre (MADIC). The centre created an innovative approach to develop and deploy wireless applications produced by independent third parties.

This paper is structured as follows. The first section provides a brief overview of the systems development literature and introduces the concept of systems development methodologies being developed in- house. The second section will introduce the MADIC case study and use the AR framework to report the findings from the study prior to the ANT analysis. By adopting a socio-technical perspective we used ANT concepts for developing the theoretical sensitivity that allows us to view the large volumes of data in a specific mode for interpretation. This is followed by the conclusion.

BACKGROUND

There is evidence that the profile of systems development on innovative projects is very different from previously promoted systems development methodologies (28). The pace of change that is a characteristic of the business environment facing organizations today is a common theme in contemporary research (12). The Information Communication and Technology (ICT) industry is characterized by a high rate of product and process innovation, rapidly shrinking product and technology life cycles (22), which require new system development ideas. Baskerville (5) discuses the nature of this prevailing business environment and the impact of new technologies to the systems development issue within it. Both integration and customization of packages and outsourcing are quite prevalent in today's environment, yet few methodologies cater to this phenomenon (28).

Integrating mobile systems is more complicated than developing traditional systems due to the number of stakeholders necessary to the development of a single system. The trends in mobile technologies reflect sharing among organizations for mutual benefits (24). For example, the wireless network infrastructure may be provided by one organization while other organizations may provide the applications software and hardware platforms. Increasingly, organizations make use of shared infrastructures and applications provided by specialist third party organizations. These third party organizations may consist of technology companies, business service providers, or possibly a consortium of several organizations from various industries. The complexity of mobile systems is also increased by the inherent nature of mobility and computing.

Although the information systems field has an overabundance of methodologies, organizations are rarely satisfied with existing methodologies (13). Surveys as well as case studies (1, 14, 25, 27, 30) reveal that some organizations tend to develop their own "variants or adapt methodologies available according to their situation-specific needs." Methodologies from outside an organization in many instances do not meet the requirements for its systems development efforts, i.e., they are not considered applicable (15). This leaves organizations with the following options: abandon the methodology, try another "off the shelf methodology," continue the use of the methodology or, as this research reveals, develop methodologies specific to the problem context. This study reveals the process that was undertaken to develop a methodology specific to the context of wireless applications integration. The result of which is that the new methodology includes aspects not found within other methodologies. Surveys investigating the use of methodologies have shown that 38% (14) or 36% (CASE Research Corporation) (32) of the organizations, used methodologies developed in-house. There are also studies that investigate the adaptation or customization of existing methodologies. A survey by Hardy (14) showed that 88% of the organizations studied had adapted methodologies in-house. Though many organizations develop or adapt methodologies in-house, we know little as to why or how this is done or which of these methodologies work best (16). In-house methodology development is not unusual but is often based upon unproven principles (16). These principles include: how to construct and adapt methodologies for specific contexts, how to identify which research methodologies are best suited to the prevailing circumstances, ensure by which means their viability is guaranteed, and how to organize methodology development efforts.

ENROLLING THE "ACTION RESEARCH" ACTOR

The research approach in this case entailed using action research (AR) cycles to examine professional practices surrounding wireless methodology development and application, whose use then underpinned theory building. The research framework in Figure 1 was developed specifically for this study. The initial stage of the methodologies development fashioned the conceptual insights (9) that informed the case study. The conceptual insights come from the research themes and existing knowledge gathered from the literature. The AR case study consisted of four stages: diagnosing, action planning, action taking and reflection (20).

FIGURE 1

Adapted from Research Framework for Action Research (22)

Three types of data collection methodologies were used during this study: in-depth open-ended interviews; direct observations; reflection upon written documents as being part of the methodology development process. The data from interviews consisted of interview notes and taped records which were later transcribed. The recorded elements include direct quotations from people about their experiences, opinions, feelings, and knowledge. The data from observations consisted of detailed descriptions of people's activities, roles, tasks and the full range of interpersonal interactions and organizational processes. The documents were analyzed using qualitative document analysis (11), resulting in the production of wireless application classifications, expert development, quotations, and complete passages from organizational documents.

MOBILE APPLICATION AND DEVELOPMENT CENTER CASE STUDY

This section describes the Mobile Application and Development Center (MADIC) organization, its constituents and the function they perform. It also presents the types of mobile applications they evaluate and integrate into their parent organizations mobile network.

MADIC was initiated to act as a "Creative Innovation Center" for third party developers to expedite their ideas and concepts for MMO2's wireless data networks. MADIC's objective was to strike up partnerships with third party application developers and to speed up time to market profitable products that would exploit MMO2's wireless infrastructure. MADIC operated by creating strong links with the third party development community. An Application Developer Forum (ADF) was hosted at regular intervals where prospective partners were briefed on the benefits of developing products with MADIC. A website was also available to the third parties and was the primary means by which MADIC realized new opportunities. Third party developers registered on the MADIC website promoting themselves and their products to MMO2. The products are usually not purchased outright. Instead, revenues generated by the product or service compensate developers on a revenue share basis, and this approach is taken to mitigate risk.

MADIC consists of skilled resources and a technical infrastructure. The technical wire\less infrastructure includes testing environments to allow third party developers to test their applications and services working over this wireless infrastructure. Experience with the introduction of new services over the wireless data network have shown that Internet Protocol (IP) based applications that work perfectly well over wired networks may not perform reliably over a wireless data network.

The MADIC comprises of a multi-disciplinary team capable of replicating MMO2's business functions. These are made up of the typical wireless operator staff comprising Information Technology, Network Technology, Marketing, Regulatory, Legal and Commercial. The MADIC comprises three different types of people with differing yet worldviews and behaviors:

1. Samurai: They are referred to as the "deal makers" and are the intermediary between the third party developer and the delivery function for the MADIC.

2. Noah's Ark: This team is made up of the typical wireless operator centric staff comprising Information Technology, Network Technology, Marketing, Regulatory, Legal and Commercial.

3. Operations Staff: They are the core team to run the operations and administrative function of the center.

In addition to which was the researcher:

4. Researcher: the researcher's role was to represent the IS/IT functions in a way comparable to the Noah's Ark representative within MADIC, defining the development processes and methodology, intervening to ensure they were implemented correctly.

METHODOLOGY DEVELOPMENT USING THE AR CYCLES

The "Research Framework for Action Research" used by the researcher is set out in Figure 1. This section describes the diagnosis of the situation in MADIC, its problematic history of using methodologies in mobile applications development, the planning of the intervention to address this situation, the action research intervention undertaken as a result and subsequent reflective learning and theory emergence that took place.

During the setting up of the organization, MADIC was encouraged to use the same systems development methodology as that used by the parent organization called the Cellnet Project Planning Methodology CPPM. The intention of MMO2 was to provide uniformity write across the organization and its offshoots to the development approaches, deliverables produced and the project phases. This notion was rejected by MADIC because CPPM proved to be not only highly bureaucratic but also very complicated to implement. A standard project would require approvals from three steering committees. Thirty different documents would be required from the seven project phases; the bid (initiation) phase alone takes an average of 24 weeks to complete.

MADIC argued that the CPPM process worked against the entrepreneurial quick turnaround concept they promoted to third party developers and underpinned the way intended to operate. They proposed the creation of a new process, which for the purpose of this paper is described as the "experimental process." The experimental development process practiced by MADIC at the beginning of the AR study, however, proved to be ineffective. This was due to a variety of issues described below. From the analysis by the researcher of the working practice of the failing "experimental processes" a workable methodology was developed. This incorporated some elements from the "experimental process" but also had many more features that conveyed competitive advantage and quick third party development turnaround to MADIC.

Diagnosing

The initial diagnosis discovered that the development methodology from the parent company CPPM was rejected due to the following reasons: It generated too much documentation and it took too long to implement third party projects. As a result, a new methodology was intrinsically evolving which would, they hoped, deliver projects quickly and with minimal documentation. The team, initially, had no idea what the process would be and did not consult any established theoretical knowledge base, but accepted that they would define and refine the development process as they progressed. The process in place at the beginning of the study was based on the principle of ensuring the involvement of all the roles and functions to the decision making process at an early stage.

Further Diagnosis

Further diagnosis revealed that the experimental development process being tested was, also itself, not working well. The lack of project deliverables that issued from its use created three fundamental problems:

1. It hampered the training of new team members who had not been involved from the project conception.

2. Team members were unable to complete design specification effectively when collaborating with project stakeholders external to MADIC.

3. There was no model to determine the complexity of the individual projects.

The experimental methodology being evolved, the "experimental process" was based on a single activity being repeated over and over. This entailed all the stakeholders meeting at regular intervals to discuss and progress the work concurrently. The result of which were very long meetings in which very few decisions were made. Interviews with the Samurai revealed that they were not capable of reviewing all the products registered. As a result, backlogs of undeveloped application built up.

The unit was under a lot of pressure to identify and implement a "killer mobile application" which would make a significant return on investment. They also had a significant amount of pressure on reducing the timeframe to deliver these mobile applications, all of which were being hindered by this experimental development process. The initial stage of this research was to overtly surface this as an issue. Having overtly acknowledged that this experimental approach was the problem, the next step for the organization was to develop an action plan which would address the problem.

Planning to Address the Problem

The action plan was to apply some structure to the systems development process by only involving people and functions "as and when" they were required on a project. There was, for example, no need for a Commercial Noah's Ark involvement until the technical feasibility had been completed. The researcher was also tasked with analyzing the documentation requirement with a view to produce templates that only had the required information to avoid the production of large documents. After the planning stage the next step was to specify organizational actions that would relieve or improve the problems defined above and enhance the process of third party application development, none of which was exactly the same. This was not a process of third party application mass production but of initial bespoke development.

Action Taking

From studies, interviews and discussions, a third party "project" was to be divided into five phases, which would not only allow for adequate monitoring and reporting but also facilitate wireless application development. This led to a five-phase life cycle called the Arrow Methodology. Each phase was to be assigned an owner who was responsible for the project until it was handed over to the next phase. The next step was to define and agree on the documentation set. A workshop was held and the following actions agreed and principles for the emergent methodology established.

a) The templates for an agreed set of documents were set down. These included: feasibility report, technical specification, application exit report, Memorandum of Understanding (MOU), Non- Disclosure Agreement (NDA) and Service Level Agreements (SLA).

b) The owners, contributors for each document were established through discussion and agreement.

c) The phase in the project life cycle when the documents should be produced was set down.

d) This process was documented in a development approach report by the researcher and was signed off by a representative from each of the functions.

e) A taxonomy of mobile applications was developed, along with a model for determining project complexity, with which to scope the application put forward and its development.

The methodology encompassing these principles was developed with the people in the situation. It was called "ARROW" because the resource allocation structured resembled the tip of an ARROW depicted in Figure 3. The approach is based on the principle of minimizing resource involvement on a project while maximizing developmental response.

FIGURE 2

Arrow Human Resource Allocation Structure

Reflection

The most important lesson learned from the organization was the fact that projects cannot be delivered efficiently without a methodology that everybody accepts and agrees with. The Arrow methodology developed was seen as a "success" because it allowed the projects to be monitored, resources allocated, and controlled. It also increased the number of applications that could be evaluated concurrently.

The "experimental process" had broken down due to two factors. The level of new products being registered for evaluation increased while the staff levels stayed the same, and there was also an increase in the different types of projects of varying complexity that were being registered. At the peak of the experimental process the Samurai rejected a total of 192 products, and 162 products were concurrently processed by MADIC over a period of five months illustrated in Table 1. After implementing the methodology, the number of applications that could be concurrently dealt with significantly increased. Table 2 illustrates the number of products entering the Arrow methodology cycle in the first quarter after the methodology sign-off. These figures do not take into account the two month transition period when the 1st cycle of the AR process was in progress.

Theory Emergence

The study revealed that the process of configuration development, which includes the integration and customization of mobile applications, platforms and servic\es is a very complex process, which involves more than just technical integration. Other factors that need to be incorporated into the systems development process were contractual issues such as agreeing on Service Level Agreements (SLA), support, maintenance and content management and interorganizational relationship. To conceptualize the complexity of wireless systems, a five dimensional reference model was derived which includes the following layers: Device Layer; Network Layer, Application Layer; Program Layer and Payment Layer. The elements from this reference model were used to determine project complexity. Each project is classified as High (H), Medium (M) or Low (L). The classification was provided to depict the level of complexity of the project tasks to coordinate the integration of the application to the wireless network, to the complexity of the product that was initially registered by the third party. These classes were applied for the following reason: to enable better skill to task allocation; allow more accurate development costing; and improve project management on complex projects. The following three elements are used as guidelines to aid the classification of projects: firstly Number of Partners (P), Projects categorized as Low involve less than two partners, Medium projects have less than four and Highly complex projects have in excess of five partners. Secondly, the Integration Factor (IntF). This factor relates the application and program layers and is associated with the level of integration required to integrate the application registered by the third party into MADIC wireless network and back end systems. The third element relates to the device such as phone, laptop or PDA the target application utilized.

TABLE 1

Number of Products Processed by the Experimental Process as of January 2002

TABLE 2

Number of Products Processed Using the Arrow Methodology

Because the Arrow methodology was created to resolve specific problems and to cater only for the MADIC systems development approach, it can be described as a "strong methodology" (12), one that was considered to be a success by the organization as it encouraged skill specialization and accurate division of labor. The next section describes the use of ANT for the interpretation of the successful implementation of the Arrow methodology at the MADIC organization.

ANT ANALYSIS OF MADIC STUDY

The idea of Actor Network Theory (ANT) was first proposed by Michel Callon and Bruno Latour (7, 8). The theory provides a unique sociotechnical approach for understanding the creation of networks of aligned interests. ANT declares that the world is full of hybrid entities (19) containing both human and non-human elements, corporeal and non-corporeal artifacts of all kinds, and was developed to analyze situations where separation of these elements is difficult (6, p. 3). Not only does ANT place equivalent importance on humans and non-humans but also on corporeal and non- corporeal actors. ANT considers both social and technical determinism to be flawed and proposes instead a socio-technical account (18) in which neither social nor technical positions are privileged. Thus an actor may be human, non-human or a networked hybrid combination of both. MADIC, for example, is a networked actor. MADIC's CEO is deemed to be in this study both the focal actor as well as the organizational actor's interlocutor. It was the CEO who problematized overtly the failure of the "experimental approach" actor and who sought to address it by bringing in the action researcher.

Table 3 describes the ANT concepts used for analysis in this study. This paper adheres to the philosophy of ANT to analyze the role of the AR methodology as a change actor and in doing so defines the most efficient working process of an emergent and more effective MADIC network.

TABLE 3

ANT Concepts

Problematization

The overall problematization centered, initially, on the betrayal of the experimental process artifact when it was enacted by other actors to define and facilitate an effective working process for the focal actor, the organization, MADIC. The human and technological actors were having difficulties working with the experimental process under the pressure of evaluating, designing and delivering multiple wireless applications concurrently. The experimental process as a methodological actor worked well when the majority of the actors' tasks focused on evaluating and deciding what applications to progress in the initial development stages. When more applications progressed to a stage, they required more detailed designs and integration tasks the process was incapable of coping with the required level of detail. The methodological actor betrayed the other elements in the network. Another problem that became apparent to the Focal Actor "MADIC" CEO was the inability of the experimental process to deal with the increased complexity of the applications that had to be progressed simultaneously.

The focal actor, the CEO, and others, were also concerned about the task and resource allocation as it seemed that some human actors were heavily loaded with tasks while others were bored from lack of work that required immediate attention and were little mobilized.

The Focal Actor addressed the preliminary problematization effectively traversing the series of OPPs listed below and described in Table 4. This was facilitated by the fact that:

1. The action researcher was enrolled to review the experimental process and produce a Project Life Cycle document to guide the development process.

2. A review of the required documentation was ordered with a view to produce templates for guidance.

3. Workshops were organized to discuss the changes in working practice, which led to the working forums being replaced by a phased approach to development.

4. A review of the roles and responsibility was initiated to address the resource allocation imbalance.

5. The action researcher was tasked with producing a model that would enable the human actors to determine the complexity of the technical artifacts to feed into the new methodology.

The process added to the solution component of the problematization being addressed, namely, the betrayal of the "experimental" process as a methodological actor, which when enacted by the other actors in MADIC, would prove capable of facilitating third party application development.

Outcome of the ANT Analysis

The CEO Focal Actor emerges from MMO2 intent on creating the MADIC organization through traversing the series of obligatory passage points listed above by translating the human professional, technical and methodological actors. MADIC's aim was to translate third party wireless technologies into its own organizational network in order to test, assess and reject or enroll them as potential applications delivered using MMO2 wireless computing platforms. Methodological actors were also translated to facilitate this process. The incumbent CPPM development methodology was enrolled by the focal actor to facilitate MADIC's workings and to enable it to traverse the OPPs necessary to achieve initial effective functioning and survival in the development of applications for the wireless arena. However, this methodology had the MMO2 interests inscribed in it, rather than MADIC's. It was a mechanism, a machination, by which MMO2 sought to maintain control over MADIC. Its enactment by the MADIC actors resulted in their being betrayed by the CPPM methodology. This was due to its inflexibility and the over-engineering of the wireless applications decision-making and development process, which resulted from the complexity of its stages and the multiplicity of its milestones. It inhibited the way in which the emergent MADIC network met the contingencies of the fast moving business environment for new wireless applications. In this way it failed to facilitate the MADIC network in their traversing a series of OPPs by which it would achieve effective organizational form, working practices and performance in "wireless technology" development for the parent actor network

TABLE 4

ANT Analysis Summary of the MADIC Methodology Implementation

A new application development methodology emerged from the intrinsic working practices of MADIC, prompted by the CEO focal actor and enacted by same professional actors comprising the Samurai, Noah's Ark, and Operations Staff. This methodological actor also betrayed them. This was due to it convening all the different categories of professional human actors in a single decision-making process at the same time and location, which proved ineffectual and time consuming without delivering the decision that was required on an enrolled application. It too was configuring the way that the organizational actors were working, which was incommensurate with how the organizational actor network needed to function in order to meet the contingencies it was faced with, and produce the deliverables required by the parent organization. This new methodology did not have MADIC's network's interests effectively inscribed in it and therefore failed to consolidate the organizational network and facilitate it as effective functioning as it should have. The emerging methodology worked in the early stages when applications developed were of a similar nature, mainly consumer products such as games, ringtones and multimedia content; these could be classed as low complexity applications. The focal actor decided that MADIC should have a change in focus to try and generate a return on the investments. This change of focus was achieved by instructing the Samurai to evaluate more data applications suited for business solutions. The outcome of this change revealed that more products needed to be put through the MADIC for evaluation. The focal actor was under pressure not to recruit additional staff. The action researcher was translated into the network to research and addr\ess the ongoing problematization, supported by the Action Research Methodological (ARM) factor. When enacted by the action researcher and MADIC's actors, the research methodological ARM actor first revealed the betrayal of the MADIC network's actors. Their incumbent experimental methodological actor was overtly exposed by ARM as being inadequate in shaping the organization's form and function. It inhibited the MADIC from traversing the OPP of higher functionality being demanded of it by the market and the CEO. It betrayed them, hence the bringing into being of the Arrow methodological actor.

The enactment, by the human professional actors, of the five stage Arrow methodological actor encouraged skill specialization and accurate division of labor between MADIC's developmental actors. Standardization of the enactment of the development process was also possible through the overt and acknowledged performance of the methodology by the actors in MADIC network. This led to increased productivity and quality, as actor resource requirements could be predicted and translated whenever necessary, as well as further consolidation of the network. The methodology enabled the management to know who was responsible for a project at a certain point in time, along with the status of that project in the development life cycle. The Arrow methodology also allowed the management to translate technical and marketing actor resources from the parent organization when they were required for the project. It consolidated the MADIC network.

It was not practical to compare mobile applications time to delivery or costs associated with delivering products due to their diversity. Therefore, a resource allocation model was used to evaluate the performance of one methodology over the other. The number of mobile products processed was used as an indication of the effectiveness of the methodology. This approach is taken because the development of mobile applications is not an industrial type production line but a set of unique bespoke development tasks, which exhibit a broad product mix of varying complexity. The number of human actors associated with development was consistent through the course of the study by the number of applications they were capable of processing varied significantly dependant on the development process being used and the application being put forward, as illustrated in Table 5.

TABLE 5

Summary of Methodologies Implemented

An increasing degree of complexity was introduced to the development process at the introduction of the Arrow methodology. This was due to the degree of sophistication of the data products required to create wireless business solution as demanded by the focal actor.

CONCLUSION AND FURTHER RESEARCH

This paper has reported the findings from an action research study at an organization called MADIC. MADIC evaluates and integrates wireless applications developed by third party developers into the parent organization MMO2 network. The aim of the action research initiative was to define a systems development methodology suited to organization aims. This paper provides the diagnosis of the situation in MADIC, its problematic history of using methodologies in mobile applications development, the planning of the intervention to address this situation, the action research intervention undertaken as a result and subsequent reflective learning and theory emergence.

This paper used ANT to analyze the rationale behind the successful implementation of the Arrow systems development methodology and the failure of the previous methodologies. It described how all the actors within MADIC were translated and mobilized into an effective problem-solving network to address the problematization brought about by the experimental approach's betrayal.

At the implementation of the Arrow methodology the wireless products actors were becoming increasingly more complicated, matched by the human actors becoming more skilled as task specialization being encouraged due to identification of roles and responsibilities. The improved resource allocation and use facilitated by enactment of the Arrow Methodology actor, was the basis for the improved performance and effectiveness, which addressed the original problemization. The network was able to operate more efficiently not only allowing more wireless applications to be evaluated by also increasing the mobile applications being progressed into the delivery phase for launch.

REFERENCES

1. Aaen, I. "CASE Tool Bootstrapping - How Little Strokes Fell Great Oaks." Lyytinen, K. and V.-P. Tahvanainen (Eds.). Next Generation CASE Tools. Amsterdam: IOS Press, 1992, pp. 8-17.

2. Aalto, J.-M. "Experiences on Applying OMT to Large Scale Systems." In Lehtola, A. and J. Jokiniemi (Eds.). Proceedings of the Seminar on Conceptual Modeling and Object-Oriented Programming. Finnish Artificial Intelligence Society, 1993, pp. 39-47.

3. Avison, D. and G. Fitzgerald. Information Systems Development: Methodologies, Techniques and Tools. Oxford: Blackwell Scientific Publications, 1995.

4. Baskerville, R.L. "Investigating Information Systems with Action Research," Communications of the Association for Information Systems, 2:19 1999.

5. Baskerville, R., J. Travis, and D. Truex. "systems Without Method: The Impact of New Technologies on Information Systems Development Projects." In Kendall, K., J. DeGross, and K. Lyytinen (Eds.). The Impact of Computer Supported Technologies on Information Systems Development. Elsevier Science Publishers B.V. North Holland Press, 1992, pp. 241-269.

6. Callon, M. "Actor-Network Theory - The Market Test." In Workshop Centre for Social Theory and Technology (CSTT), Keele University, UK, 1997: http://www.keele.ac.uk/depts/stt/ant/ callon.htm.

7. Callon, M. "The Sociology of an Actor-Network: The Case of the Electric Vehicle." In Callon, M., J. Law and A. Rip (Eds.). Mapping the Dynamics of Science and Technology: Sociology of Science in the Real World. London: Macmillan, 1986, pp. 19-34.

8. Callon, M. and B. Latour, "Unscrewing the Big Leviathan: How Actors Macrostructure Reality and How Sociologists Help Them to Do So." In Knorr-Cetina, K.D. and A.V. Cicourel (Eds.). Advances in Social Theory and Methodology: Toward an Integration of Micro- and Macro-Sociologies. Boston: Routledge and Kegan Paul, 1981, pp. 277- 303.

9. Carroll, J. and P. Swatman. "Structured-case: A Methodological Framework for Building Theory in Information Systems Research," European Journal of Information Systems, 9:235, 2000, pp. 235-242.

10. Checkland, P. Systems Thinking, Systems Practice. Chichester: Wiley, 1981.

11. Denzin, N.K. and Y.S. Lincoln (Eds.). Handbook of Qualitative Research. Thousand Oaks, CA: Sage, 1994.

12. Fitzgerald, B. "Systems Development Methodologies: The Problem of Tenses," Information Technology & People, 13:3, 2000, pp. 13-22.

13. Fitzgerald, B., N. Russo, and E. Stolterman. Information Systems Development: Method-in-Action. UK: McGraw-Hill, 2002.

14. Hardy, C., J. Thompson, and H. Edwards. "The Use, Limitations and Customization of Structured Systems Development Methods in the United Kingdom," Information and Software Technology, 37:9, 1995, pp. 467-477.

15. Howard, G., et al. "The Efficacy of Matching Information Systems Development Methodologies with Application Characteristics - an Empirical Study," Journal of Systems and Software, 45, 1999, pp. 177-195.

16. Juha-Pekka, T. "Incremental Method Engineering with Modeling Tools: Theoretical Principles and Empirical Evidence," Computer Science, Economics and Statistics (Thesis), University of Jyvaskyla, Jyvaskyla, 1998. http://www.cs.jyu.fi/~jpt/doc/thesis/ime.html.

17. Kodama, M. "The Mobile Internet Revolution in Japan: A Case Study in NTT DoCoMo," Long Range Planning, 4:35, 2002, pp. 349-365.

18. Latour, B. The Powers of Association. Power, Action and Belief. A New Sociology of Knowledge? Sociological Review Monograph. London: Routledge & Kegan Paul, 1986, pp. 264-280.

19. Latour, B. We Have Never Been Modem. Hemel Hempstead: Harvester Wheatsheaf, 1993.

20. Lewin, K. "Frontiers in Group Dynamics II. Channels of Group Life; Social Planning and Action Research," Human Relations, 1:2, 1946, pp. 143-153.

21. Makay, J. and P. Marshall. "Imperatives of Action Research," Information Technology & People, 14:1, 2001, pp. 46-59.

22. Nambisan, S. "Software Firm Evolution and Innovation- orientation," J. Engineering Technology, Management, 29, 2002, pp. 141-165.

23. Nissen, H.E. "Responsible Action in the Use, Management and Development of Information Systems." In Jayaratna, N. and B. Fitzgerald (Eds.). Lessons Learned from the Use of Methodologies (Proceedings of the 4th Conference on Information Systems Methodologies). British Computer Society, 1996, pp. 1-16.

24. Olla, P. and N.V. Patel. "A Value Chain Model for Mobile Data Service Providers," Telecommunications Policy, 26:9-10, 2002, pp. 551-571.

25. Pyburn, P. "Linking the MIS Plan with Corporate Strategy: An Exploratory Study," MIS Quarterly, June 1983, pp. 1-14.

26. Russo, N. L. "Exploring the Assumptions Underlying Information Systems Methodologies. Their Impact on Past, Present and Future ISM Research," Information Technology & People, 34:4, 2000, pp. 313-327.

27. Russo, N., J. Wynekoop, and D. Walz. "The Use and Adaptation of System Development Methodologies." In Proceedings of International Conference of IRMA (International Resources Management Association, Atlanta, ay 21-24, 1995.

28. Sawyer, S. "A Market-based Perspective on Information Systems Development," Communications of the ACM,44:11, 2001.

29. Solander, K. and K. Lyytinen. "How to Combine Tools and Methods in Practice - A Field Study. In Steinholtz, B., A. Solvberg, and L. Bergman. Lecture Notes in Computer Science, Second Nordic Conference CAiSE'90. Stockholm, Sweden, May 1990, pp. 195-211.

30. Wijers, G. and H. van Dort. "Experiences with the Use of CASE Tools in The Netherlands." SOlvberg, A,and L. Bergman (Eds.). Advanced Information Systems Engineering, e.B.S., Advanced Information Systems Engineering. Germany: Springer-Verlag, 1990, pp. 5-20.

31. Wynekoop, J. and N. Russo. "Studying System Development Methodologies: An Examination of Research Methods," Information System Journal, 7, 1997, pp. 47-65.

32. Yourdon, E. The Decline and Fall of the American Programmer. Englewood Cliffs, NJ: Prentice-Hall, 1992.

PHILLIP OLLA, CHRIS ATKINSON, and RESHMA GANDCEHA

Brunel University

Uxbridge, Middlesex, UB8 3PH

Copyright International Association for Computer Information Systems Fall 2003

More News in this Category