Scandinavian Journal of Information Systems Volume 29|Issue 1 Article 3 6-30-2017 Work Informatics—Mon Amour Markku I. Nurminen University of Turku, [email protected] Follow this and additional works at:http://aisel.aisnet.org/sjis Recommended Citation Nurminen, Markku I. (2017) "Work Informatics—Mon Amour,"Scandinavian Journal of Information Systems: Vol. 29 : Iss. 1 , Article 3. Available at:http://aisel.aisnet.org/sjis/vol29/iss1/3 This material is brought to you by the Journals at AIS Electronic Library (AISeL). It has been accepted for inclusion in Scandinavian Journal of Information Systems by an authorized administrator of AIS Electronic Library (AISeL). For more information, please [email protected]. Nurminen: Work Informatics—Mon Amour Work Informatics—Mon Amour Markku I. Nurminen Professor emeritus, Turku School of Economics University of Turku, [email protected] Abstract. This article displays the professional career of the author. It is start- ed with a chronological narrative of his academic work highlighting some cru- cial events in it. The main contribution of his work is a conceptual framework called ‘Work Informatics’ that regards the use of information technology as an inherent part of the work of the actual user. This framework is briefly explained. The fi- nal discussion tries to transcend the restrictions created by the unusual subjective char- acter of this paper: what are the lessons learned and how are the two stories connected? Key words: Information system, work, actor, social interpretation, organisational implemen- tation. 1 Introduction Shortly after retiring in 2009 I started to think whether I had accomplished anything during my academic years in research. I picked up the list of my publications. I must say I was quite disap- pointed with the image it gave others about my career, but not in terms of quantity or quality. The beef of my work was not noted on the list; somehow it emphasised the wrong aspects. Then I extracted a few more essential aspects that could be called theories or frameworks that I had created or that I had contributed to their invention. This list clearly gave a better answer to my original question. These were something I really had done. This list thus became the beginning of these memoirs and the frameworks summarised below. The first part of this paper tells the story that needs to be told to connect the items together. The second part tells more about Work Informatics (WI), that is, the most comprehensive framework to which I have contributed. Fi- nally I conclude with some remarks on the period that has followed my working life. I am rather confused about speaking about my life in the first person. Yet I never had an academic life that was distinct from my personal life. My research has always been part of my Accepting editors: Arto Lanamäki, Rudy Hirschheim and Jaana Porra © Scandinavian Journal of Information Systems, 2017, 29(1), 57–78 Published by AIS Electronic Library (AISeL), 1 Scandinavian Journal of Information Systems, Vol. 29 [], Iss. 1, Art. 3 social and political activities. Therefore, I believe this offered personal context will bring the true taste of my life to these academic contributions. 2 Year by year I received my student grade simultaneously with my sister, two years older than me. For most of my school years, all my classmates were two or more years older than I was. Yet I survived those teenage years by using two tricks: First, try to avoid open confrontation because you will be the loser; and secondly, learn to create alternative interpretations of all situations that may develop into conflicts. In my case, it became clear that I have been repeating these cornerstones of Finnlandisierung1 throughout my whole life and in particular during my research. Not even research can be explained in terms of the object of the work alone, as the researcher makes the result. In the special issue of careers, however, it is legitimate to refer to personal background of the kind spoken of here. 2.1 The path to computers and research I started my university studies in Mathematics at the University of Turku, Finland, in 1960, to achieve a professional competence for a related and appropriate job. Just a few months before receiving my Bachelor’s degree in 1963 I went to test for programming courses organised by IBM. After a successful test, I participated privately in a course that opened access to a summer job at IBM for a few subsequent summers. My programming skills also opened the doors to em- ployment at the university to initiate the very first programming courses there, which was more an opportunity to earn money than my major vocation. My Master’s Thesis was on Operation Analysis wherein I contributed to a particular search problem. I was happy and motivated because I learned that when a small child gets lost in a forest, good search methods are invaluable. It is what I was told at the start, but only later did I understand that the object of a search could become an unfriendly submarine as well. Life in Turku in the 1960’s was not isolated from international cultural trends. Woodstock and other hippie movements also declared, “Make peace, not war!” This atmosphere stimulated us in our small research group to design experiments using computer music, also used for small breaks from military applications. I was playing clarinet in the students’ wind orchestra, and I also conducted this band for a few years. This hobby led me to study more musical theory. In 1966, I received a Master’s degree, and during the next year, I published my first composition: A Finnish tango, created using a multi-level Markovian process run in the University’s then only computer (Nurminen 1971)2. It was my first and hopefully last contribution to Artificial Intelligence. The model turned out to be successful in finding small melodic fragments, but it was quite poor at mastering larger musical structures. The composition did receive, however, reasonable publicity. During the composition period, many other researchers produced com- positions using computers. This tango project was original, however, in attempting to compose pieces within an existing genre of popular music. For me, the project was a failure because I had 58 • Nurminen http://aisel.aisnet.org/sjis/vol29/iss1/3 2 Nurminen: Work Informatics—Mon Amour to finish the composition manually. However, it was the starting point of my passion to get rid of using computers as a subject; as an actor on its own. When I returned from my military service in 1969, I had to decide whether I should or should not continue my research activity. My licentiate thesis (the intermediate level between MSc and the PhD) before military service continued the optimisation genre within search prob- lems. At the time of my return to civil society I was lucky, because the then Humanities faculty was inviting applications for the position of mathematician within various areas of the Humani- ties. Thus, during the next two years, I had the privilege of creating an archive of spoken dialects over all areas of Finland (Professor Osmo Ikola) (Nurminen and Karjalainen 1972). Another big project was the morphology of the Finnish language: A generation of a computerised set of rules for verb and noun conjugation (Professor Kalevi Wiik) (Wiik 1967). Wiik was also working on Chomskian generative grammars, which were useful in pars- ing both natural and programming languages. My programming skills were challenged in the coding (using machine language!) of the driver for the speech synthesiser ‘Ove’. Many other smaller projects were part of my adventure into humanistic research. Probably the most im- portant lesson I learned was that whereas social sciences focused on collectives of people, in the Humanities, the single individual was a central and legitimate object of study. It is allowed to study a single book written by a single author without having to collect a representative sample of books and authors. Little could I see then the potential accumulating of competence around computers and the humanities that was later to significantly direct my research work and career. 2.2 Moving toward the Ph.D. My life with the Humanities was a dream, exactly like a child who is happy in a candy shop. It had one obvious disadvantage, however; the work was unfocused, so I had to jump from one project to another. I could not formulate a personal research trajectory for myself, so my knowl- edge was not cumulating towards anything that could become a PhD Dissertation. Then, in 1971, I decided to apply for a position as a lecturer, later an assistant professor, in information systems at the Turku School of Economics (TSE). At the same time, the Scandinavian School of Information Systems was taking its first baby steps. Börje Langefors had published the seminal monograph (Langefors 1970) Theoretical Anal- ysis of Information Systems, THAIS, in two volumes in 1970. Along with his core group, he or- ganised some very popular summer courses located in and including participants from all Scan- dinavian countries and Finland. I attended one of them in summer 1972 in Oppdal, Norway. This was a starting step to my own development as IS researcher. I started to write small arti- cles and publish them mainly in the publication series of the Turku School of Economics (TSE). Soon I had contributions that seemed to offer a basis of a PhD thesis. Yet it turned out that many of my colleagues at the TSE treated me as an outsider who was not properly initiated into the spirit of a management school. Instead of support, I experienced the sense that they were not very eager to let me present my thesis at the TSE. Then I went back to the university, where Professor Timo Järvi (Computer Science) welcomed me. There I presented my thesis in autumn of 1976 at the University of Turku and received a grade from a genuine philosophy (rather than a business) faculty. As an extra credit I even had a rare minor grade in the discipline philosophy. Work Informatics—Mon Amour • 59 Published by AIS Electronic Library (AISeL), 3 Scandinavian Journal of Information Systems, Vol. 29 [], Iss. 1, Art. 3 As a conceptual framework, I had adopted the theory of fuzzy sets, probably because the idea was quite popular during that time. Still, and perhaps more significant for my future work, was the reformulation of the Infological Equation (Dahlbom 1995) introduced by Börje Langefors. In his formulation the infological meaning of a message was constituted not only of the data, but also the time and place of the presentation. I wanted to add the purpose of information to this list. I illustrated my claim using the example of the monetary value of the company’s storage. That information may be requested by the financing manager, the insurance investigator, or the tax authority, each with different interests and goals. It is obvious, therefore, that the numerical value will be different in each case. This was the first step toward my later interpretation of in- formation and its processing as inherent parts of purposeful human activity. The opponent of this dissertation was Professor Pertti Järvinen from the University of Tam- pere. This contact was maintained, when I started to use the monographs written by Järvinen and Pentti Kerola (University of Oulu): Systemointi I and II in my teaching (Järvinen and Kerola 1975). Thus, I was not at all surprised when I received an invitation to Tampere for a summer seminar in 1978 that is better known now as the first meeting of the IRIS tradition. Even though the initiative came from Pertti and Pentti, I was the most active participant at these meetings over their first three decades by attending all of the first thirty seminars. For me, it was at the time an excellent forum to present my work in its early phase and receive solid feedback. IS research in Scandinavian countries was then divided into two fractions, namely, the Langefors group (THAIS) and the work-oriented action research done together with the local and later even with the national trade unions. The latter fraction soon became an iconic repre- sentative of Scandinavian IS research, whereas the Langefors approach led to several early formu- lations of IS development methodologies, quite similar to the later popular structural analysis and design methods. Probably the best known of these was the ISAC methodology. There was a certain tension between the two directions, because the trade union wing inter- preted the THAIS researchers as being committed to the owners’ interests, not to the workers. Interestingly, the IRIS meetings fostered both wings with only with minor friction. At several of the meetings of the Boards of the IRIS Association and the Scandinavian Journal of IS, we had debates about the orthodox interpretation of the Scandinavian School. Some members wanted a more restrictive policy; only the radical (often in terms of political meaning) enough approaches could qualify as truly Scandinavian. Each time a liberal view was selected (Iivari and Lyytinen 1998). For me, this choice was fine, because I hoped to receive tolerance of my sometimes-dissident approaches in an atmosphere that I could sense was not too exclusive. The Scandinavian School of Information Systems is probably impossible to define; it probably de- fines itself through its practice in the spirit of Wittgenstein. The openness described above made it also free to foster creative exploration of approaches that did not subscribe to many of the mainstream movements. Meanwhile (1978) the Turku School of Economics had established the position of associ- ate professor in Information Systems. Of course, I applied because I thought that in that role I could better contribute to the School and the discipline. I must say I was very disappointed when the School decided to nominate one candidate who came from another discipline, Man- agement Science. I had problems both in terms of trust and the motivation for collaboration, and I was happy when I received an office room in a small distant building. This destiny gave me an opportunity to enjoy daily coffee breaks spent with a small group of sociologists. These 60 • Nurminen http://aisel.aisnet.org/sjis/vol29/iss1/3 4 Nurminen: Work Informatics—Mon Amour discussions gave me broad and in some respects also deep insights into people problems; i.e.; sociological thinking. These new circumstances did not end my research activity. I was one of the organisers of the IRIS 2 in 1979 in Dragsfjärd. In IRIS 4 (Oulu), I presented a paper with the title entitled “Against System; A Human Perspective on Information Processing” (Nurminen 1981). It illus- trated the strictly questioning character of a young researcher with a strong self-confidence who questioned a predefined commitment to the development of an information system, typically with an integrated architecture. One could also anticipate the early formulation of Work Infor- matics in this title. My commitment to the work environment at the TSE was, however, low. I kept my eyes open for promising opportunities. That trigger came from a surprising direction: the University of Bergen (Norway) had a department that was shared by two faculties, the Social Sciences and the Humanities. They had two full professorships, one for each faculty orientation. I decided to apply for the Humanist chair although I must confess that I did not seriously expect to receive the position. However, the evaluators (Börje Langefors and Sture Allén) gave strong weight to my experience with the humanistic applications I had undertaken earlier. In January of 1982, I thus travelled to the west coast of Norway and started to teach and also figure out what human- istic research in information systems could be. My earlier summer job in Stockholm as a teen- ager turned out to have been a good investment, because I had learned then to use the Swedish language in actual practice. I could start at Bergen as a ‘competent Scandinavian’. 2.3 Humanistic research The Department of Information Sciences at the University of Bergen was quite exceptional. Two faculties with a spelled-out focus on human beings wanted to utilise and promote information technology. The origin of this policy may be found in the growing interest (especially in social sciences) in factor analysis and other multivariate analysis methods that were eagerly welcoming the calculation capacity offered by computers. This view was, however, very close to the situa- tion that I already had faced in my time as the mathematician for the Humanities. My clients then regarded computers as tools for solving research problems formulated by other disciplines. I wanted something more from my work. I wanted to make the phenomenon of information processing itself to the main objective of this research. I formulated my first version during my first term, Spring 1982, in the course entitled “Hu- man-scale Information Systems” (HIS). That version was ‘published’ in the form of local lecture notes. In the document, the new system architecture offered by HIS was analysed from multiple perspectives, among others, the ontological, epistemological, cognitive, behavioural, organisa- tional, sociological, and technical. HIS is a conceptual construct of system architecture that is decentralised to consist of subsystems that are used by one individual employee. Such system units have a set of locally stored files and application software. This design integrates the IT tasks to an inherent part of the work tasks of the related actors. In addition, the structure divides the responsibility for IS deployment to the respective actors. Communication between these very personal systems is naturally interpreted as organisational coordination. Work Informatics—Mon Amour • 61 Published by AIS Electronic Library (AISeL), 5 Scandinavian Journal of Information Systems, Vol. 29 [], Iss. 1, Art. 3 In the HIS construct, my concept of a human being takes its first steps towards the work role in Work Informatics. This notion builds on the person as the active subject of her life and work and is open to responsibility and competence. For example, her work is not regarded only from the outside as in many business processes, but also from the inside; values, knowledge, commit- ment, and understanding are relevant factors, as well as the will. The will is not completely free, however; it is restricted by physical (she can swim but not fly) and social (do not kill people, but keep your promises) constraints. One more contribution to the Bergen department was an organisation of the IRIS 6 seminar (1983 in Øystese, close to Bergen) with the student group from Bergen. This task can be seen in the context of the mission to make the originally Finnish idea about IRIS truly shared among all Scandinavian countries. The four first meetings took place in Finland. IRIS 5 took place in Stockholm, where Kalle Lyytinen happened to reside as a guest researcher. The next logical step of this Trojan horse trick had to be Nurminen in Bergen. After this step, Helsinki could take its turn again (1984, Markku Sääksjärvi). Then it was easy for Lars Mathiassen to invite IRIS 8 to Aarhus as an appendix to the second decennial Aarhus conference in 1985. During my Bergen years, I never really lived in this city. I worked two weeks in Bergen and then spent one week at home in Finland. The Scandinavian Airline System (SAS) helped me organise my travelling, so I could pay a visit to many places close to my landing airports on my way to or from Bergen. I could visit Oslo, Stockholm, Gothenburg, Lund, Copenhagen, and Aarhus. I remember these visits for their stimulating discussions with and the good contacts I made with many Scandinavian researchers. I still am thankful to all who were involved. This surfing across Scandinavian universities later led to many contacts. For example, I was invited to act as the opponent in multiple PhD dissertations. I choose to pick up just a few of these, and they became the most significant experiences for my own development: Göran Gold- kuhl (1980), Kalle Lyytinen (1986), Pelle Ehn (1988), Kari Kuutti (1994), and Susanne Bødker (Dr. scient. 1998). All five have had remarkable successes in their own research careers. I am convinced that this success is not simply from the merit of having me as their PhD opponent; rather I am happy for the opportunity to have learned of their early achievements. Once again, accidental, almost arbitrary, conditions outside the academic work itself turned out to have an effect on my career, this time my commuting. Today the international nomadic life is more likely to belong to the planned life cycle of research education. It opened the chal- lenge for me to explain my thinking to new people with different backgrounds. As an extra bonus I found both friends and a professional contact network. 2.4 Farewell to Bergen The experiences from my years in Bergen started to take the form of a framework made up of three perspectives to information systems with the title, People or Computers (Nurminen 1988). It was published 1986 in Finnish (WSOY) and in 1988 in English (Studentlitteratur), even if most of the material was already presented in the department in Bergen before I had to return back to Finland in 1984. This move happened partly because my family would not move to Norway and partly because I was unable to convince the faculty about my intention to turn the development and use of IT into the object of research from a humanistic perspective (Nurminen 62 • Nurminen http://aisel.aisnet.org/sjis/vol29/iss1/3 6 Nurminen: Work Informatics—Mon Amour and Rønning 1984). In the monograph, two alternative perspectives accompany the humanistic perspective. The framework of three perspectives for seeing information systems was directly derived from the HIS approach. These three perspectives were systems-theoretical, socio-technical, and humanistic. The humanistic perspective was fully consistent with the HIS architecture. The two other perspectives were needed as scaffoldings for the humanistic perspective; it had turned out that not even all my co-workers in Bergen were able to understand the HIS concept and see the role of paradigms in research. The revolutionary ideas concerning paradigms by Kuhn (1962) started to have an influence in various disciplines during the 1980’s, at least in the Scandinavian countries. The strength of the three parallel perspectives was their ability to give different interpreta- tions to one existing phenomenon, an IS in actual use. These perspectives were characterised by ideal types, and differences in the notions of knowledge, a human being, and organisation were identified. The three perspectives were constructed based on the relationship between the computer and the human being. In the systems theoretical perspective the main focus was the computer and the people needed to operate (serve) the computers. The humanistic perspective was the oppo- site of this viewpoint. The human being was at the core, and the computer was seen as simply a means to realise human needs. The intermediate perspective, the socio-technical one, aimed at finding a balance between these two extremes. Boström and Heinen (Bostrom and Heinen 1977a; 1977b) had introduced the main principles of socio-technical design applied to ISD, and Enid Mumford was successful in giving these ideas a more concrete methodology called ETHICS (Mumford 1986). A balance was created by the acknowledgement of two parallel sys- tems, the social and the technical. These systems were designed in separate tracks, and the most promising parts of both were finally connected to create a joint optimal solution to the original problem. It was a revolutionary setting, because most of the earlier approaches to the design of the social system had been based on the machine metaphor of organisation. According to this idea, organisations had been designed according to the same mechanistic principles as machines on a shop floor. The idea of two distinct systems argued that social systems are fundamentally different from technical systems, and therefore, they deserve to be designed using their own principles and methodologies. This invention had a huge impact on our discipline, as the practices of research and de- velopment of information systems became richer and certainly achieved improved quality in many ways. Yet at the same time, however, a socio-technical trap emerged as an unintended consequence. The setting of two parallel systems implied that they were different, sometimes even distinctively so, from each other. The technical system thereby received an autonomy that made it independent of the social system. The technical imperative seemed to be to generate neutral and unquestionable objectives. In my mind, behind these objectives were the interests of human beings, and they were fundamentally social. The social origin of technically designed information systems has still too often, therefore, disappeared from sight, and interpreted in terms of Berger and Luckman (1966), so they were reified, which means inhumane. Most of my thinking is based on the idea that all technical systems are essentially social systems and (there- fore, hopefully) controlled by their users. This Credo implies for me that IS research has always been a part of my own social and political activity. Work Informatics—Mon Amour • 63 Published by AIS Electronic Library (AISeL), 7 Scandinavian Journal of Information Systems, Vol. 29 [], Iss. 1, Art. 3 There is an interesting overlapping in the titles of the three perspectives and the three tradi- tions presented by Jørgen Bansler (1987 and 1989). Two titles have a full coincidence: Systems theoretical and Socio-technical. Bansler’s third class bears the name ‘Critical Tradition’, whereas Nurminen’s third perspective is considered ‘Humanistic’. This difference indicates that Nurmi- nen does not, along with the humanistic perspective, subscribe to the critical political agenda embedded in the critical tradition. The humanistic perspective aims at improving the actors’ understanding of their work situation both broadly and deeply. Then the actors are free to enter and become better equipped in the negotiations of their future work situation when any unin- tended consequences also become visible. I am fully aware that this stance includes a flavour of Hume’s guillotine, but if we do not keep different domains of activity apart from each other, we may find ourselves searching for partners to new coalitions; e.g; political parties, immigrants or sexual minorities. I was lucky to receive the position of associate professor at the University of Turku in 1983. I started to document my Norwegian experiences in the monograph. In parallel terms, I wanted to initiate empirical research into the three perspectives, in particular the Humanistic Perspec- tive. This circumstance created a dilemma. It was impossible to find such applications to study that were designed according to humanistic perspectives simply because none of them existed. Our research team had to find out how to undertake empirical research about an object that did not exist. The following strategy was thus outlined: First study the use situation of an information system already in use. Collect information about that use situation paying special attention to the work tasks of the users, both in individ- ual and collective settings. Then analyse the structure and functioning of the system in detail. Problem situations and breakdowns are interesting, because they make the work aspects of the IS visible. Using all this information about the current use of the IS, translate the description into the HIS format by de-integrating the data bases into personal system units and dividing the system functions so they constitute a part of the division of labour between the employees. This reconstruction is called the social interpretation of the IS and the work situation, and it was generated to be used as a reference in a comparison with the original situation. The Academy of Finland approved a grant to our research group for the project “Knowledge and Work” for three years 1986-1990. The value of IS research is ultimately demonstrated in the practical use of information tech- nology. Therefore, our research group wanted to go to actual users. It happened that we were invited to a dozen companies and other organisations to help in their deployment of existing information systems. Most of these cases were contracted research; i.e.; the customer paid our contract in full. Our niche was known as Social Interpretation and was a variation of the Human- istic Perspective. It was created by reducing computer functions to the actions of the users, often by imagining the corresponding tasks that had to be done without computers. Such an alterna- tive interpretation makes the work and its coordination visible in a manner that is independent of the current computer-based solution. It can be used as a benchmark for evaluating the benefit of the current system, namely, to what extent does the system contribute to the objectives of the organisation? Most customers were satisfied. We found really dramatic problems in actual use situations, however, more often than once, we had to make an intervention to successfully rescue the re- sponsible information manager. The top managers intended to fire him/her because s/he had 64 • Nurminen http://aisel.aisnet.org/sjis/vol29/iss1/3 8 Nurminen: Work Informatics—Mon Amour allowed these big scandals to happen. We told them that similar scandals occur in practically all organisations, so it was nothing extraordinary. The direct financing by the companies could also be seen as a proof of the relevance of our approaches. All users at different organisational levels were happily enlightened in their understanding the new interpretation. No direct confron- tation between the different groups emerged during our interventions, perhaps because both workers and their managers wanted to do their work well and professionally. Since this goal was their primary interest, the other contradictions potentially surfaced later. The integrated structure over user groups at multiple levels of the organisation was described using an ONION model; it included a description of the activity and its objectives at each level, accompanied by an evaluative view of the related parts of that IS. 2.5 The Jyväskylä intermezzo At the end of the 1980’s, I sent an application for the position of full professor to the Univer- sity of Jyväskylä. The IS department by then had grown to a significant centre of research and teaching in Information Systems, thanks much to Kalle Lyytinen. The nomination was rapid and dramatic: I made a complaint on the selection procedure, it was accepted, and suddenly I was nominated (1989) and lost my previous position in Turku. The time was economically critical in Finland. It was practically impossible to sell our house to finance the family’s move to Jyväskylä, so I had to start commuting again. My co-workers elected me to be the head of the Jyväskylä department, and there I learned a lot about university administration. The weekly travelling of four hours on the train was just enough time to read and comment on one manuscript submitted for a Master’s thesis. Again, I was able to benefit from my travelling, as Tampere was in the middle of the journey between Turku and Jyväskylä. Almost every week I stepped off the train to have a meeting in the cafeteria of the railway station in Tampere with Pertti Järvinen. We organised a conference on behalf of IFIP WG9.1 (Com- puters and Work). I held the program chair whereas Pertti was the real organiser. The conference entitled “Human Jobs and Computer Interfaces” took place in Tampere in 1991. Of course I certainly was worn out after two years 1990-92, but I was happy when I had the opportunity to get back my old position again in Turku. In Scandinavia, contacts in the informal tradition of reoccurring seminars started to became more institutionalised. In the 1988 IRIS9 meeting in Røros the Scandinavian Journal was estab- lished, and soon the IRIS Association received its first bylaws and Board. Both steps gave IRIS community continuity and identity. I had the privilege to be a part of both of these emerging traditions. In Turku we established an organisational unit within the University of Turku, called Labo- ris (labor=work, is=information system) to produce research with a focus on the relationship between work and IT, in order to have a credible counterpart when negotiating with our cus- tomers. Later we established a full-blooded commercial company. I was a stakeholder (minority owner and Board member) of that company; Ulf Forsman was the CEO). The activities of the firm were successful, and after three years we received an offer given from a great company that gave us owners multiple compensation for our investment. Work Informatics—Mon Amour • 65 Published by AIS Electronic Library (AISeL), 9