Special Issue On Distributed Participatory Design, SJIS 2009, Issue 1

C o n t en t s :
Scandinavian Journal of Information Systems 2009 Vol. 21, No. 1

Editorial 1
Consumers' Views on Privacy in E-Commerce 3

T. Kaapu & T. Tiainen
A Special Issue Editorial 23

K. D. Öberg, D. Gumm & A. M. Naghsh
Distributed Development to Enable User Participation: Multilevel design in the 27

HISP network
O. H. Titlestad, K. Staring & J. Braa
Inter-Contextual Distributed Participatory Design: Communicating design 51

philosophy and enriching user experience
H. Obendorf, M. Janneck & M. Finck
Furthering Distributed Participative Design: Unlocking the walled 77

gardens
C. Loebbecke & P. Powell Scandinavian Journal of Information Systems
2009
Vol. 21, No. 1
Established 1989
ISSN 0905-0167
Scandinavian Journal of Information Systems
Publishing
Copyright © 2009 Scandinavian Journal of Information Systems. The IRIS Association, Aal-
borg University, Department of Computer Science, Selma Lagerlöfs Vej 300, DK-9220 Aal-
borg, Denmark.
ISSN: 0905-0167
ISBN: 9788799142576
Publication date: 30th June, 2009.
Production team: Peter Axel Nielsen (director) and Annette Moss (editing).
Electronic journal: www.e-sjis.org; available free of charge.
See www.e-sjis.org for scope, submission of articles, and subscription to the printed journal.
Editorial Board
• Bjørn Erik Munkvold, University of Agder, Norway, Coordinating Editor.
• Rikard Lindgren, IT University of Gothenburg & Viktoria Institute, Sweden, Editor.
• Keld Bødker, Roskilde University, Denmark, Editor.
• Samuli Pekkola, Tampere University of Technology, Finland Editor.
Advisory Board
• Richard Boland, Case Western Reserve University.
• Tone Bratteteig, Oslo University
• Susanne Bødker, Aarhus University.
• Bo Dahlbom, IT University of Göteborg.
• Jan Damsgaard, Copenhagen Business School.
• Jukka Heikkila, University of Jyväskylä.
• Ola Henfridsson, Viktoria Institute.
• Karl Kautz, Copenhagen Business School.
• Finn Kensing, University of Copenhagen.
• Karlheinz Kautz, Copenhagen Business School.
• Kalle Lyytinen, Case Western Reserve University.
• Lars Mathiassen, Georgia State University.
• Eric Monteiro, Norwegian University of Science and Technology.
• Joe Nandhakumar, Warwick Business School.
• Ojelanki Ngwenyama, Ryerson University.
• Peter Axel Nielsen, Aalborg University.
• Sandeep Purao, Penn State University.
• Matti Rossi, Helsinki School of Economics.
• Ulrike Schultze, Southern Methodist University.
• Jesper Simonsen, Roskilde University.
• Susan Leigh Star, Santa Clara University.
• Erik Stolterman, Indiana University.
• Carsten Sørensen, London School of Economics.
• Pål Sørgaard, Telenor Research & Innovation.
• Virpi Tuunainen, Helsinki School of Economics.
• Geoff Walsham, University of Cambridge.
• Youngjin Yoo, Temple University.
Editorial
Keld Bødker
Rikard Lindgren
Bjørn Erik Munkvold
Samuli Pekkola
The first issue of the SJIS volume 21 features one research article and a special issue on challenges
and opportunities of distributed participatory design. The research article by Taina Kaapu and
Tarja Tiainen presents a phenomenographical analysis of consumers’ interpretation of privacy in
e-commerce, identifying a number of different privacy conceptions. Their results illustrate how
the consumers’ view of privacy is situated and dynamically constructed based on new informa-
tion and experience. The guest editors of the special issue were Karin Danielsson Öberg, Dorina
Gumm and Amir M. Naghsh. We thank them for their contribution to the journal, and the
considerable efforts they have put into serving as guest editors.
With this issue, Bjørn Erik Munkvold has taken over from Matti Rossi as coordinating edi-
tor for SJIS. We are grateful to Matti for his valuable service for the journal and the community,
and welcome Samuli Pekkola as the new Finnish representative on the SJIS editorial board.
© Scandinavian Journal of Information Systems, 2009, 21(1), 1–2

2
Consumers’ Views on Privacy in E-Commerce
Taina Kaapu
Department of Computer Sciences, University of Tampere, Finland
taina.kaapu@uta.fi
Tarja Tiainen
Department of Computer Sciences, University of Tampere, Finland
tarja@cs.uta.fi
Abstract. Information privacy protection and invasion of privacy in e-commerce have be-
come important topics in both everyday activities and scientific discussions. The aim of
this study is to understand how consumers regard privacy in business-to-consumer e-com-
merce. As this study focuses on consumers’ own interpretations of privacy, the research
approach is empirical, rather than theoretical. Based on a phenomenographical analysis
of consumer interviews, we identify different layers of understanding by focusing on the
referential objects and the structural components of information privacy. The result in-
cludes 25 different privacy conceptions, showing that consumers’ view of privacy is situated
and constantly under construction as the consumer gets new information or experiences.
Key words: information privacy, consumer, e-commerce, phenomenography.
1 Introduction
Understanding consumer behaviour is vitally important in online shopping. One precondition
for the growth of e-commerce is that the consumers use online channels. These channels are
chosen at each stage in the purchase process: requirements determination, vendor selection,
purchase, and after-sales service (Choughury and Karahanna 2008). The process is connected
to the consumers’ trust in e-purchase and e-vendors, and to the consumers’ perceptions of risk
(Choughury and Karahanna 2008; Verhagen et al. 2006). The number of online consumers has
grown; at the same time, the fears regarding information privacy have also increased (Malhotra
et al. 2004). The biggest concerns to Internet users are viruses, spam, spyware and hackers (Paine
et al. 2007). If these problems are not solved, the consumers whose privacy concerns have not

been addressed may delay their purchases or even forgo them, and some concerned consumers
might prefer traditional ways of purchasing (Prabhaker 2000).
To deal with these concerns, privacy enhancing technologies (PETs) have been developed:
software programs, hardware devices and even publications, which help users to regain their pri-
vacy lost on the Internet (Camp and Osorio 2003). Legal instruments for increased security have
been formulated as well: for example, the European Union (EU) requires all its member states to
legislate to ensure that their citizens have a right to privacy (Directive 95/46/EC).
On a practical level, e-vendors work to increase online purchasing. Research in informa-
tion systems (IS) and consumer studies aims to increase understanding about e-commerce and
consumers’ online behaviour (see e.g., Cassidy and Chae 2006; Hui et al. 2007; Malhotra et al.
2004). Our paper belongs to the same research area, although we focus solely on information
privacy. According to the traditional definition, information privacy is the ability of the indi-
viduals to control information about themselves (Westin 1967). Instead of concentrating on tra-
ditional and direct marketing, as done in several former studies (e.g., Smith et al. 1996, Stewart
and Segars 2002), we seek to understand how consumers view information privacy in business-
to-consumer (B2C) e-commerce. In doing so we aim to present and discuss the subject matter so
that business and legislative authorities can adequately respond to and address these consumers’
needs and fears. This is necessary to allow maximizing the potential of e-commerce.
Theory testing with surveys is a commonly used research method in studying Internet pri-
vacy. A typical study asks about informants’ attitudes towards specific privacy statements with
fixed scale (e.g., Cassidy and Chae 2006; Malhotra et al. 2004). In theory testing studies, the
researcher—based on the theory under testing—defines how information privacy is conceptual-
ized. However, it is important to take a step back and investigate how consumers understand
privacy in everyday practice.
To fill this gap in the literature, we decided to use a qualitative research approach for getting
a richer picture of consumers’ views. Instead of seeking the dominant view or an average one,
we focus on the differences in views. We decided to concentrate on the variation of consumers’
interpretations as consumers are not a homogenous group. In most consumer studies, consum-
ers are divided to groups based on their demographical variables, income or attitudes. Westin
categorized consumers based on their attitudes towards privacy; to categories of privacy funda-
mentalists (who feel that they have lost their privacy), privacy pragmatists (who protect their
personal information), and privacy unconcerned (who have no real concerns about privacy)
(Taylor 2003).
Our aim is to focus on all types of consumers’ views on information privacy without cat-
egorizing consumers beforehand. First, we seek what earlier studies say on consumers’ inter-
pretations of information privacy. In doing this, we briefly describe the concept of privacy as
discussed in the previous literature. Then, we describe the research methods used in our study:
as we look for alternative views, we use phenomenography and consumer interviews. After the
methodological part, we offer the results in a form of categorization of consumers’ views based
on the analysis of the interview material. The results show that consumers’ interpretation of
information privacy is situated; it varies between familiar, trusted cases and unknown cases per-
ceived as suspicious. Finally, we present the discussion and conclusion.
4 • Kaapu & Tiainen

2 Research background
In this section, we describe the scope of Internet privacy studies for locating our study to the
appropriate scientific field. Privacy is characterized according to the traditional definition as “the
claim of individuals, groups, or institutions to determine for themselves when, how, and to what
extent information about them is communicated to others” (Westin 1967, (pp. 6-7). Invasions
of privacy occur when individuals cannot maintain an adequate degree of control over their
personal information and its use (Chung 2003). We follow the generally accepted view of infor-
mation privacy by seeing it as the ability of individuals to control information about themselves
(e.g., Cheung and Lee 2006; Graeff and Harmon 2002; Udo 2001).
Online information privacy has been studied in the disciplines of law and public policy,
marketing, organisational behaviour, and IS (Malhotra et al. 2004). One way to see privacy is
to understand it as a legal concept (e.g., Curran and Richards 2004). Although the concept of
privacy itself may sound straightforward, the regulatory laws vary between cultures (Milberg et
al. 1995). Developed societies have made different assumptions about privacy in their societal
regulatory approaches. The societies can be roughly sorted into two categories: First, to those
who view privacy as a human right, as is the case in the EU (Bygrave 1998), for example, where
measures that address all the data collection and use within society are being introduced. Sec-
ond, to societies which view privacy as a matter for contractual negotiation, such as the United
States and Japan, for example, where the laws are specific to various sectors (e.g., medical data)
(Smith 2004).
There exist contradictory views on privacy and benefits for consumers, for example about the
collection of personal information. It can be seen as a positive matter, since personalized serv-
ices cannot be created without personal information. However, consumers’ hopes in this area
are paradoxical: easy, personalized services are in demand but collecting personal information
is resisted (Awad and Krishnan 2006). While the freedom of movement of information and its
benefits to the general public have been emphasized (e.g., Bergkamp 2002; Rubin and Lenard
2002), the somewhat opposite view sees personal information registers as unreliable, and the
aim of the laws has been to limit their use. Thus, the latter view focuses on threats such as more
widespread profiling when handling personal information (e.g., Graeff and Harmon 2002; Liu
et al. 2005) and consumers’ continuous on-line monitoring (e.g., Kruck et al. 2002; McRobb
and Rogerson 2004; Smith 2004).
Consumers’ lack of trust constitutes a major psychological barrier to the adoption of e-
commerce (Cheung and Lee 2006). Consumers’ privacy concerns have been studied with theo-
retically based surveys, with varying results (e.g., Udo 2001; Malhotra et al. 2004). For example,
consumers’ privacy concerns are stated to be related to the following aspects of data collection
and use (Smith et al. 1996; Stewart and Segars 2002): 1) unauthorized collection, 2) errors
related to the integrity of databases, 3) unauthorized secondary use, and 4) improper access to
personal data. Some studies present concerns in a more concrete way, as listed in the following
(e.g., Cassidy and Chae 2006; Chung 2003):
Consumers’ Views on Privacy in E-Commerce • 5

1. Visits to websites might be secretly tracked.
2. E
-mail addresses and other personal information could be taken and used without per-
mission for marketing or other purposes.
3. Personal information could be sold without permission to third parties.
4. Credit card information could be stolen.
One of the other types of categorizations is presented by Paine et al. (2007), which states that
the consumers’ main concerns about online privacy are viruses, spam, spyware and hackers.
Malhotra et al. (2004) conceptualize Internet users’ information privacy concerns as
The degree to which an Internet user is concerned about online marketers’ collection of
personal information, the user’s control over the collected information, and the user’s
awareness of how the collected information is used (Malhotra et al, p. 338).
They also developed a causal model to describe how concerns influence a consumer’s decision to
release or not release personally identifiable data.
E-vendors can do a lot for mitigating the consumers’ fears related to privacy. Information
on how companies maintain and use personal information increases consumers’ trust (Liu et al.
2005). The presence of a vendor’s online privacy policy decreases consumers’ privacy concerns
(Hui et al. 2007; Jensen et al. 2005; Pan and Zinkhan 2006). At least in some cases, consumers
trust e-vendors (e.g., Gefen et al. 2003) and are not afraid of privacy problems with them, such
as e-vendors selling personal information to third parties (Cheung and Lee 2006).
The above studies present at least partly contradictory findings, which makes consumers’
views on information privacy an important issue to study further. Consumers’ behaviour is usu-
ally studied in a conventional direct marketing environment (Phelps et al. 2000). These stud-
ies are most often based on demographical factors, and the differences that are found include
(Graeff and Harmon 2002):
• Gender differences: men are less concerned about privacy issues than women, and men
have more faith in purchasing on the Internet;
• lass differences: the consumers with high incomes want to know more about their
C
information after collection than other consumers;
• Age differences: older people are less likely to believe that their information might be
sold to others for marketing purposes.
Furthermore, the global nature of e-commerce makes privacy issues even more complex, because
the perceptions of privacy and fair information practices depend on government regulations and
vary across cultures (Bellman et al. 2004; Milberg et al. 1995).
Information privacy is focused in different areas, such as consumers’ behaviour (see Table
1 for a summary). Although some studies (e.g., Jensen et al. 2005; Paine et al. 2007) ask for
a deeper understanding of consumers’ thought-models, almost all studies have used a theory-
testing research approach with surveys or laboratory tests. The only exception we could find is
the study by Hui et al. (2007), in which field observations were made in a local firm focusing on
privacy statements. Our study belongs to the same empirically based approach among qualita-
tive studies, focusing on the consumers’ own interpretations of information privacy. In this ap-

proach, rather than building our research on some earlier studies with their underlying assump-
tions, we must be as open minded as possible to reach the consumers’ own thought-models. Our
aim is to find out what these consumers exactly mean when they discuss information privacy.
However, the concept of information privacy may signify different issues or concerns to
different people. This paper aims to clarify this with the help of a categorization of consumers’
views of information privacy in e-commerce. The study has both theoretical and practical con-
tributions. On the theoretical side, we discuss the consequences of our empirical results vis-à-vis
the existing literature. On the practical side, we give important guidelines in order to understand
consumers’ concerns about privacy in the online environment.
Areas of information privacy studies Example references

1. Defining privacy Westin 1967
2. Legal issues about privacy Bygrave 1998
Cassidy and Chae 2006
3. Technology for privacy (Privacy Enhancing Technologies) Camp and Osorio 2003
4. Vendors’ actions to increase privacy McRobb and Rogerson 2004
Prabhaker 2000
5. Consumers’ actions in relations to privacy, which is Awad and Krishnan 2006
studied based on: Graeff and Harmon 2002
- Consumers’ demographic factors Taylor 2003
- Consumers’ attitudes, values
and behavior
6. Consumers’ privacy concerns: Chung 2003
- Personal information: collection, unauthorized Malhotra et al. 2004
secondary use, improper access, errors, stealing Paine et al. 2007
- On-line monitoring Stewart and Segars 2002
- Viruses, spam, spyware and hackers
Table 1: Focal areas of information privacy studies
3 Methodology
Our aim is to understand how consumers see privacy and to describe the differences in their
views in e-commerce. For studying people’s own interpretations of a concept (information pri-
vacy in this case) a qualitative method which focuses on people’s narration is needed. Indi-
viduals’ views are socially constructed; however, the individuals’ own background, including
their education and experience, can also have effect on their views. Methods such as discourse
analysis, grounded theory (GT), and phenomenography can thus be considered. In choosing the
research method, we discarded discourse analysis, since it has its focus on social interaction, such
as shared (communal) views and argumentations (Alvesson and Karreman 2000).
A decision between GT and phenomenography was made based on their different study
aims. Although they both itemize individuals’ talk to its elements, the target of that process

is different. GT re-ties the elements together for a whole picture of the phenomenon under
study (Glaser and Strauss 1973), whereas phenomenography aims for reaching all the alterna-
tive views or the structures of individual thought-models (Marton and Booth 1997). As some
thought-models are richer and more versatile than others, the result forms a hierarchical struc-
ture. Phenomenography is about individual meaning construction, which results in a concep-
tion referring to conceiving and understanding something. Humans’ experience of the world
is constituted as an internal relation between the experiencing people and the world (Marton
1981). Conceptions are regarded as ground for action (Säljö 1994).
Phenomenography was introduced by educational researchers (Marton 1982). It has been
used for educational studies also in the IS field in clarifying computer science students’ concep-
tions of recursion (Booth 1992) and in finding out about moral conflicts in the project work
course (Vartiainen 2007). The method is also used for analysing IS professionals’ assumptions
about the human being (e.g., Isomäki 2002). Here we use the method for analysing IS users’
views. Our study focuses on non-professional people whose knowledge of technology is lim-
ited; they might have erronous views about what is possible and what is not over the Internet.
Regardless, in this study we do not evaluate the workability of their presented assumptions. It
is enough that the informant believes in them, and that they thus may affect his/her behaviour.
These concepts will be made explicit to allow IS professionals to understand them.
In phenomenography, empirical material is typically collected by interviewing a relatively
small number of relevant informants. The main point when choosing these is to reach the largest
possible differentiation in their views (Marton and Booth 1997), similar to theoretical samples
in other qualitative methods (see e.g., Glauser and Strauss 1973). We maximize the differentia-
tion in privacy conceptions with the help of two interview settings. The first one focused on
privacy, and the views there were directly elicited from the interviewees: they were asked to de-
scribe privacy in the Internet setting in their own words. When a question about a phenomenon
is asked directly, there is a chance that the informants repeat the dominant discourse of the issue;
this can be regarded as the shared view by the society (Hynes et al. 2006). In this study the aim
was to reach all the alternative views: therefore, besides of direct asking, we also discussed privacy
indirectly. The second interview focused on the use of electronic services; privacy was expected
to be an underlying assumption to emerge in the interviews.
The interviewees were sought in several ways. One criterion in the selection was age. Unlike
in many studies that use young informants (e.g., Cheung and Lee 2006; Gefen et al. 2003), we
also sought older people’s views. Having them in our interview group increased the chance of
eliciting a larger variety of concepts: the older people seem to be the most passive in data protec-
tion issues while well-educated, young, heavy Internet users are the most active group (Grable
and Joo 1999; Muttilainen 2006).
In the first interview set, individuals known by the researchers were directly asked for inter-
views. Also, the snowball method was used; the interviewees were asked to name other possible
interviewees, especially those who might have (different) opinions on privacy. The first interview
set based on direct questions about information privacy included twelve interviews.
For the second interview set, volunteers were found by advertising on a local newspaper’s
website and in an e-commerce seminar. Volunteers needed to fill in an Internet form indicating
their age, sex, and values. For identifying the values, the volunteers rated each of the listed values
and specified the most and the least important values. The list of values included: 1. sense of

belonging, 2. excitement, 3. warm relationship with others, 4. self-fulfillment, 5. being well-re-
spected, 6. fun and enjoyment in life, 7. security, 8. self-respect, and 9. sense of accomplishment
(based on List of Values (LOV) by Kahle et al. 1986). In the selection of interviewees our aim
was to maximize the differentiation in participants’ views. As the former studies highlight in-
formation privacy concerns, we decided to use two values that are related to safety (security and
excitement). Of the twenty volunteers who filled in the form we selected five security-minded
and five excitement-minded consumers. Their interview focused on the use of e-services.
We did not study the relationship between the interviewees’ background and the privacy
view, so the interviewees’ backgrounds are not significant for the analysis. However, we are aware
that the variation among interviewees’ background might help in reaching the largest possible
variation in views. A brief description of their backgrounds is shown in Table 2. This may help
the reader to better understand the empirical base of this study.
The aim of phenomenographical study is to describe the differentiation between individuals
as regards the phenomenon under study. In collecting data, the interview situation affects what
people say and how they say it; furthermore, due to bias caused by personal education and other
background, researchers may ignore some ideas mentioned by the informants (Eriksen 2001).
To minimize this problem, we varied the interview situation: in the first interview set there was
one interviewer, and in the second set there were two interviewers.
Interviewees Female Male Total

Age
Between 25 and 35 years 4 4 8
Total 13 9 22
Occupational background
Business and administration 4 4 8
Health and social affairs 4 2 6
Teaching and education 3 1 4
Technological sector 2 1 3
Agriculture - 1 1
Total 13 9 22
Table 2: Consumers interviewed
Otherwise, the interviewing progress was alike in both of the interview sets. The individual in-
terviews were open-ended, and only the topics were decided beforehand. The interviews started
from a general discussion about the interviewees’ backgrounds (as consumers in e-commerce)
and were followed by a discussion about the main issue (which concerned, in the first set,
privacy issues in e-commerce and, in the second set, e-services in the form of e-journals and e-
commerce). The duration of the interview situations varied from thirty minutes up to two hours
and thirty minutes. The interviewer’s role was to follow the interviewees’ ideas and explore their
narration.

The interviews were collected during spring and summer 2004 in Finland. They were tran-
scribed and their text then analysed. In phenomenographical studies, the analysis focuses on
two components in the informants’ experiences of the phenomenon; the referential compo-
nent—which describes what the phenomenon means in everyday language—and the structural
component—which refers to a deeper level of phenomenal meaning (Marton and Booth 1997).
The what aspect directs individuals’ thought to the object, which can be physical or mental by
nature. The how aspect refers to the thought processes by which an object of thought is limited
in relation to its environment (Marton 1981). In phenomenography, the conceptions are in-
tentional with respect to the two intertwined aspects, which signify the qualitative differences
among conceptions. The aspects render the relation that a conception constitutes between an
individual and the surrounding world as contextual (Marton and Booth 1997).
In our study, at first the focus in analysis was on the referential component, i.e., on what the
interviewees meant with privacy on the level of everyday language. They described the details of
privacy and the problems related to keeping their privacy. The interview texts were split in small
items—each of them included one aspect or a problem of privacy. The items were categorized
in order to obtain a single dimension of the categorization at time—first, the what aspect of the
final categorization. The analysis continued by focusing on the structural component of privacy
views. Structure is reached by analysing the target of the referential component. In the final
categorization, this is the how aspect.
4 Result: Categorization of consumers’ views

Our study deals with how consumers see information privacy. The results present different layers
of understanding in two dimensions. The first dimension is the referential component which
focuses on the meaning of privacy in the interviewees’ everyday language (Table 3 columns
A-E; What). The second dimension is the structural component which focuses on the form of
What A. Use and
D. Danger of
misuse of B. Monitoring C. Threat of E. Risk with
hackers and
customer consumers spam payment
viruses
How information
1. Product and
1A 1B 1C 1D 1E
e-vendor
2. Technology 2A 2B 2C 2D 2E
3. Societal 3A
3B 3C 3D 3E
norms
4. Consumer 4A
4B 4C 4D 4E
him/herself
5. Fellow men 5A 5B 5C 5D 5E
Table 3: Summary of the categorization of consumers’ information privacy conceptions in

e-commerce

thought when the interviewees talk about privacy (Table 3 rows 1-5; How). By the two dimen-
sions we identify a total of 25 different privacy conceptions. Table 4 includes one example of
each conception to illustrate the content.
The referential component (the what aspect) of information privacy consists of five objects.
Use and misuse of customer information includes consumers’ personal information and how it
is used, especially concerning its misuse. The second object, monitoring consumers, refers to
monitoring consumers’ actions when they are using e-commerce systems. Threat of spam refers
to e-mails that a consumer interprets as spam as a part of privacy demands. Danger of hackers and
viruses relates to hackers (who spread viruses) as a factor to intrude privacy. Risk with payment
refers to issues related to making payments on the Internet that can create problems of privacy.
The other dimension, structural components (rows in Table 3, or the how aspect) shows differ-
ent forms of thought and stress the structural aspect of the conceptions. It includes five objects.
When a consumer gives personal information in order to purchase a product in a certain e-shop,
the focus is on products and maybe also on the e-vendor. Furthermore, there are systems (or
technology) to deliver the product to the right person. It may be regarded as safer to order from
the home country than from abroad (societal norms), the consumer may make an error when
writing the order (consumer him/herself ), and there may be a family member watching over his/
her shoulder (fellow men).
4.1 Referential objects of information privacy

The referential objects (Table 3: columns) are presented in the order of which they were empha-
sized by the interviewees as a group, taking into account how much they discussed each object:
• All (22) interviewees talked about use and misuse of customer information.
• 20 interviewees talked about monitoring consumers.
• 15 interviewees talked about threat of spam.
• 14 interviewees talked about danger of hackers and viruses.
• 9 interviewees talked about risk with payment.
When the interviewees described information privacy, in most of the cases they connected it
to possible problems. Four categories of the referential components focus on problems only
and just one on both use and misuse. The result of our study does not state anything about the
frequency of specific views among the whole population; however, all the five referential objects
exist in people’s thinking about privacy.
Object A: Use and misuse of customer information. In the interviews, the consumers mostly
discuss their personal information as customer information. This is information which e-ven-
dors collect by asking it from consumers. Some of the interviewees use the concept “customer
information”, others underline their own viewpoint by talking about “my information that my
e-vendor has or knows”. In general, the interviewees are reluctant to give their information
and they are afraid of misuse of their personal information – but only if they do not know the
e-vendor beforehand. Nevertheless, the consumers interviewed also understand the benefits of

Content illustrated with an
Conception
example from the interviews
Use and
The consumer gives personal details, e.g., when:
misuse 1A (product)
- he/she is familiar with the product.
of customer
2A (technology) - the system of registration is easy to use.
informa-
3A (soc norms) - he/she feels that laws ensure security.
tion
4A (consumer) - he/she can give as little information as possible.
5A (fellow men) - his/her son helps to log in to e-vendor’s pages.
The consumer feels that the use of e-commerce and personal
1B (product) information are monitored when:
- he/she is interested in products such as explosives.
Monitoring 2B (technology) - he/she thinks that the e-vendor gets logs of web visits without
consumers consumers’ knowledge.
3B (soc norms) - a certain intelligence service may be monitoring.
4B (consumer) - he/she is not able to see the statistics of visits.
5B (fellow men) - a family member is watching over the shoulder.
The consumer receives mail that disturbs and invades personal
1C (product) privacy when:
- he/she participates in the e-vendor’s lotteries.
Threat of 2C (technology) - the system in his/her computer does not work properly.
spam 3C (soc norms) - he/she uses foreign e-services, for example, newspapers.
4C (consumer) - until he/she started to use the filter program.
5C (fellow men) - until the son installed the filter program.
The consumer believes that hackers use viruses to steal personal
1D (product) information when:
Danger of - the consumer uses “some strange” e-services.
hackers and 2D (technology) - the computer is using a wireless connection.
viruses 3D (soc norms) - the consumer uses foreign e-vendors.
4D (consumer) - the consumer does not have a firewall.
5D (fellow men) - the son hasn’t installed virus protection.
The consumer is not afraid to use e-bank and give personal
1E (product) information when:
- he/she knows that the e-vendor is reliable.
Risk with 2E (technology) - the payment system uses secure actions.
payment - the payment transaction is conducted within one’s own
3E (soc norms)
country.
4E (consumer) - he/she is familiar with the secure transactions.
5E (fellow men) - he/she asks advice from a net community.
Table 4: The contents of the conceptions illustrated with interview examples
getting personalized offers from e-vendors. One problem is that giving information is compli-
cated (Conception 2A):

But it (the registration) must not be made too difficult or complicated. There are so
many registration forms; fill in this field, this one, that one, so I won’t do it. I think, leave
it. Anyway, I can’t stand writing my whole biography to some registration (forms).
Object B: Monitoring consumers. The interviewees describe two kinds of monitoring. They
tell that some one may watch when they are using the computer. This is the same as traditional
monitoring: the one who is monitored and the one who monitors are in the same physical space.
Besides of the traditional monitoring the interviewees are concerned if there exists virtual moni-
toring which happens over the Internet. For example, the interviewees state that some e-vendors
keep an eye on consumer purchase behaviour or that the police monitors the Internet. Virtual
monitoring means that the visits to websites are monitored secretly and information about In-
ternet use is added to visitors’ personal information. The interviewees claim that it should be a
fundamental right (of Internet privacy) to visit web pages anonymously.
Some of the interviewees are more afraid of monitoring than others. The most careless ones
say that monitoring occurs only on a small scale or that their personal information is not rel-
evant to strangers. Some interviewees admit that they do not know enough – and they also do
not like giving their personal information, because they do not actually know what happens to
the information or what is possible to do over the Internet. One of the interviewees gives an
example where her own level of knowledge about technology affects her concerns with consumer
monitoring (Conception 2B):
I’ve heard about user tracing. When I hadn’t used the net for a very long time I got a
notification saying ‘you are running out of ink’. Oh my god, I thought, did I run out of
ink so fast? Then I realized that it was an advertisement. I’m still wondering if they could
know whether I was running out of ink.
Object C: Threat of spam. All the interviewees agree that unwanted e-mail is annoying and
most of them see a conflict between spam and privacy. However, they clearly differentiate be-
tween spam and other uses of customer information (object: use and misuse of customer infor-
mation); especially when the unsolicited mail or directed ads originate from their own e-vendor,
it seems acceptable. We use the word “spam’ because the interviewees cite the term; in addition,
“garbage” and “junk” were also used. Spam usually refers to unwanted e-mails, which causes
harm at least by filling, and occasionally choking (malicious attacks), one’s mailbox. Still, it is
difficult to determine whether an e-mail is wanted or not; for defining that, the recipients’ inter-
pretation is needed. Besides of defining what spam is, protection against it, and the consumers’
own actions are important, as seen in the next quotation (Conception 3C):
Of course, one factor is that if lots of spam start coming. -- But I haven’t done any busi-
ness with unfamiliar vendors and I have indeed avoided foreign firms - any contacts to
them.
Object D: Danger of hackers and viruses. The concept of hacker is used here because the
interviewees use it. They usually add that hackers use viruses to get their personal information

from e-vendors. While the interviewees trust in their own e-vendors’ (in believing that they
do not misuse consumers’ personal information), the trust to the e-vendors’ ability to protect
consumers’ information is rather low. The interviewees blame the e-vendors of not taking good
care of their information. The next quotation from among the interviews presents an example
(Conception 2D):
If someone gets my personal information and can take money from my bank account.
... The way anyone can get to one’s files and find whatever from them. Then someone
indeed might empty my bank account … my empty bank account.
In the interviews, the danger of hackers and viruses is mentioned often, but the interviewees do
not specify their views, not even when probed by the interviewer.
Object E: Risk with payment. This object refers to how issues related to making payment on
the Internet entail privacy problems: the interviewees are concerned that credit card information
can be stolen or used somewhere without their permission. Only a few of the interviewees report
having used their credit cards on the web. Although most of them frequently use Internet bank-
ing, they do not talk about it in a context of risk with payments; the bank is seen as an institu-
tion that can be trusted it seems. E-shopping could be paid in two main ways according to the
interviewees: with a credit card or in a post office when the consumer receives the product. The
interviewees said that now it is also possible to pay e-shopping directly via an Internet bank.
The interview quoted below shows that even though some of the consumers interviewed
may feel afraid using their credit cards, they have nevertheless used them because of the benefits
perceived (Conception 1E):
When you use some trustworthy and large vendors you get (the ordered product) in a
week or so… Then you will get the right product at the right time, and the bill comes at
the same time (with the product). When you pay with a Visa (credit card) it is chancy. I
wouldn’t otherwise use it (Visa credit card), but you have to use it (for paying) abroad.
The products are cheaper there.
4.2 Structural components of information privacy

Besides the above mentioned referential objects, the categorization includes a second dimension,
the structural components of information privacy. The components show different modes of
thought and stress the structural aspect of the conceptions.
Component 1: Product and e-vendor. When privacy is understood as related to the compo-
nent of the product and the e-vendor, all interviewees express very similar opinions. They say
that it is easier to order and give information if you know exactly what you are getting and who
the e-vendor is. First, the interviewees underline that they trust well-known e-vendors. They
mentioned many familiar Finnish brands – such as Veikkaus (a betting company, also working
online) and NetAnttila (an e-shop for clothes and household goods) – as places whereto it is safe

to give their personal information. The main principle seemed to be: you can only trust some
companies on the web, not all.
Besides the familiarity of the e-vendor, the familiarity of the product also matters. It is easi-
est to buy online if the product is familiar and standard, as the following interview quotation
illustrates (Conception 1E):
If I want to order a bottle of Scotch once a month then I could type it there… as a pay-
ment to any international bank account. But if you do not know anything about the
product…
Component 2: Technology. In the second component, privacy is understood through technolo-

gy. The interviewees refer to technology using words such as “the system”, “their computer”, “my
computer”, and “e-vendors’ register”. In addition, the e-vendor’s www-pages and their mainte-
nance are understood as technology. The interviewees usually claim that information security,
as far as privacy is concerned, refers to the security of a bank account or the confidentiality of
a credit card number; they also mention firewalls or virus protection. However, the interview-
ees also mention many positive sides of e-commerce systems: for example, a system can check
if there are typing errors; it is easier to compare different products; the use of an e-commerce
system may be cheaper and faster. The use of technology may also be worthwhile from the con-
sumer’s viewpoint (Conception 2A):
Computers are good, as they can check right away that you are doing the right thing.
With a person, mistakes may happen. A computer admits without scruples when some-
thing is not working. If there was a person, it would take time before getting a comment
that something is wrong.
Component 3: Societal norms. The interviewees describe that differences in national laws and
behavioural norms affect on their actions in e-commerce. The third component is called Societal
norms since it concentrates on others’ expected behaviour based on images of commerce habits
and technological expertise in certain countries. The interviewees have very strong views on
security in different countries and the opinions are similar among all interviewees. They note
that ordering from abroad is not safe; in other words, they would not give their personal infor-
mation abroad. The interviewees emphasize their perception that information security is better
in Finland, and that it is easier to give personal information to Finnish vendors than to others
(Conception 3D):
I have been interested in high quality artistic tools. And there (abroad) are plenty of
them (artistic tools) that are not available here (in Finland). Sometimes I have tried to
order them. However, there is still this problem of privacy... Some hacker may steal my
information and I don’t receive my packet at all.
Component 4: Consumer him/herself. The interviewees can also understand the implications
of their own actions regarding privacy in e-commerce. The interviewees describe problems and
how their own behaviour can provide solutions. The problems mentioned in this respect are

connected to the use of e-services, for example, to the difficulties of changing one’s own per-
sonal information or remembering user names and passwords. The interviewees’ policies vary.
One solution is to avoid giving any information to the e-vendor, thus refusing to deal with e-
commerce—the reason stated can be: “it is not safe”. Another solution is to avoid giving one’s
own personal information, using a fake personality instead, such as Donald Duck. The third
solution is to use the power of a community, which is easier via the Internet, as explained in the
following quotation (Conception 4E):
I’ve often noted that it (virtual community) is a very good deterrence. If after buying you
go somewhere (to a virtual community) and you ask whether this should be really so and
so… Usually when the firm finds out that somebody has been making inquiries about
their products, the service can improve suddenly.
Component 5: Fellow men. Besides of the interviewee’s own actions there are other people
whose actions affect the interviewee’s privacy situation. Focusing on the whole interviewing situ-
ation the interviewees also described what problems other people could cause to them. In the
category of Fellow men, the people are known by name and they belong to the same community
as the interviewee. They can be family members, work colleagues, or friends – also from a virtual
community. These known persons can help in the use of e-commerce, but they can also cause
harm, as in the next interview quoted, in which the interviewee discusses monitoring which
injures privacy (Conception 5B):
When I am using the computer, there can be so-called distractions. I find them disturb-
ing. (-) Sometimes my husband is trying to peek at what I am doing.
5 Discussion
Our study dealt with how consumers view information privacy. The result is a categorization,
which reveals consumers’ anxiety in the referential objects of privacy (use and misuse of cus-
tomer information, monitoring consumers, threat of spam, danger of hackers and viruses, risk
with payment). When we discussed privacy with the interviewees, they pointed our several risks
and threats. In the categorization, only one object, use and misuse of customer information,
contains both positive and negative sides of privacy; all other objects of privacy are regarded as
rather negative. Besides of the objects, we also identified the structural components of thought
about privacy. These were: product and e-vendor, technology, societal norms, consumer him/
herself, and fellow men.
As we described earlier, there are contradictory views about consumers’ fears of information
privacy. While several studies (Cassidy and Chae 2006; Chung 2003; Liu et al. 2005) state
that consumers are worried of their personal information use in e-commerce, Cheung and Lee
(2006) state that consumers are not afraid of invasions of privacy. Our study explains one reason
for the contradictory findings. Cheung and Lee studied the fears of privacy concerns in the con-

text of e-commerce with known e-vendors. The consumers were asked to evaluate the following
statements (Cheung and Lee 2006):
1. Internet vendors will sell my personal information to third parties without my permis-
sion,
2. Internet vendors’ are concerned about consumers’ privacy,
3. Internet vendors will not divulge consumers’ personal data to other parties.
Cheung and Lee (2006) found that consumers trust the e-vendor with who they make com-
merce and have no fear of the e-vendor misusing their personal information. Our study verifies
that only few consumers fear invasion of privacy caused directly by their own vendors. Neverthe-
less, this does not mean that consumers do not have any concerns. Instead, the interviewees were
afraid of anonymous surveillance, spam (other than that related to their vendors) and hackers
using viruses – which might be the context of other information privacy studies.
Already before our study, we knew that consumers’ privacy behaviour varies. Westin present-
ed three types of people in this respect: privacy fundamentalists, unconcerned and pragmatists
(Taylor 2003). Westin’s segmentation is a useful conceptual framework when thinking about
how an e-service might be used; however, it does not work when the focus is a consumers’ point
of view. The consumers interviewed about privacy do not demonstrate a single attitude; instead,
they act differently in different situations and contexts.
Our study indicates that Internet privacy is not a stable and homogenous concept to con-
sumers. While in one case the interviewee described herself as being careful about what infor-
mation she gave to an e-vendor, in another case she was not that concerned about information
privacy. She was not concerned when she acted with her own bank and e-supplier, since she had
used their e-services several times. She seemed to feel safe in familiar situations, but in a new
situation with an unknown vendor she is careful and does not provide whatever information.
To get a good idea of the consumer’s privacy views, the informant needs to be asked to describe
both familiar situations, such as e-banking and e-actions with a long-standing vendor, and new
situations in which the vendor is unknown as in casual Internet purchasing activities. In some
studies (e.g., Cheung and Lee 2006) this differentiation was not used.
Furthermore, the chosen method (phenomenography) focuses on the second-order perspec-
tive, which means that we described consumers’ views on privacy in e-commerce as perceived by
a certain group of people. This kind of approach helps to understand the variety of conceptions
of privacy. When the nature of the concerns is understood, it offers building blocks for further
research. For example, the researchers have to take situationality into account also in the case of
information privacy.
In phenomenography the world is described the way informants see it. The validity of the
study is based on that some informants (or at least one informant) see the phenomena via the
described conceptions. However, some of the conceptions can be totally unrealistic and against
the current scientific knowledge about the phenomenon. For example, one of the interviewees
quoted previously was wondering whether somebody, through the Internet, could know when
the ink was running short in her printer. The example illustrates how consumers’ unjustified
fears might affect their judgement. Thus, unrealistic statements forming part of (some) consum-

ers’ views on the phenomenon, need to be analysed as well in order to understand the variety of
consumers’ thought-models.
Our experiences indicate that the method of phenomenography is rich and rewarding, but
at the same time, that the method requires time and hard work. Since there are no formal data
gathering and analysis methods (for example as compared to grounded theory) and as the re-
search is conducted phase by phase, phenomenography demands good organizing skills from
the researcher.
Phenomenography is a qualitative method, and is therefore prone to the same potential
limitations as other qualitative studies. One of these limitations is the impossibility of deter-
mining when the number of informants is sufficiently large. About twenty informants is said
to be a large enough group for a theoretical saturation (e.g., Alexandersson 1994). We had 22
interviewees which seem to be enough, since the same objects and views were repeated, and the
last informants did not contribute new elements to the categorization.
For e-vendors, the results of this study show that it is crucially important to understand
online consumers’ concerns about privacy in order to maximize the potential of e-commerce.
Understanding how consumers view privacy issues provides a means to understand whether
people would be open to marketing efforts which require information sharing and information
exchange. In addition, the results help to understand what safeguards and other actions must
be in place to ensure that consumers are willing to give their information and use e-commerce
systems. The interviewees regard e-vendors as benevolent but not sufficiently competent in tak-
ing care of information privacy. For taking better care about privacy and making the care-taking
visible to their customers the e-vendors should:
• describe how consumers’ personal information is taken care of and used,
• give consumers more advice on giving and handling personal information safely,
• state if some third party is allowed to use their personal information for marketing
purposes,
• take good care of their own brands, reputation and usability,
• enable their customers to check, correct and delete their own personal information,
and
• be worthy of trust.
6 Conclusion
The aim of this study is to describe the differences in how consumers view privacy in e-com-
merce. In privacy issues we need more understanding from a consumer viewpoint (Cassidy and
Chae 2006; Dinev et al. 2006; Malhotra et al. 2004). The result of our study is a categorization
of consumers’ views on information privacy. It includes different layers of understanding by
focusing on referential objects and structural components of information privacy.

Besides of the categorization itself, an important finding is that consumers’ views on privacy
(and privacy concerns) are situated; it depends on the context. A familiar situation—e.g., acting
with a known e-vendor—is regarded as safe, whereas a new, unknown situation is seen as fearful
and risky. New experiences (e.g., using a new web site several times) and new information (e.g.,
from media) affect the consumer’s behaviour. The view of privacy should thus not be regarded
as stable, but as constantly under social construction.
7 Acknowledgment
The authors would like to gratefully acknowledge the helpful comments on earlier versions of
this paper from the Associate Editor, Professor Bjørn Erik Munkvold, the three anonymous
reviewers, Dr Hannakaisa Isomäki, Professor Pertti Järvinen, Dr Minna-Kristiina Paakki, Dr
Tero Vartiainen, and the members of Dr Viveca Asproth’s working group in IRIS28 - 2005,
Kristiansand, Norway. We thank Steve Legrand for making our English more readable.
8 References
Alexandersson, M., Metod och medvetande, (Method and Awareness, in Swedish), Acta Universita-
tis Gothoburgensis, Göteborg, 1994.
Alvesson M., and Karreman, D., “Varieties of discourse: On the study of organizations through
discourse analysis”, Human Relation, (53:9), 2000, pp. 1125-1149.
Awad, N. F. and Krishnan, M. S., “The Personalization Privacy Paradox: An Empirical Evalua-
tion of Information Transparency and the Willingness to be Profiled Online for Personaliza-
tion”, MIS Quarterly, (30:1), 2006, pp. 13-28.
Bellman, S., Johnson, E. J., Kobrin, S. J., and Lohse, G. L., “International differences in infor-
mation privacy concerns: a global survey of consumers”, The Information Society, (20), 2004,
pp. 313-324.
Bergkamp, L., “EU data protection policy. The privacy fallacy: adverse effects of Europe’s data
protection policy in an information-driven economy”, Computer Law & Security Report,
(18:1), 2002, pp. 31-47.
Booth, S. A., Learning to program: A phenomenographic perspective, Acta Universitatis Gothobur-
gensis, Göteborg, 1992.
Bygrave, L. A. “Data Protection Pursuant to the Right to Privacy in Human Rights Treaties”,
International Journal of Law and Information Technology, (6), 1998, pp. 247-284
Camp, L. J., and Osorio, C., Privacy Enhancing Technologies for Internet Commerce, Trust in the
Network Economy, Springer-Verlag, Berlin, Germany, 2003.
Cassidy, C. M., and Chae, B., “Consumer Information Use and Misuse in Electronic Business:
An Alternative to Privacy Regulation”, Information Systems Management, Summer 2006,
pp. 75-87.

Cheung, C. M. K., and Lee, M. K. O., “Understanding Consumer Trust in Internet Shopping:
A Multidisciplinary Approach”, Journal of the American Society for Information Science and
Technology, (57:4), 2006, pp. 479-492.
Choughury, V. and Karahanna, E., “The Relative Advantage of Electronic Channels: A Multidi-
mensional View”, MIS Quarterly (32:1), 2008, pp. 179-200.
Chung, W., “A Snoop at Privacy Issues on the Internet in New Zealand”, Business Review, Uni-
versity of Auckland, (4:3), 2003, pp. 2-15.
Curran, C. M., and Richards, J. I., “Misplaced Marketing. Public Privacy and Politics”, Journal
of Consumer Marketing, (21:1), 2004, pp. 7-9.
Dinev, T., Bellotto, M., Hart, P., Russo, V., Serra, I., and Colautti, C., “Privacy calculus model
in e-commerce – a study of Italy and the United States”, European Journal of Information
Systems, (15), 2006, pp. 389-402.
Directive (95/46/EC) of the European Parliament and of the Council of 24 October 1995 on
the protection of individuals with regard to the processing of personal data and on the free
movement of such data.
Eriksen, T. H., Small Places, Big Issues. An Introduction to Social and Cultural Anthropology, Pluto
Press, London, UK, 2001.
Gefen, D., Karahanna, E., and Straub, D. W., “Trust and TAM in Online Shopping: an Inte-
grated Model”, Management Information Systems Quarterly, (27:1), 2003, pp. 51-90.
Glaser, B. G., and Strauss, A., The Discovery of Grounded Theory, Strategies for Qualitative Re-
search, Aldine publishing company, Chicago, 1973.
Grable, J. E. and Joo, S., “Factors related to risk tolerance: a further examination”, Consumer
Interests Annual, (45), 1999, pp. 53-58.
Graeff, T. R., and Harmon, S., “Collecting and using personal data: consumers’ awareness and
concerns”, Journal of Consumer Marketing, (19:4), 2002, pp. 302-318.
Hui, K.-L., Teo, H. H., and Lee, S.-Y. T., “The Value of Privacy Assurance: An Exploratory Field
Experiment, MIS Quarterly (31:1), 2007, pp. 19-33.
Hynes, D., Tiainen, T., Koivunen, E.-R., and Paakki, M.-K., “Articulating ICT Use Narratives
in Everyday Life”, In EM Trauth (Ed.), Encyclopedia of Gender and Information Technology.
Idea Group Reference, London, UK. 2006, pp. 37-43.
Isomäki, H., The Prevailing Conceptions of the Human Being in Information Systems Development:
Systems Designers’ Reflection, Department of computer and information sciences, University
of Tampere, Tampere, 2002.
Jensen, C., Potts, C., and Jensen, C., “Privacy practices of Internet users: Self-reports versus
observed behavior”, International Journal of Human-Computer Studies (63), 2005, pp. 203-
227.
Kahle, L. R., Beatty, S. E., and Homer, P, “Alternative Measurement Approaches to Consumer
Values: The List of Values (LOV) and Values and Life Style (VALS)”, Journal of Consumer
Research, (13), 1986, pp. 405-409.
Kruck, S. E., Gottovi, D., Moghadami, F., Broom, R., and Forcht, K. A. “Protecting personal
privacy on the Internet”, Information Management & Computer Security, (10:2), 2002, pp.
77-84.

Liu, C., Marchewka, J. T., Lu, J., and Yu, C-S., “Beyond concern—a privacy-trust-behavioral
intention model of electronic commerce”, Information and Management, (42), 2005, pp.
289-304.
Malhotra, N. K., Kim, S. S., and Agarwal, J., “Internet Users’ Information Privacy Concerns
(IUIPC): The Construct, the Scale, and a Causal Model”, Information Systems Research,
(15:4), 2004, pp. 336-355.
Marton, F., “Phenomenography – describing conceptions of the world around us”, Instructional
science, (10), 1981, pp. 177-200.
Marton, F., Towards phenomenography of learning, Integratial experiments aspects, University of
Gothenburg Dept. Education, Gothenburg, 1982.
Marton, F., and Booth, S., Learning and awareness. Mahwah, N.J.: Lawrence Erlbaum, 1997.
McRobb, S., and Rogerson, S., “Are they really listening? An investigation into published online
privacy policies at the beginning of the third millennium”, Information Technology & People,
(17:4), 2004, pp. 442-461.
Milberg, S. J., Burke, S. J., Smith, H. J., and Kallman, E. A., “Values, personal information
privacy concerns, and regulatory approaches”, Comm. ACM, (38:12), 1995, pp. 65-74.
Muttilainen, V., “Suomalaiset ja henkilötietojen suoja. Kyselytutkimusten ja viranomaistilas-
tojen tietoja 1990-luvulta ja 2000-luvun alusta”, (Finnish people and security of personal
information. Surveys and statistics of state authority 1990’s and in the beginning of 2000,
in Finnish), Oikeuspoliittisen tutkimuslaitoksen julkaisuja, (218), Oikeuspoliittinen tut-
kimuslaitos, Helsinki, 2006.
Paine, C., Reips, U.-D., Stieger, S., Joinson, A., and Buchanan, T., “Internet users’ perceptions
of ‘privacy concerns’ and ‘privacy actions’”, International Journal of Human-Computer Stud-
ies, (65), 2007, pp. 526-536.
Pan, Y. and Zinkhan, G. M., “Exploring the impact of online privacy disclosures on consumer
trust”, Journal of Retailing Volume (82:4), 2006, pp. 331-338.
Phelps, J., Nowak, G., and Ferrell, E. “Privacy Concerns and Consumer Willingness to provide
Personal Information”, Journal of Public Policy Marketing, (19:1), 2000, pp. 27-41.
Prabhaker, P. R., “Who owns the online consumer?”, Journal of Consumer Marketing, (17:2),
2000, pp. 158-171.
Rubin, P. H., and Lenard, T. M., Privacy and the commercial use of personal information, Boston:
The Progress & Freedom Foundation / Kluwer Academic Publishers, 2002.
Smith, H. J., “Information Privacy and Its Management”, MIS Quarterly Executive, (3:4), 2004,
pp. 291-313.
Smith, H. J., Milberg, S. J., and Burke, S. J., “Information privacy: Measuring individuals’ con-
cerns about organizational practices”, MIS Quarterly, (20:2), 1996, pp. 167-196.
Stewart, K. A., and Segars, A. H., “An empirical examination of the concern for information
privacy instrument”, Information Systems Research, (13:1), 2002, pp. 36-49.
Säljö, R., “Minding action. Conceiving the world versus participating in cultural practices”,
Nordisk Pedagogik, (14), 1994, pp. 71-80.
Taylor, H., “Most people are “privacy pragmatists” who, while concerned about privacy, will
sometimes trade it off for other benefits”, The Harris Poll, (17), 2003, March 19.
Udo, G. J., “Privacy and security concerns as major barriers for e-commerce: a survey study”,
Information Management & Computer Security, (9:4), 2001, pp. 165-174.

Vartiainen, T., “Moral Conflicts in Teaching Project Work: A Job Burdened by Role
Strains”, Communications of the Association for Information Systems, (20, article 43), 2007.
Verhagen, T., Meents, S., and Tan, Y.-H., “Perceived risk and trust associated with purchas-
ing at electronic marketplaces”, European Journal of Information Systems (15), 2006, pp.
542–555.
Westin, A. F., Privacy and Freedom, Atheneum, New York, 1967.

A Special Issue Editorial
Distributed PD: Challenges and opportunities
Karin Danielsson Öberg

Umeå University, Sweden
kdson@informatik.umu.se
Dorina Gumm
effective WEBWORK GmbH, Germany
gumm@effective-webwork.de
Amir M Naghsh
Sheffield Hallam University, UK
a.naghsh@shu.ac.uk
Participatory design (PD) has been one of the important contributions of Scandinavian re-
searchers to information systems (Floyd et al. 1989). For a considerable amount of time, the
importance and usefulness of the approach has been illustrated during the design of computer
systems for work. The Scandinavian Journal of Information Systems (SJIS) serves as an impor-
tant contribution to the history and development of the approach. Already in its first volume
published in 1989, PD research was represented. The PD approach focuses on the collaboration
between designers and users, supporting direct involvement of users in the design process itself
while allowing designers to participate in the users’ world (cf., Kyng 1998).
Globalisation and organisational networks—and the implicit distribution coming along
with this—increasingly determine software development (Audy et al. 2004). The distribution of
collaboration in design and development processes has become a key challenge for PD, espe-
cially when one considers participation that was originally developed with a focus on co-located
design activities. This view of activity was developed to allow users and designers to collaborate
face-to-face, building on their mutual learning through the design process. However, the in-
volvement of a large number of users and other stakeholders who may be distributed across time,
space and organisational structures, has challenged developed PD research and activities.
Distributed design teams are by no means a new aspect in participatory design. A variety
of research work reports on projects that face some kind of distributedness (e.g., Divitini et al.
2000; Irestig et al. 2002, Bødker et al. 2004). Moreover, support of such distributed projects,

in terms of an effective facilitation of communication and knowledge sharing in distributed set-
tings, has become one of the main considerations in the field of computer supported cooperative
work (CSCW). Although PD does not necessarily argue for, or include, computers in order to
support group work during design (Bannon 1993), the two communities could benefit from
each other (Kensing and Blomberg 1998).
In addition other research domains like open source software development (OSSD) (Bar-
cellini et al. 2008) or distributed software development (DSD) (Heeks et al. 2001; Damian et
al. 2003; Coar 2004, Gumm 2006) explore the complexities and challenges related to distrib-
uted people and groups. However, although this research has developed a considerable body
of knowledge, the understanding about distinct challenges of and solutions for participatory
design in distributed environments is still evolving.
This SJIS special issue on distributed PD, has its origins in a series of events including: two
workshops conducted at NordiCHI 2006 and CHI 2008 and a panel discussion held at PDC
2008. Our intention for the workshops and the panel was to enable researchers with this joint
interest to meet and share experiences and knowledge. Several interesting articles from research
and practice were presented at these workshops, enabling creative and encouraging discussions
among participants.
The call for papers for this special issue attracted 13 submissions and we are pleased to note
that a variety of European countries were represented by the submitting authors such as Greece,
Italy, United Kingdom, Germany, Belgium, and the Scandinavian countries Sweden, Norway
and Finland. This illustrates the widespread and ongoing interest regarding Distributed PD in
Europe, but also a special attention among the Scandinavian countries. Through three rounds
of double-blind reviews, three papers were selected for publication in this special issue. They in-
clude experiences gathered from extensive distributed PD project research as well as theoretical
reflections about PD’s relation to other research areas.
Longitudinal case studies provide the possibility to reflect on changing project settings and
the evolution of distribution within a project. The first two papers of this special issue report on
very different case projects in which a software system, initially developed for a particular user
group, has been adapted to a growing range of use contexts. Such expansion requires coping
with the diversity of roles and cultures. In the first article entitled “Distributed development to
enable user participation: Multilevel design in the HISP network”, Ola Hodne Titlestad, Knut
Staring and Jørn Braa explore distributed participation within and across countries through a
long term, globally targeted effort of networked action research projects. They introduce and
discuss the concepts of scaffolding and boundary spanners as approaches to cope with the geo-
graphical and organisational distribution and diversity of roles.
The development and expansion of the domain application also requires new approaches to
cope with requirements emerging from different contexts of usage. In the second article entitled
“Intercontextual distributed participatory design: Communicating design philosophy and en-
riching user experience”, Hartmut Obendorf, Monique Janneck and Matthias Finck report on
a project in which a software application has been developed for a growing amount of use con-
texts, ranging from universities and schools across freelancer networks and other kinds of firms
and organisations. The authors introduce two new methods, intercontextual user workshops
and commented case studies that support the participation of users and mutual understanding
across different communities of practice.
24 • A Special Issue Editorial

Finally, the challenges accompanied by distribution are not limited to Participatory Design
but arise whenever collaborative methods are applied. In the third and final article of the special
issue, entitled “Furthering distributed participative design: Unlocking the walled gardens”, Clau-
dia Loebbecke and Philip Powell analyse the similarities between participatory design, design
science and action research. By thoroughly analysing 15 papers from these research approaches,
they demonstrate how they match in research contributions, roots and methodological guide-
lines. The authors argue that insight into some of the issues raised by participatory design in
distributed contexts may arise if PD looks outside of its ”walled garden”.
Given distributed participatory design’s growing importance, the reflection of existing par-
ticipatory methods, experiences with new methods and the connection to other important re-
search areas are required. Although significant progress has taken place in the area of distributed
PD, much still needs to be done.
With this special issue we hope the knowledge and expertise among researchers and practi-
tioners will reach further than before; hopefully, this is the departure and not the end. It is with
pride that we hand this special issue over to you and hope that you will enjoy the reading it as
much as we have preparing this valuable set of papers.
At last but not the least, we would like to acknowledge the important work of our reviewers
for this special issue: Andrea Herrman, Andrew Warr, Andy Dearden, Anna Croon Fors, Bir-
gitta Bergvall-Kåreborn, Brendon Clark, Charlotte Wiberg, Chris Roast, Christina Mörtberg,
Daniela Damian, Erik Stolterman, Jonny Holmström, José Abdelnour-Nocera, Marcel Christ,
Mikael Wiberg, Monique Janneck, Peter Wright, Sari Kujala, Steve Sawyer, Tony Gorschek, and
Victor Kaptelinin. Thank you for your hard and quality work.
We would also like to take this opportunity to thank Andy Dearden and Andy Warr who
have been co-workers at the workshops, and the invited panellists at PDC 2008: Jeanette Blomb-
erg, Tone Bratteteig, Jose Abdelnour-Nocera, and Gerhard Fischer.
We would like to send our greatest gratitude to the participants of the two workshops; with-
out you there would not have been this special issue.
Finally, we would like to thank editor Matti Rossi for his support and work with this special
issue.
1 References
Audy, J., Evaristo, R. & Watson-Manheim, M. B., “Distributed analysis: The last frontier?”, in
HICSS ’04, IEEE Computer Society, 2004, p. 10010.2.
Barcellini, F., Détienne, F., Burkhardt, J.M. “Users and developers mediation in an Open Source
Software Community: boundary spanning through cross participation in online discus-
sions,” Int. J. Human-Computer Studies, 66, 2008 pp. 558–570.
Bannon, L. CSCW: An initial exploration, in Scandinavian Journal of Information Systems, 1993,
Vol. 5, pp. 3-24.
Bødker, K., Pors, J. K. & Simonsen, J., ‘Implementation of web-based information systems in
distributed organizations’, in Scandinavian Journal of Information Systems, 2004, Vol 16,
85–116.
A Special Issue Editorial • 25

Coar, K., “The sun never sets on distributed development”, Distributed Development 1(9), 2004,
pp. 32–39.
Damian, D. E., Eberlein, A., Shaw, M. L. & Gaines, B. R., ‘An exploratory study of facilitation
in distributed requirements engineering’, Requirements Engineering 8(1), 2003, 23–41.
Divitini, M., Farshchian, B. A. & Tuikka, T., “Internet-based groupware for user participation
in product development”, SIGCHI Bulletin 32(1), 2000, pp. 31–35.
Floyd, C., Mehl, W.M., Reisin, F.M., Schmidt, G. & Wolf, G. , “Out of Scandinavia: Alterna-
tive approaches to software design and system development”, Human-Computer Interaction
4(4), 1989, pp. 253–350.
Gumm, D.-C. (2006b), ‘Distribution dimensions in software development projects: A taxono-
my’, IEEE Software. Special Issue on GSD, 2006, 23(5), 45–51.
Heeks, R., Krishna, S., Nicholson, B. & Sahay, S., “Synching or sinking: Global software out-
sourcing relationships”, IEEE Software - Global Software Development 18(2), 2001, 54–60.
Irestig, M. & Timpka, T., “Dynamic interactive participatory design: a method for extending
participatory design to large system development projects”, in Proceedings Participatory De-
sign Conference, CPSR, 2002, pp. 317–322.
Kyng, M., Users and computers: A contextual approach to design of computer artefacts, in
Scandinavian Journal of Information Systems, 1998, Vol. 10, No. 1&2, pp. 7-44.
26 • A Special Issue Editorial

Distributed Development to Enable User
Participation
Multilevel design in the HISP network
Ola Hodne Titlestad

Department of Informatics, University of Oslo, Norway
olati@ifi.uio.no
Knut Staring
knutst@ifi.uio.no
Jørn Braa
jbraa@ifi.uio.no
Abstract. Through the study of a long term, globally targeted effort to design health in-
formation systems in the Global South, we explore challenges to distributed participation
within and across countries, and describe efforts at addressing these. Networked action
research projects can enable pooling of resources, skills, best practices and tools, and cross-
country collaboration does not have to preclude local ownership, as illustrated by the case
material in this article. We highlight specifically the need for circulation of people, artefacts,
and standards, to both support local practices and foster the capacity of all stakeholders to
take active part in the design and implementation of information systems. The deep effects
of global technological change call for a multilevel approach bridging local implementations
with global research and participatory design efforts and co-evolution of standardised tools.
Key words: Distributed participatory design, boundary spanning, scaffolding, innofusion,

Global South

1 Introduction
This article addresses the issue of participatory design (PD) of information systems in resource
-constrained settings, with a particular focus on the diverse public health care sector in the
Global South. The insights from the PD tradition are widely recognised, and have become an
increasingly mainstream part of information system development methodology in rich coun-
tries. However, their potential has yet to be fully exploited in regions such as Africa and Asia,
where design-reality gaps remain a challenge to computerisation attempts (Avgerou 2002; Heeks
2006). Examining the case of a long term effort to design health information systems across a
number of countries, we discuss the scope for PD approaches in networks that are distributed
along several dimensions.
In the South, the lack of funds and skilled people underscores the importance of being able
to draw on inputs from similar settings, as participation in design is costly. Key challenges thus
relate to developing means to leverage achievements and replicating participatory learning proc-
esses in new places. Therefore, Distributed Participatory Design (DPD) becomes an important
goal – both for empowering developer and user communities in various countries, and for build-
ing reusable tools. The spread of the internet is a major factor in enabling PD across space and
time (Gumm 2006a). The aim of this article is to explore what DPD can entail in a network of
South-South-North collaboration.
There is a strong need for consolidation, integration and collaboration in health information
systems globally (AbouZahr and Boerma 2005; Tierney et al. 2008). The HIV/AIDS pandemic
and initiatives such as the UN Millennium Development Goals (MDG) have lead to priori-
tisation of the complicated task of improving health services and information flow, and have
increased donor support for many health programs (such as TB, malaria, vaccination programs,
maternal health care, etc.). Unfortunately, donors often introduce bespoke systems that sidestep
established reporting routines. The resulting complexity, coupled with meagre human resources
and infrastructure, easily leads to fragmentation and poor data quality. The launch of the Health
Metrics Network (HMN) in partnership with the World Health Organisation (WHO) in 2005
was an important step towards promoting standards. A key feature of the HMN Technical
Framework is integration of essential data in a shared repository.
Such emphasis on integration and standardisation is quite removed from the original focus
of participatory design projects. The PD tradition grew from a number of Scandinavian work-
ing life action research projects in the 1970’s and early 80’s. In particular, union based projects
focused on empowerment of workers to:
• Improve their capacity to negotiate the introduction of new technology with manage-
ment (for example, a project for the Norwegian Iron and Metal workers’ union; Nyg-
aard 1977; Sandberg 1979), and
• Foster participatory design of alternative computer applications with the workers’ skills
and interests in focus (for example, the UTOPIA project, which aimed to create good
tools for typographers; Bødker et al 1987).
While the direct outcomes from these projects were relatively insignificant, they did have an
effect at the political level, and gave rise to new legislation regulating the introduction of new
28 • Titlestad, Staring & Braa

technology (Bjerknes and Bratteteig 1995). Even so, the early PD projects had little impact out-
side singular research sites, and therefore also failed to become sustainable. Summarizing these
experiments, Engelstad and Gustavsen (1993) suggested that the action research community
needed to move beyond isolated efforts; focus should be shifted “from single organisations and
work places [...] to networks” (ibid., p. 219). Adding to this the globalisation afforded since
then by the Internet makes it interesting to examine approaches to user involvement in design
in distributed networks (Obendorf et al this issue).
The Health Information Systems Programme (HISP) is a multinational research and devel-
opment network aiming at developing information systems to strengthen local health services
in the Global South. Explicitly rooted in the PD tradition (e.g., receiving the Artful Integrator’s
Award at the Participatory Design Conference in 2006 for action research and design across
country and community contexts), the project has increasingly imbibed lessons from the dis-
tributed software development (DSD) literature. HISP was initiated in South Africa in 1994 by
researchers from local and Norwegian universities, and health activists from the anti-apartheid
struggle. The project was part of the new ANC government’s Reconstruction and Development
Program (ANC 1994a), and began as a participatory system design project in three pilot dis-
tricts, aimed at supporting newly defined health districts (ANC 1994b). The project had two
major elements: 1) definition of an integrated essential data set, including data from the full
range of health programs and services, to be collected from all health facilities, and 2) develop-
ment of a software application called the District Health Information Software (DHIS) to man-
age the data being collected and integrate with other data sources such as population census. In
the period 1999 - 2001, use of the DHIS application gradually spread, leading it to become an
official national system in South Africa (Braa and Hedberg 2002).
After the turn of the millennium, HISP activities spread to other countries in Africa and
Asia. Today, HISP comprises a global network of universities, health authorities and NGOs
from around 15 countries in the Global South. Through persistent efforts, the network involved
in design and development of the software also slowly expanded from a two-man team in South
Africa to a global team distributed across 7-8 countries in Africa, Asia, and Scandinavia. Distri-
bution and coordination of activities within HISP have been enabled by an international PhD
program with more than 25 students from Africa and Asia and the establishment of Masters pro-
grams in health informatics in several countries (Tanzania, Mozambique, Malawi, Ethiopia, Sri
Lanka) funded by the Norwegian government, where students conduct research within the net-
work and circulate between countries, enabling the sharing of experiences and best practices.
Organisation of the article: The next section reviews the literature on participatory design
in distributed contexts and related concepts, followed by a section outlining the methodology.
Section four details the increasing distribution of design and development activities in the HISP
project. This is followed by a discussion where we draw on the case study to describe a global
approach to participation across boundaries. The last section sums up our contributions.
Distributed Development to Enable User Participation • 29

2 Literature and background
This section discusses the concept of participatory design in relation to recent trends in the direc-
tion of widely dispersed software development and deployment of standard information systems
in diverse settings. To analyze the role of participatory design in a globalized world, we introduce
concepts around learning and innovation across boundaries with the help of standards.
2.1 DPD, DSD and standards

There has been a marked growth in standards based, generic information systems, designed for
use throughout whole industry sectors, such as health care, or even across sectors, such as ERP
systems (Pollock et al. 2007). Thus, there is a need for approaches to distributed systems de-
velopment that both ensure coherence and can support diverse local settings, striking a balance
between global standards and local needs (Rolland and Monteiro 2002). To be able to develop
global software systems, modularisation and generification strategies become paramount, and
this is reflected in both organisational arrangements and software architecture (Staring and Ti-
tlestad 2008).
The diversity of distributed use settings should be distinguished analytically from the dis-
tribution of software developers. The latter is explored in the literature on distributed software
development, which discusses challenges in terms of coordination and sharing of knowledge
beyond those faced by collocated teams (Lings et al. 2006). This is particularly relevant when
developers are spread across cultures and time zones, as is common in offshore outsourcing
(Sahay et al. 2003). Offshoring arrangements are usually seen mostly as cost-saving measures,
but as pointed out by the Open Innovation literature (Chesbrough 2003; West and Gallagher
2006), the preponderance of ingenuity in the world is distributed outside of any particular or-
ganisation, and one should seek to harness it. This would seem to be all the more the case when
potential users are also widely dispersed.
1 Have Clear Distribution Rationale 6 Manage Processes

2 Clarify All Understandings 7 Develop a Sense of “Teamness”
3 Leverage Modularity 8 Encourage Temporary Collocation
4 Use Cultural Mediation 9 Encompass Heterogeneity
5 Facilitate Human Communication 10 Develop an Effective Tool Base
Table 1: Ten strategies for DSD, from Lings et al. (2006)
The distribution of developers geographically introduces problems of mutual understanding

which are very similar to issues that PD approaches seek to address, and Gumm (2006a) rightly
asks whether the concept of DPD embodies a contradiction in terms. Furthermore, whereas
DSD has so far been studied mostly in commercial settings, PD is often associated with research
or public sector efforts. In our view, this makes it even more pertinent to compare the two bod-
ies of research. For example, it is interesting to note that the strategies suggested for successful
DSD (see Table 1) are quite similar to PD methods for bridging the gap between users and

developers, for example items 2, 5, 8 and 9. In other words, DSD research provides methods to
alleviate many of the challenges increased distribution poses, which can help make DPD viable.
Similarly, Loebbecke and Powel (2009) analyse links to action research and design science.
2.2 PD in the Global South and learning

Over time, the focus of much PD research has shifted in a more pragmatic direction (Green-
baum and Kyng 1991). A key PD principle is to bridge and blur the user-designer distinction
from both directions, through mutual learning processes. By this approach, users must be able to
engage with the artefact under design, and designers should build a thorough understanding of
the life-world(s) of the users, to have a more complete perspective of the system in use (Bjerknes
and Bratteteig 1995). Effective methods to achieve this usually rely on prototyping and intensive
face-to-face interaction between users and designers, such as close observation of the work prac-
tices of users, joint workshops, and scenarios (Kensing 2003). In the Global South, computer-
ised information systems are still few and far between, and potential users in the health care sec-
tor have no or extremely limited prior exposure to them. This amounts to a significant threshold
hindering participation, and a visual, interactive prototype is essential to overcome such barriers
and enable meaningful reciprocal learning between designers and users (Kimaro and Titlestad
2005). Such prototypes become significant boundary objects shared between problem solving
contexts (Star 1989). In new product development, boundary objects help establish a “boundary
infrastructure” used to manage knowledge across a given boundary (Carlile 2002).
Mutual design processes are therefore likely to initially have characteristics of exploration
and struggle to make things work, captured in the related concepts of innofusion (Fleck 1988)
and learning by trying (Fleck 1994). Complex artefacts such as organisational technologies are
likely to require processes of experimentation in the contexts in which they are implemented.
They may need to be taken apart, broken down, adapted and reconfigured, and sometimes the
technology may be more or less completely re-invented in its implementation and use. Repeated
reconfiguration of components into systems that fit particular contexts thus blurs the line be-
tween innovation and diffusion, leading to the notion of innofusion (Fleck 1988).
Similarly, learning by trying refers to “the knowledge created during innofusion and may
occasionally amount to significant changes in technological knowledge bases” (Peine 2008, p.
5). Learning occurs “due to an initial misfit between product characteristics and its use environ-
ment that leaves room for improvements through the operating experience” (ibid). According to
Peine, if the initial pioneering phase of radical change results in a working product, the process
becomes one of learning by using, with smaller, progressive modifications to the functioning
system. Once the product stabilises and there is less scope for further improvement, learning
takes on more of a character of on-the-job training, referred to as learning by doing. However,
in the Global South, with sparse technological infrastructure and knowledge base, this clean
sequence becomes blurred, with the categories of learning occurring in parallel: familiarisation
with technology and design of an appropriate system of necessity become intertwined and take
place simultaneously.
Furthermore, when considering the Scandinavian PD lessons in the context of the Global
South, an important difference is that the typical arena for movements for social change in de-

veloping countries is the political ‘grassroots’, rather than the workplace (Braa 1996). Therefore,
a community based approach is recommended (Byrne and Sahay 2003). As the user-designer
gap in knowledge of and experience with IT solutions is often wider in the Global South than
in Scandinavian settings, PD approaches aiming at fostering mutual learning are actually even
more appropriate and relevant in such contexts (Braa et al. 1993).
The concept of innofusion is particularly apt when we make the transition from bespoke
systems for a single site to more distributed information systems. In order to cope with the
variety of present and future contexts, Fischer (2008) introduces the concept of meta-design,
meaning design that aims to empower users to engage actively in continuous, iterative develop-
ment, rather than being restricted to the use of existing systems. Meta-design strategies such as
flexible standards, modularisation and black-boxing can facilitate generification processes (Braa
et al. 2007). The meta-design approach strives at creating not only a flexible technical basis for
design, but also social infrastructures in which users can participate actively as co-designers to
shape and reshape socio-technical systems.
2.3 Boundary spanning and scaffolding

Successful examples of global scale DSD can be found when studying the phenomenon of Free
and Open Source Software (FOSS), which is usually characterised by a communitarian, asyn-
chronous and mediated design practice where both users and developers are widely dispersed
(Barcellini et al. 2008). FOSS projects typically blur the separation of roles through transparent
processes and open communication and intensive use of online media (Fogel 2005; Lanzara and
Morner 2005). Indeed, the FOSS approach can be considered as a continuous form of open
ended distributed participatory design, where new functionalities can always be proposed by
different kinds of participants, regardless of their stake in the project, and users can potentially
be involved in all phases of the development process (Barcellini et al 2008; von Hippel 2005).
In order to create successful organisational systems, design teams should include both software
architects and domain experts. Thus, mutual learning processes are crucial, though difficult to
accomplish (Gumm 2006a). A successful FOSS project will consist of people filling different
roles (sometimes simultaneously), such as project leader, administrator, developer, and user, with
varying levels of engagement. In particular, the FOSS literature stresses the role of active users
who participate by assisting newcomers, reporting bugs, or proposing new features (Fogel 2005;
Preece et al. 2004).
Prominent ways to cope with the geographical and organisational distribution and diversity
of roles in FOSS projects include boundary spanners and scaffolding. Boundary spanners are
persons who act as mediators, traversing boundaries between organisations and teams, enhanc-
ing informal communication across networks (Sarant 2004; Sonnenwald 1996). “Becoming
boundary spanners implies having developed skills and competencies in the different fields that
are spanned. Boundary spanners are well aware of all practices and have achieved legitimacy and
credibility in the domains they span” (Barcellini et al. 2008:560). The multiple dimensions of
distribution in FOSS projects (and in DPD more generally) mean that many boundaries need
to be bridged, and this implies a strong need for developers, implementers and coordinators who
can span organisational contexts and geographical boundaries. Whereas with infrastructural

software like databases or operating systems, developers are usually also users, the differences in
perspective is larger for organisational software, and thus the need for boundary spanners all the
greater. The gap can best be bridged by technically conversant people who engage in implement-
ing the system and training end-users, and who are also adept at communicating with the core
developer team. More than just super-users, such mediators normally work closely with users to
make systems work and therefore understand their problems, but also know at least some of the
developers by name and can relate to them (Finck et al 2004).
Supplementing and enabling the activities of boundary spanners are material means of me-
diation and support for the design process. Seeing parallels between design and learning, we
find Orlikowski’s (2006) use of the concept of scaffolding useful for understanding how DPD
can succeed. A scaffold is not part of the final product, but serves temporary support functions.
Scaffolds are flexible and can be set up according to local conditions, and are also portable, i.e.
they can quickly be assembled and modified in different places. Thus the metaphor helps us
focus on the “temporary and situated engagement of technology in knowledgeable activity. That
is, for the duration of a particular human practice, actors draw on various artefacts, spaces, and
infrastructures to conduct their activities” (ibid., p. 462), or what Orlikowski terms the ‘scaf-
folding of knowledgeability’. Moreover, scaffolds are generative, in the sense that they serve as a
“basis for other (creative) work, facilitating the performance of activities that would have been
impractical without material augmentation” (ibid., p. 462).
Thus, the challenges of DPD can be addressed through the combination of a network of
collaborators formed around boundary spanners and scaffolding infrastructure such as training
material or the typical technical artefacts employed by FOSS communities (see section 3). Ac-
cording to O’Reilly (2003), the truly distinguishing characteristic that FOSS offers is not the
availability of the source code, but an overall “architecture of participation”, both material and
social. The experiences from the case presented below suggest the contours of such an architec-
ture for the Global South.
3 Method
Our empirical investigations are grounded in the Scandinavian action research tradition, which
emphasises engagement in the field. A long-term distributed action research effort, the approach
followed in HISP builds on the literature on networked action research (Elden and Chisholm
1993; Engelstad and Gustavsen 1993), and has elsewhere been termed Networks of Action (Braa
et al. 2004a). In essence, it is based on the premise that action should not take place in isolated
units, but only as part of a greater community where sharing of experiences, learning processes,
and support are facilitated. The Networks of Action approach is specifically designed for the
resource limited conditions in the Global South. The 15 years time-span of HISP exceeds tra-
ditional ‘projects’ and is more akin to social movements or programmes (Elden and Chisholm
1993).
All three authors are actively engaged in the Health Information Systems Programme on a
daily basis. One of the authors was the originator of the project in the mid-1990’s, and the two
other authors have been engaged in the action research project for about five years, working on

software development, design and implementation, as well as education activities. Since 2004,
this group has lead efforts to distribute participatory processes of development and design of
version 2 of the software (DHIS v2). More specifically, we have engaged in the management of
developers in the Norwegian node, facilitating collaboration between nodes, guiding PD proc-
esses in local implementation projects, and building design capacity in various countries.
The ongoing activities started more than a decade ago. Such a long-term action research
project inevitably gives rise to various data sources, both formal and informal, involving multi-
ple types of data and methods of collection. It is therefore difficult to provide extensive quantita-
tive details of the number and timing of interviews and repeat interviews conducted. However,
the project has produced vast amounts data from primary and secondary sources. In line with
FOSS best practices (Fogel, 2005), extensive use of open communication tools such as mailing
lists, issue trackers, and wikis means that much of the informal communication in the project is
publicly accessible (von Krogh and Spaeth 2007).
Primary data sources include notes form field observations and discussions with health
workers, administrators, technical staff and local developers as we engaged in project implemen-
tation. Secondary sources include journal papers, conference presentations, PhD and masters
theses (many supervised by us), and various workshops and seminars. Also, since the project
received funding from a variety of donors, in particular through the European Commission’s 6th
Framework Program, extensive project reports have been developed, with detailed description of
activities and progress, supplemented by evaluation reports, funding proposals and notes from
meetings involving managers and staff. These data sources are listed in Table 2.
Academic Theses, dissertations, papers, conference presentations

Funding proposals, progress reports, evaluations, presentations to
Project related
stakeholders
Communication Mailing lists, wiki, issue tracker, source code commit messages, instant
tools messaging meeting records
Field Meeting notes, user observations, developer workshops
Table 2: Data source categories
The empirical focus of this paper is on the trajectory of participatory design in the project,
from its beginnings following a traditional PD approach in a few districts in South Africa, to an
increasingly globally dispersed processes of design and implementation taking place around the
network today. Taking Gumm’s (2006b) four axes of distribution (spatial, temporal, organisa-
tional, and stakeholder related) as an organizing principle, we identified a number of important
distributed interactions impacting on the design, both of the software and of the overall proc-
ess, including implementation. These interactions were then grouped according to similarities
in how they were affected by distribution. Surveying these groups, the following main themes
emerged:
• Design taking place at multiple levels, and the interaction between local and global
design processes

• The organisation and balancing of teams and roles, and their effect on both traditional
PD and DPD
• The role of DSD to foster DPD
These grouped themes form the structure of the case description in the next section.
4 Case – distribution in the HISP network

The Health Information Systems Programme (HISP) spans a 15 year period and represents a
reservoir of field stories across a wide variety of research topics and country contexts. The project
has undergone major transformations from being a small district pilot to becoming a recognised
actor on the global arena of health information systems for the South. This case study focuses
on the trajectory of PD in HISP, particularly during the shift from a locally focused to a globally
distributed project.
We start by presenting a condensed timeline of design activities in HISP (4.1), followed
by a description of what exactly is being designed (4.2) in this project. With this background,
we discuss how various constellations of teams and roles both strengthen and impede local and
global design activities (4.3), and finally describe how the challenges of global software design
have been met in and across the various local contexts (4.4) in HISP.
4.1 Timeline of HISP design

The first phase of DHIS development can be characterised as an intensive three year evolu-
tionary PD process which took the system from a district pilot to a country-wide standard for
Health Information Systems (HIS) in South Africa (Braa and Hedberg 2002). These efforts
were carried out in line with PD practices, and a series of increasingly refined prototypes were
tested in close collaboration with users, to enable information for local action. To some extent
the prevailing post-apartheid reform goals of decentralisation and local empowerment were “in-
scribed” into the software. While the iterative design process produced a close fit with the needs
for sweeping reform, the system accumulated both rigidities and a messy architecture overall.
This proved problematic when it was subsequently introduced in countries such as Mozam-
bique, India, Vietnam, and Cuba after the turn of the millennium. To address this situation, the
project embarked on a completely revised and internationalised version of the software in 2004,
starting from a full remodelling of the database. The developer team was still confined to Cape
Town, and employed the same technology (MS Access), but this time the users were primarily
elsewhere (South Africa kept the existing installations, which were stable and well adapted to
the local situation). Through extensive travelling of project staff, supplemented by e-mail com-
munication, the new internationalised DHIS version was developed as a participatory process
between the developers in Cape Town and implementation teams in Botswana and Zanzibar
(Sheikh and Titlestad 2008).
As the new, internationalised v1 was introduced to new countries, the two-person develop-
ment team became a bottleneck to supporting the expanding network of users with specific re-

quirements. While the technology enabled rapid prototyping, the v1 architecture was not suited
for distributed development, and because of the small and co-located team, no source code shar-
ing tools had been employed. Another shortcoming of the existing architecture (called v1) was
the dependence on the Microsoft platform, which meant that even though DHIS v1 itself was
open source, it required the full MS Windows and MS Office stack to run. These factors trig-
gered yet another revamp of the software. Development of v2 began in 2004 under leadership
of the University of Oslo, but aimed at distributing development activities to a number of the
countries in the network, in order to bring software development closer to the contexts of use.
A stack of “bleeding edge” Java-based technologies was selected for v2, and in parallel a
distributed development platform similar to those employed by many FOSS projects was set
up. However, re-implementing DHIS as a modular web application proved quite difficult. Also,
the radical break in technologies as well as an over-emphasis on the new online communica-
tion platform presented a formidable obstacle to the involvement of existing technical staff in
various sites. The new flexible but complex architecture in effect hindered PD efforts, as it took
over a year and a half before v2 could initially be deployed, first in India, and even then much
important functionality was lacking. The system improved significantly through early use in
India and Vietnam and later also in Sierra Leone, as well as through the involvement of new
software developers recruited locally. While engaging with the global source code, their main
task was to support local implementations, in the process bridging the divide between users and
developers.
After the first pilot in Kerala in the beginning of 2006, use of v2 spread greatly in India,
establishing the credibility of some of the HISP team members to participate in a national HIS
reform process. As a consequence, v2 is planned for installation at the sub-district level all over
the country. The maturation of v2 made it a candidate for inclusion in the Health Metrics Net-
work (HMN) initiative to implement an integrated HIS in Sierra Leone and also meant that it
became part of a WHO recommended stack of FOSS tools for public health.
Table 3 summarises different stages of distribution of developers and user settings.
Timeline Stage Use and Development

Pilot and national Users and collocated software developers, all in South
1994-1999 system Africa, network of users
Expansion Multiple countries, core development isolated from local

2000-2004 modifications
Technological transition Two branches (v1 & v2), infrastructure for sharing, but
2004-2007
fragmented processes, isolated modifications
Integration Multiple local teams, travelling, local developments
2008->
contributing to global software
Table 3: Different stages of distribution of developers and user settings.
Sub-sections 4.3 and 4.4 describe in more detail how the organisation of both people and tech-
nologies was decisive in the design of DHIS during the stages depicted above.

4.2 Global tools and in-country system design
Design of in-country systems – example South Africa

As the existing South African HIS was extremely fragmented, with poor data quality and lacking
standards, the first objective was to develop shared essential definitions of data to be collected
and key indicators to be calculated across the country. The initial development of a minimum
data set in Western Cape Province and software to handle it (DHIS) took place in close collabo-
ration with end users and health professionals. Through a gradual process, several features were
added which to some extent inscribed reform aims and best practices into the software; 1) the
new health policies with a focus on decentralisation and local management and 2) information
for local action approach to public health. The breakthrough came in 1997/98 when first one
province, then a second, implemented new minimum data sets using the DHIS.
The experiences from South Africa have been instrumental for HISP in the development of
participatory approaches to identify and to support information needs within national health
systems. In each country, the more concrete design outputs needed for a well-functioning system
include standardised and clearly defined data sets and indicators, good data flow routines, data
validation, and various standard formats (tables, charts, maps etc.) for data analysis, presenta-
tion and feedback for each organisational level. While the standardisation of health information
has made progress at the global level, there are still pervasive contextual differences between and
within countries, making in-country design processes essential to the implementation of locally
relevant information systems.
A shared toolbox
We use the term global toolbox for systems design to refer to what is being distributed and ap-
plied in the various in-country design processes in the global HISP network. The toolbox con-
tains software (the DHIS), best practices and recommended strategies for overall development
of in-country systems, as well as training material and other documentation.
DHIS is a generic tool rather than a pre-configured database application, with an open
metadata model and a flexible user interface that allow the user to design the contents of a spe-
cific information system without the need for programming. This also allows for rapid prototyp-
ing: for example, during a short initial visit in Sierra Leone, DHIS v2 was quickly customised
with the local organisational hierarchy, and data was imported from various fragmented systems.
The implementation team could then present the rapidly assembled system to major stakehold-
ers in the project, visualizing their own data in indicators and charts. By displaying concretely
what the final integrated system could provide, both technically and in terms of information, the
DHIS served as a powerful interactive tool supporting early discussions.
In a typical implementation sequence, the initial system designed at the national level un-
dergoes a validation process at selected pilot sites, where local managers and health workers play
a central role in modifying and further adapting the system to their needs. Being able to com-
municate through a visual and easily editable tool facilitates local PD processes. For example,

district managers in Sierra Leone became actively engaged in designing additional calculated
indicators through the user interface and could quickly see the results in tables and charts, using
their recently collected data. In Cuba, the sub-district statisticians played a key role in localizing
the organisational hierarchy by replacing “dummy” names such as “Family Doctor Office 1, 2,
3” with meaningful facility names not known at the national level (Braa et al. 2004b). Such
design processes were often carried out as part of on-the-job training. In this way, an important
requirement in cooperative prototyping was achieved; to bring the current skills of the users in
contact with new technological possibilities (Bødker and Grønbæk 1991).
4.3 Teams and roles in HIS design

A number of actors and stakeholders are part of the HISP network; health workers, administra-
tors, public health experts, graduate students, project leaders, and technical staff and trainers
in the various sites where the system is introduced. To strengthen the many local implementa-
tion teams, both from the technical and medical information domains, the University of Oslo
(UiO) introduced graduate programs in health information systems and research projects re-
lated to HISP activities. A number of key project members, with either medical or technical
backgrounds from Ethiopia, India, Malawi, Mozambique, Tanzania, and Vietnam have so far
completed Masters or PhD degrees.
A key strategy in HISP has been to ensure circulation of people across countries, and PhD
students, researchers and consultants from many different countries have through their travel-
ling established bilateral links and learning between countries and also helped to both design
and disseminate the global toolbox. E.g. in preparation for an upcoming implementation of v2
in Tanzania, a team attended a workshop in India to learn about relevant experiences. Other
examples include Ethiopian Masters and PhD students travelling to Tajikistan and India to
support v2 implementation, and a Vietnamese Masters student who spent time in Ethiopia to
strengthen the local development team there.
The influence of the network on local implementation

When v1 was revamped with global use in mind, the South African development team depend-
ed on feedback from Botswana and Zanzibar. Interaction with users in neighbouring Botswana
was mainly facilitated by developer field visits, but with the more distant Zanzibar the approach
was to communicate through experienced global HISP staff. The process of customising DHIS
in Zanzibar revealed a lot of bugs and instability in the new version, causing considerable frus-
tration in the local team, who were simultaneously learning to use the system. From 2005 to
2007, there was an almost permanent presence of HISP members (PhD students and consult-
ants) from Norway, South Africa, and India, both from technical and public health domains.
Furthermore, Masters and PhD students from Tanzania mainland (a three hour boat-ride away)
who enrolled in Oslo programs did extensive field work in Zanzibar, and participated in the
local implementation team. Experienced DHIS implementers from Norway and India had in-
depth knowledge of the DHIS, and frequently reported back to the developers in South Africa

(whom they knew well) on problems. The following e-mail exchange illustrates the importance
of having technically skilled people in the field who are able to provide constructive feedback to
the programming team:
Indian PhD student working in Zanzibar:
Hi, we have some problem in calculated Data element. I tried both in DHIS 1.4.0.37
and 40 but both didnt work. I tried to add some calculated data element and then tried
to export to data mart but the field show blank. And also there is no DataelementCalcalt-
edfield table in datamart file. Acutally, i was going through the code but all the steps for
export to data mart are about aggreation and indicators and others but not the anything
about caluclated data element. I dont know whether i am doing any mistake in setting
up the dataelement.
South African developer:
There is no procedure to auto-calculate these values in the export to datamart process. It

is a process that is currently outstanding but it will be included in build 42.
The Zanzibari implementers were recruited from local technical universities and had no prior
experience with either the DHIS or the health information field. They were immersed in an in-
tensive learning by trying process, working under the guidance of the visiting implementers on
designing data sets and indicators and customising the slowly maturing software. The local team
gradually increased their overall command of the domain and the maintenance of the system,
and gained the capacity to conduct extensive training workshops and facilitate on-the-job learn-
ing by doing in district offices. Fluent in the local language and culture, they could more easily
understand the feedback from local users and engage in often heated design discussions with
key stakeholders. Such intensive engagement with users often led to queries or suggested im-
provements which the implementers then raised with the international team, and subsequently
discussed on the global mailing list or directly with the core developers, in effect forming a
multi-level chain of mediation. However, as they became more familiar with the core team, the
local implementers also communicated directly via email or instant messaging.
Results of user-developer gaps in HIS implementation

When DHIS v2 was first introduced in the field in 2006, the HISP India team members, who
had several years of experience with v1, found it hard to make it useable in their context. At the
time, no report generation or graphing tools were in place. Additionally, the initial implementa-
tion had poor capacity for processing the large amounts of data collected in India, having been
tested mostly with small sample data sets, as illustrated by this e-mail from the local team to the
Norwegian developers:

We couldn’t export data values even after waiting for one complete night after clicking
the button export. Even the calculation of 10 indicators for a year it took more than 7
hours which is not acceptable.
The Indian team also complained that they could not find their data, as the database model was
difficult to understand. Although the model was quite similar to v1, the new version incorporat-
ed a database abstraction layer, inscribing the use of a Java Application Programming Interface
(API) instead of directly accessing the database. However, the Indian team found this hard to
understand, and chose to bypass the API in order to obtain quick results and fill in the gaps in
the global v2 software, e.g., providing “hard-coded” (but well-performing) reports.
In response to immediate user needs, a Dashboard module was developed by the Indian
team in collaboration with health managers, with the purpose of enabling graphical analysis
of data and indicators in charts. This quickly became an important instrument for the Indian
coordinators when demonstrating the capabilities of the system to local authorities. For quite
some time, this module remained outside of the global code repository, and due to many local
workarounds and hacks, it was not compatible or able to plug into the globally shared code. As
such, this became a constant source of friction between the local team and the core developers,
who were asked to help out on problems, yet did not want to deviate from the architectural prin-
ciples they saw as the foundation of the system. The tension eased gradually as the Norwegian
team gained a better understanding of the context and pressures facing the Indian team.
4.4 Expanding the developer network - DSD

Over the expansion stage of the project, local requests for additional functionality in v1 peaked,
and increased the pressure on the small development team in South Africa. Where skills were
available, local additions were developed in-country, as was the case with an Ethiopian ICD
(International Classification of Diseases) module and a Mozambican custom data entry module.
However, due to the architecture of DHIS v1 and lack of tools to support a distributed devel-
opment process, these initiatives resulted in “forks”, i.e. local versions incompatible with the
centrally maintained source code. The isolation caused by such forks hindered appropriation of
global level improvements, and neither did the global toolbox benefit from innovations in the
locally forked software.
This situation triggered the development of v2, which emphasised distribution of design in
order to better utilise the skills around the network as well as meeting local user requirements.
In this way, a DSD process was partly initiated because of the needs to support PD in many
locations. However, setting up architecture for distributed development while at the same time
replicating all the functionality of v1 proved difficult, as illustrated by the initial failure of v2 to
meet user needs in India. The expansion of the developer network meant a wider distribution of
skills, mitigated by travel, mentoring and enrolment in graduate education programs, while still
contributing to the project, both in the home country and globally.
After a long term trial and error process, during which both the coordinators and the net-
work of developers increased their skills and knowledge of the core architecture and processes,
many were able to bring local innovations into the global code base, thus forming a true DPD

network designing the global toolbox. Through close interactions with users and stakeholders
in the various settings where the DHIS was deployed, several local teams created add-ons to the
core application, often quite outside the purview of most of the key developers. The need for
synchronisation with the steadily evolving core code base, combined with the potential use of
the functionality in other countries, meant that there was pressure to integrate these add-ons
with the global repository and “domesticate” them as generic (i.e. not country specific) mod-
ules.
For example, the Dashboard module developed in India quickly became a success there, and
caught the interest of the Norwegian coordinators as having functionality with a clear poten-
tial for more general use. However, the Dashboard reflected a lack of understanding of the v2
architecture and did not follow the intended separation of concerns between global (core) and
local modules. A Norwegian developer visiting India spent a significant length of time pair-pro-
gramming with its developer and committed it to the repository. Shortly after, the new module
was used to great effect at meetings with the World Health Organization in Geneva. Later, the
Indian developer stayed in Oslo for a couple of months, working with the team and attending
the master level course in v2 technologies.
The Ethiopian team faced an important requirement to collect information with a relatively
intricate break down by categories of status, sex, and age, and linked to the International Clas-
sification of Diseases (ICD). Some years earlier the Ethiopians had created a fork of v1 to accom-
modate this functionality. This time, a local developer successfully implemented a module in v2
to support the multidimensional data sets, replacing parts of the core. Although this resolved
the urgent need in Ethiopia, the solution was incompatible with central parts of the global code,
such as the calculation of indicators, and therefore could not be easily shared by other countries.
Shortly afterwards, the Ethiopian developer was recruited into the PhD program in Oslo, where
he started discussions with the core developers on how to merge it into the global code base.
Despite generating several good ideas, a concrete solution remained elusive. However, when the
Ethiopian developer visited Tajikistan a few months later in relation to the customisation of v2,
he quickly saw the benefit of his multidimensional model for their detailed report requirements,
and under severe time pressure, the module was finalised and large parts of the core system were
re-factored. The new module was subsequently utilised to its full potential in Sierra Leone, and
underwent several improvements and further stabilisation as a joint effort between Ethiopian
and Norwegian developers.
5 Discussion
The stages in the HISP case—Pilot and national system, Expansion, Technological transition,
and Integration—illustrate several aspects of the relationship between DSD and DPD. We now
proceed to analyze the distribution of design activities through the interplay between local and
global levels, and the crucial role of implementers and networks.

5.1 Global design through local use
HISP design can be seen as two distinct yet related processes taking place at different levels: 1)
in-country design of health information systems with an emphasis on the system rather than on
software and 2) across-country design of a globally distributed toolbox of software and best prac-
tices for in-country HIS design. In the development model for DHIS v2 as described in the In-
tegration stage, these processes continuously feed into each other; globally distributed solutions
grow out of local designs and use, and the global toolbox is utilised in local design processes (see
Figure 2). The multidimensional data module was designed in an evolutionary process through
iterations of user interaction and prototyping first in Ethiopia, then in Tajikistan and finally in
Sierra Leone, and in order to meet the architectural requirements of the global toolbox it went
through an adaptation process in Norway. In this way innovation takes place through local use;
through cycles of innofusion where the global and local mutually influence each other.
Evolutionary global toolbox design

time
Global Global Global Global

release 1 release 2 release 3 release 4
Local innovations
Global standards
Local use Local use Local use Local use

and PD in and PD in and PD in and PD in
country 1 country 2 country 3 country 4
Figure 1: Evolutionary global toolbox design
Maintaining the focus on PD as the network expanded necessitated distributed teams of de-
velopers engaging in locally relevant design. In time, and through training and the circulation
of people and artefacts, these teams took part in the design of generic modules which had de-
veloped from their own or similar contexts, thereby forming a DSD network. While the DSD
literature mainly discusses virtual teams and offshore outsourcing for the commercial sector in
the US or Europe, the HISP case turns the setting on its head and looks at “outsourcing” of large

parts of development of information systems for the social sector in the Global South, explicitly
aiming to train developers in many locations. In fact, HISP tries to apply the DSD techniques
listed in section 2, in combination with FOSS approaches to dispersed collaboration.
A key design method to spread local innovations across a heterogeneous network is abstrac-
tion or generification (Pollock et al. 2007). The struggle to make the Indian Dashboard module
part of the global toolbox illustrates that it takes considerable effort to make the products of local
design both relevant and known. To achieve this, the software architecture must take the form
of flexible meta-design (Fischer 2008) to balance local and global needs (Rolland and Monteiro
2002). Thus, DPD raises two important issues; 1) the results take the form of standards, if they
are to be distributed and shared, and 2) the participants customise the results to suite their own
needs. The benefit of standards, combined with the democratic principle of users being able to
adapt the system according to their needs, balances the global-local tensions. The standards in
question therefore have to be flexible (Braa et al. 2007) in order to facilitate innofusion.
As long as the initial technological transition was mainly confined to Oslo, there was little
room for sharing of local improvements (which were derided by the Norwegian developers as
“hacks”). However, as the requisite infrastructure and skills gradually evolved through learn-
ing by trying (Fleck 1994), local designs became available as material in the global toolbox,
in a process of generification. Thus, as Figure 1 illustrates, innofusion in a network creates op-
portunities for learning by trying designs developed elsewhere. In this way, the sharing of local
innovations within the network means that there is extensive scaffolding material (Orlikowski
2006) available to the local design of information systems.
5.2 Coping with distribution

The case study demonstrates that the global toolbox does not travel well across geographical
boundaries without being linked to the circulation of people who care about it. The imple-
mentation team in Zanzibar with experienced foreign HISP consultants and PhD students co-
designing with locally recruited implementers combined best practices from the global toolbox
with a thorough understanding of the local context and needs. HISP implementers take the role
as boundary spanners to balance global standards with local needs and to bridge the technical
and health domains (see Figure 2). Implementers acting as mediators (Finck et al 2004) are
involved with extensive negotiations in the redesign of existing paper forms, the determination
of data to be collected, and the determination of appropriate indicators to be calculated and
reported. The HISP project do not only seek to automate existing processes, but act politically to
support the reform of work practices in the direction of effective use of information for action.
Through co-design with more experienced HISP implementers from other countries and
long term processes of learning by using, local team members gradually become effective bound-
ary spanners and more active participants in the global network. Furthermore, PhD students
from other countries learn important lessons from a new context and from interactions with
other implementers in the network.

Global Technical domain Health domain
Standardisation Global developers Global public health experts

Innovation
Implementers National managers
Local developers Local end users

Local
Figure 2: Implementers as mediators
The platform-like character of the software means that it is simultaneously part of an end
product which includes reports and local data standards, as well as scaffolding used by boundary
spanners in collaboration with local users to make systems function in their work settings. We
have seen from the initial phase in Sierra Leone how the flexibility of the platform enabled rapid
production of relatively elaborate boundary objects that facilitated communication between
users and developers.
The HISP variety of DPD is one where local implementers fill crucial roles as intermediaries
with end users. While the project has seen instances of enthusiastic health managers provid-
ing excellent feedback to core developers, such direct communication with end users are very
much the exception. The South African user mailing list shows some promise in this regard,
but strong “champions” and other boundary spanners are needed in each country to enable
anything similar to strong FOSS communities (Barcellini et al. 2008). The long term global
support to capacity building for local design has been crucial to providing sustainability. In most
of the HISP countries implementers were enrolled in graduate programs in health information
systems. However, emphasis was placed on maintaining strong links with field activities, and
all students were encouraged to link their studies to ongoing design projects, as illustrated i.e.
through the long term involvement of Tanzanian graduates in the Zanzibar project.
5.3 The case for networked PD

While acknowledging the very real obstacles to PD introduced by increased distribution, we
believe that the HISP case shows that Engelstad and Gustavsen’s (1993) call for networked ac-
tion research is relevant to DPD, and this emphasis on networking is also very much part of
FOSS best practices (Fogel 2005). Thus, one has to find ways of coping with use in a growing
number of contexts (Obendorf et al, 2009). The benefits of networking and distribution can be
explored along two dimensions; one political and one practical, which share common needs for
standardisation.

The political dimension may be seen in the continuation of the Scandinavian projects from
the 1970’s and 80’s. These projects addressed a key political issue at the time: the threat felt by
workers from the introduction of new technologies. Similarly, HISP is addressing the “hot” issue
regarding equity of health services in the Global South. Also, there was increasing global focus
on reliable health data and practical ways to design HIS. Sustainable development of HISP has
repeatedly involved bottom-up distribution and networking to gain significant traction. Once
a sufficiently broad basic coverage was established, a kind of snowballing effect attracted more
top-down interest from political (and administrative) quarters, as illustrated in the three exam-
ples below.
First, the momentum created by results (and good timing, as other efforts faltered) enabled
gradually widened distribution of use and design in South Africa and a spread of HISP activities
from three districts to the provincial level, then to more provinces, and finally to recognition at
the ministerial level and approval as a national standard. Second, the expansion in India from a
pilot district to state-wide rollouts in several regions of the country established the credibility of
HISP India team members to participate in a national reform process. Problems with the exist-
ing system were highlighted by visualisation of the data using DHIS v2, which thus became a
key reform tool. The revised data sets are currently being rolled out nationally, and v2 has been
formally adopted for the process, which is one of the largest HIS implementations in the Global
South.
Third, the national success in South Africa and new initiatives in a number of countries in
Africa occurred at the same time that international actors placed national health information
systems at the top of the global health agenda, generating interest from a wider audience. Since
action was needed in many countries, and few other actors were operating in several countries
with the same generic and free software, HMN asked HISP to support Sierra Leone with DHIS
v2, and WHO initiated collaboration on software development. As the scale of efforts expanded,
increased emphasis was unavoidably put on efficiency and practical considerations, seemingly
overshadowing the overt political agenda, i.e., somewhat similar to the trajectory of the PD
tradition. Nevertheless, the focus on deprived communities remains the same.
On the practical side, the HISP case also displays clear benefits from distribution and net-
working. On one hand, a string of local implementations have benefited from the global toolbox
of best practices and software, stemming from multiple implementations and cross country
collaboration. On the other hand, as illustrated in Figure 1, the material support afforded by
the global toolbox has been greatly enhanced through the extensive distribution, and the shared
material fit well with Orlikowski’s (2006) notion of scaffolding.
However, the HISP case shows that to ensure working information systems in the South, we
must look beyond temporary scaffolds. The circulation of people is essential to properly exploit
the toolbox. An overall observation from the case is that the most efficient design happens in
many different contexts where technical developers, domain experts and stakeholders meet.
Day-to-day distribution should therefore, if possible, be interspersed with co-location, in line
with the DSD recommendations from Lings et al (2006) listed in section 2.
But even isolated boundary spanning is not sufficient. Thorough local capacity building is
crucial, underscoring the importance of engaging students and health staff in learning by trying,
using and doing. These efforts must have a long term vision, and we propose the concept of insti-
tutional scaffolding to emphasise the importance of activities such as the development of training

programs, academic degrees, and various formal standards, as well as the creation of information
system cadres and ministerial HIS units. Such efforts must be durable to have real impact, and
therefore, this kind of scaffolding must be relatively permanent (analogously with the century
perspective on constructing the Sagrada Familia in Barcelona), and involve institutionalisation.
As these become fully institutionalised and independent, they can be said to have morphed into
Fischer’s (2008) technical and social infrastructures for meta-design.
6 Conclusion
The contributions of this paper consist of adding perspectives to the study of distributed partici-
patory design, by 1) extending it to the Global South and 2) emphasizing the role of implement-
ers as boundary spanning mediators, 3) enabling active generification of innovations that arise in
local settings and 4) collaboration on the DPD of a common toolbox which in turn 5) affords
learning-by-trying through scaffolding.
In the perspective of the PD literature, parallels emerge between the early Scandinavian
projects and the current challenges facing information systems in the Global South. The HISP
case shares many features with for example the influential UTOPIA project (Bødker et al. 1987),
in terms of both practical methods and political aims. The approaches are similar in their focus
on empowerment through participatory capacity building and on the generation of useful tools
for and by people at the margins in global development, technological and otherwise. However,
while the UTOPIA project stranded as commercial technologies improved, the timing is bet-
ter for HISP. Today, FOSS technologies are at the cutting edge of technological development,
and huge corporations are embracing them. For example, IBM is using FOSS for competitive
advantage (Eclipse Foundation 2007). These practices are potentially of immense importance
in the Global South (Weerawarana and Weeratunga 2004). However, as discussed in this paper,
much scaffolding and learning by trying is needed before developing countries will be able to
master them “as their own language”.
Still, as we have shown, the HISP approach to institutional scaffolding of DPD contains
elements of a model that can plausibly be extended beyond its origins in the public health care
sector. Moreover, the major contribution of the HISP case lies in the demonstration that the
learning engendered by use of the system in the Global South can make considerable contribu-
tions to generic software design. In other words, emphasis should be put on facilitating two-way
learning and innofusion processes. Despite the challenges, the design of global software poten-
tially stands to benefit from wide distribution of use.
7 References
AbouZhar, C., and Boerma, T., “Health information systems: the foundations of public
health,” Bulletin of the World Health Organisation, (83:8), 2005, pp. 578-583.

ANC “The Reconstruction and Development Programme,” African National Congress, Johan-
nesburg, 1994a.
ANC “A National Health Plan for South Africa,” African National Congress, Johannesburg,
1994b.
Avgerou, C. Information Systems and Global Diversity, Oxford University Press, 2002.
Barcellini, F., Détienne, F., Burkhardt, J.M. “Users and developers mediation in an Open Source
Software Community: boundary spanning through cross participation in online discus-
sions,” Int. J. Human-Computer Studies, (66:7), 2008 pp. 558–570.
Bjerknes, G. and Bratteteig, T., “User Participation and Democracy: A Discussion of Scandina-
vian Research on System Development,” Scandinavian Journal of Information Systems, (7:1),
1995, pp. 73-98.
Bødker S. and Grønbæk, K., “Design in Action: From Prototyping by Demonstration to Coop-
erative Prototyping,” in Cooperative Design of Computer Systems, Greenbaum, J. and Kyng,
M., editors, Lawrence Erlbaum Associates, Hillsdale, NJ, 1991, pp. 197–218.
Bødker, S., Ehn, P., Kammersgaard, J., Kyng, M. and Sundblad, Y., “A Utopian Experience –
On Design of Powerful Computer-based Tools for Skilled Graphic Workers,” in Computers
and Democracy – A Scandinavian Challenge, Bjerknes, G. et al., editors, Avebury, 1987, pp.
251-278.
Braa, J., Monteiro, E., Riiser, V., Vedel, E., Aanonsen, K., “Experiences from European health
informatics: Lessons for African systems development?,” Health Informatics in Africa: HELI-
NA 93, Mandil, S., Moidu, K., Korpela, M., Byass, P., Forster, D., editors, Elsevier, 1993,
pp. 161-165.
Braa, J., “Community-based participatory design in the Third World,” PDC ’96: Proceedings of
the Participatory Design Conference, Computer Professionals for Social responsibility, Blomb-
erg, J., Kensing, F., and Dykstra-Erickson, E, editors, USA, 1996.
Braa, J., Hedberg, C., “The struggle for developing District health information systems in South
Africa,” Information Society, (18:3), 2002.
Braa, J., Monteiro, E., and Sahay, S., “Networks of action: sustainable health information sys-
tems across developing countries,” MIS Quarterly, (28:3), 2004a, pp. 337-362.
Braa, J., Titlestad, O.H., and Sæbø, J., “Participatory health information systems development
in Cuba – the challenge of addressing multiple levels in a centralised setting,” Proceedings of
PDC 2004 (1), Toronto, Canada, 2004.
Braa, J., Hanseth, O., Mohammed, W., Heywood, A., and Shaw, V., ”Developing Health Infor-
mation Systems in Developing Countries: The Flexible Standards Approach,” MIS Quar-
terly, (31:2), 2007, pp. 381-402.
Byrne, E., and Sahay, S., “Health information systems for primary health care: Thinking about
participation,” Proceedings of the International Federation of Information Processing, IFIP 9.4
and 8.2 Joint Conference on Organisational Information Systems in the Context of Globalisa-
tion, Korpela M., Montealegre R., and Poulymenakou, A., editiors, Dordrecht, The Neth-
erlands, Kluwer, 2003, pp. 237–249.
Carlile, P. R., “A Pragmatic View of Knowledge and Boundaries: Boundary Objects in New
Product Development,” Organisation Science, (13:4), 2002, pp. 442-455.
Chesbrough, H., Open Innovation: The New Imperative for Creating and Profiting from Technol-
ogy, Harvard Business School Press, Cambridge, MA, 2003.

Eclipse Foundation, Eclipse Open Innovation Networks, http://www.eclipse.org/org/founda-
tion/membersminutes/20070920MembersMeeting/07.09.12-Eclipse-Open-Innovation.
pdf, 2007, visited October 2008.
Elden, M., and Chisholm, R. F., “Emerging varieties of action research: Introduction to the
special issue,” Human Relations, (46:2), 1993, pp. 121-142.
Engelstad, P. H., and Gustavsen, B., “Swedish network development for implementing national
work reform strategy,” Human Relations, (46:2), 1993, pp. 219-248.
Finck, M., Gumm, D., Pape, B., “Using Groupware for Mediated Feedback,” Proceedings of
PDC 2004, (2),Toronto, Canada, 2004.
Fischer G., “Rethinking software design in participation cultures,” Automated Software Engineer-
ing, (15:3), 2008, pp. 365-377.
Fleck, J., “Innofusion or Diffusation? The Nature of Technological Development in Robotics,”
Edinburgh University PICT Working Paper (4), 1988.
Fleck, J., “Learning by Trying: The Implementation of Configurational Technology,” Resarch
Policy, (23), 1994, pp. 637-652.
Fogel, K., Producing open source software, Sebastopol, CA, O’Reilly, 2005.
Greenbaum, J., and Kyng, M., (editors), Design at work: Cooperative design of computer systems,
Lawrence Erlbaum Associates, Publishers, 1991.
Gumm, D.C., “Distributed Participatory Design: An inherent Paradoxon?,” Proceedings of
IRIS29, Helsingør, Denmark, 2006a.
Gumm, D. C., “Distribution Dimensions in Software Development Projects: A Taxonomy,”
IEEE Software, (23:5), 2006b.
Hanseth, O., Monteiro, E., and Hatling, M., “Developing information infrastructure: The ten-
sion between standardisation and exibility,” Science, Technology, & Human Values, (21:4),
1996, pp. 407-426.
Heeks, R., “Health information systems: failure, success and improvisation,” Int J Med Inform,
(75:2), 2006, pp.125-37.
von Hippel, E., Democratizing Innovation, Cambridge, MIT Press, 2005.
Imsland, V., and Sahay, S., “Negotiating Knowledge: The Case of a Russian–Norwegian Soft-
ware Outsourcing Project,” Scandinavian Journal of Information Systems, (17:1), 2005, pp.
101–130.
Kensing, F., Methods and Practices in Participatory Design, ITU Press, 2003.
Kimaro, H., and Titlestad, O.H., “Challenges of user participation in the design of a computer
based system: The possibility of participatory customisation in low income countries,” Pro-
ceedings of IFIP WG. 9.4, Abuja, Nigeria, 2005.
von Krogh, G., and Spaeth, S., “The open source software phenomenon: Characteristics that
promote research,” Journal of Strategic Information Systems, (16:3), 2007, pp. 236-53.
Lanzara G. F., Morner M., “Artifacts rule! How organizing happens in open-source software
projects,” in Actor-Network Theory and Organizing, Czarniawska B., and Hernes T., (edi-
tors), Copenhagen Business School Press, Copenhagen, 2005, pp. 67-90.
Lings, B., Lundell, B., Ågerfalk, P. J., and Fitzgerald, B., “Ten Strategies for Successful Distribut-
ed Development,” in The Transfer and Diffusion of IT for Organisational Resilience, Donnel-
lan, B., Larsen, T. J., and Levine, L., (editors), Springer Science & Business Media, 2006.

Loebbecke, and Powell, “Furthering Distributed Participatory Design: Unlocking the walled
gardens,” Scandinavian Journal of Information Systems, (21:1), 2009, pp. 79-108.
Nygaard, K., “The Iron and Metal Project. Trade Union Participation,” Proceedings of the CREST
Conference on Management Information Systems, London, Cambridge University Press,
1977.
Obendorf, Janneck, and Finck, ”Intercontextual Distributed Participatory Design: Communi-
cating Design Philosophy and Enriching User Experience,” Scandinavian Journal of Infor-
mation Systems, (21:1), 2009, pp. 53-78.
O’Reilly, T., The Architecture of Participation, Available from http://www.oreillynet.com/pub/
wlg/3017?wlg=yes, 2003, visited October 2008.
Orlikowski, W., “Material knowing: the scaffolding of human knowledgeability,” European Jour-
nal of Information Systems, (15:5), 2006, pp. 460-466.
Peine, A., “The sources of knowledge: Towards a framework about use, consumption and indus-
trial dynamics,” 25th Celebration Conference on Entrepreneurship and innovation: organisa-
tions, institutions, systems and regions, Copenhagen, CBS, Denmark, 2008.
Pollock, N., Williams R., and D’Adderio, L., “Global software & its provenance: generifica-
tion work in the production of organisational software packages,” Social Studies of Science,
(37:2), 2007, pp. 254-280.
Preece, J., Nonnecke, B., and Andrews, D., “The top five reasons for lurking: improving commu-
nity experience for everyone,” Computer in Human Behavior, (20:2), 2004, pp. 201–223.
Rolland, K. H., and Monteiro, E., “Balancing the local and the global in infrastructural infor-
mation systems,” The Information Society, (18:2), 2002, pp. 87-100.
Sahay, S., Nicholson, B., and Krishna S., Global IT Outsourcing, Cambridge University Press,
2003.
Sandberg Å., (editor), Computers dividing man and work, Swedish Center for working life,
Demos project report no 13, Utbildingsproduktion, Malmø, 1979.
Sarant, S. A., The role of organizational boundary spanners in industry/university collaborative
relationship, Doctor of Philosophy in Psychology Dissertation Thesis, North Carolina State
University, USA, 2004.
Sonnenwald, D.H., “Communication role that support collaboration during the design proc-
ess,” Design Studies, (17:3), 1996, pp. 277–301.
Sheikh, Y. H., and Titlestad, O. H., “Implementing Health Information System in Zanzibar:
Using Internet for communication, information sharing and learning,” IST Africa confer-
ence, Windhoek, Namibia, 2008.
Star, S. L., “The structure of ill-structured solutions: Boundary objects and heterogeneous dis-
tributed problem solving,” Morgan Kaufmann Series In Research Notes In Artificial Intelli-
gence, (2), Huhns M., and Gasser, L., (editors), Morgan Kaufman, Menlo Park, CA, 1989,
pp. 37-54.
Staring, K., and Titlestad, O. H., “Development As Free Software: Extending Commons Based
Peer Production To The South,” ICIS conference, Paris, 2008.
Tierney, W. M., Kanter, A. S., Fraser, H. S. F, and Bailey, C., “Establishing Partnerships to
Promote eHealth in Developing Countries: Lessons from Africa,” available from http://
ehealth-connection.org/files/conf-materials/Establishing%20Partnerships%20to%20Pro-
mote%20eHealth_0.pdf, 2008, visited January 2009.

Weerawarana, S., and Weeratunga, J., Open Source in Developing Countries, Swedish Interna-
tional Development Cooperation Agency, Stockholm, Sweden, 2004.
West, J., and Gallagher, S., “Patterns of Open Innovation in Open Source Software,” in Open
Innovation: Researching a New Paradigm, Chesbrough, H., Vanhaverbeke, W., and West, J.,
(editors), Oxford University Press, 2006, pp. 82-106.

Inter-Contextual Distributed Participatory Design
Communicating design philosophy and
enriching user experience
Hartmut Obendorf
C1 WPS, Germany
hartmut.obendorf@c1-wps.de
Monique Janneck
University of Hamburg, Germany
monique.janneck@uni-hamburg.de
Matthias Finck
effective webwork, Germany
matthias.finck@effective-webwork.de
Abstract. Most studies of participatory design examine the development of a single, cus-
tomized software system that supports typical workflows within a single client organization.
To cope with other use contexts and new forms of work – such as communities and virtual
networks – the traditional repertoire of PD methodology needs to be expanded to deal with
distribution and diversification of users. Based on a ten-year case study, we describe experi-
ences with PD in the development of a groupware system that initially targeted a single use
context, but was continually extended and adapted to new contexts of use with new require-
ments and work practices. To enable distributed participation across contexts, new methods
had to be established: inter-contextual user workshops bring users from different contexts
and developers together to reflect on the usage and design of the software and its further de-
velopment. Commented case studies make this face-to-face interaction persistent, providing
a written documentation of distributed use experiences and design decisions. In the process
of building an inter-contextual community of users, the PD focus shifts from custom soft-
ware development to empowering users in assessing their own practice and technology use.
Key words: Distributed participatory design, community of practice, community of interest,

community building, workshop, case studies.

1 Introduction
Participatory design (PD) has traditionally focused on the quality aspect of software develop-
ment (e.g., Bjerknes & Bratteteig 1994): as users are considered to be the foremost experts
for their work, their involvement in the development process yields better requirements spe
cifications, and result in better system design and more usable software. The developed software
must be adapted to the task, not otherwise.
Another motivation for participatory design originates in its background in the European la
bor movement following the goals of humanization and democratization of work (Czyzewski et
al. 1990; see also Braa and Vidgen 1995 for a historical synopsis); the ‘empowerment’ of workers
exemplified by the construction of tools enabling users to change their own work environment
(cf., Elovaara et al. 2006; Karasti and Syrjänen 2004; Herrmann et al. 2004; DePaula 2004;
Irestig et al. 2004). Users develop skills and acquire tools that empower them to decide how to
change their work practice.
However, PD’s established focus on large organizations and ‘workers’—and on the develop-
ment of a single, contiguous, customized software system representing and supporting typical
workflows within one organization—limits the applicability of PD methods. While very suc-
cessful in bringing together different stakeholders and involving them in design, they need to
be extended to cope with new forms of work. New organizational structures, such as ‘virtual
networks,’ are difficult to include in development using traditional PD methodology as or-
ganizational structures are missing and boundaries between stakeholders become more fluid (cf.,
Finck and Janneck 2005; Janneck et al. 2006).
The traditional PD approach includes many methods that depend on the possibility for
immediate negotiation; it is thus difficult to follow when co-location and shared work contexts
are no longer given for participating users. Distributed participatory design (DPD) needs to cope
with settings where stakeholders are distributed across various dimensions of time, space, and/or
organization (Gumm 2006; Fischer 2004; Franssila and Pehkonen 2006; D’Andrea et al. 2008;
Barcellini et al. 2008).
In this paper, we report on experiences with user participation in a long-term Open Source
development project. The Open Source software, CommSy, is a web-based groupware system
developed to support communication and coordination in working and learning groups by
facilitating the exchange of documents and the sharing of important notes and dates between
users, comparable to, e.g., BSCW, phpBB, or Moodle (Schümmer and Lukosch 2007; Stefanov
et al. 2005; Cole and Foster 2007). The development of CommSy was initiated in an academic
working group that experimented with groupware technology for their own needs. Over the
years, use of CommSy spread rapidly to other departments and universities, and—most recent-
ly—secondary schools and vocational communities. The participatory development process had
to deal with changing user requirements and also with less and less familiar and increasingly
distributed contexts of use—in addition to CommSy becoming the base for a commercial ven-
ture (cf., Kensing 2000). The software was used in new work environments and for new work
practices, and the amount of users and application domains increased continously. Users as
well as designers became more and more distributed, both organizationally and physically. This
challenged and changed the PD process that was started when the user group was still small and
52 • Obendorf, Janneck & Finck

development was confined to a single location (cf., Gumm et al. 2006). Today, the CommSy
development process shows typical characteristics of a DPD process regarding the temporal,
physical and especially organizational dimension of distribution (cf., Gumm 2006).
In CommSy’s ten project years, the development team continually applied participatory
design methods. However, while today’s methods still openly display their PD roots, the inten-
tion for involving users has shifted from the traditional PD paradigm of designing and tailoring
custom software to enabling reflection and appropriation of software by building a community
that is able to communicate and negotiate even as the user base broadens and the application
contexts diversify.
Instead of developing a single application for tasks and work practice within a single organi-
zation, one application needed to meet the needs of distributed groups of users without losing
the focus on ‘empowerment’. Much as in the related PD work for designing for communities
(e.g., Braa 1995; Bødker et al. 1995; Henderson and Kyng 1991; Korpela et al. 1998; Merkel
et al. 2004; Karasti and Syrjänen 2004), ‘empowerment’ of users means to encourage them to
reflect on their own tasks and work processes and to analyze organizational structure and pro
cesses. As several very different communities of practice (Wenger 1998) who all used CommSy
were targeted, another layer was added to analysis: the reflection of the work of users in com-
pletely different contexts, creating an awareness for the requirements of others—a process we
termed inter-contextual community building.
As a consequence, methods for building a community of interest (Fischer 2001) were de-
veloped – people linked not by common practice in their respective work contexts, but rather
by a shared interest, manifested here in the use of the same piece of software (cf., Janneck and
Finck 2006), across several communities of practice. Thus, inter-contextual participatory design
is a form of distributed participatory design focusing on users who are not part of a single com-
munity of practice (examples for distributed CoPs are described, e.g., in Amaury et al. 2006;
D’Andrea et al. 2008) but distributed across several organizational contexts—with spatial and
temporal distribution playing an additional role.
In the following sections, we will describe experiences and results based on our case study
exploring the distributed participation in the development of CommSy. Throughout the Com-
mSy development process, system development and research on system development and design
have been intertwined in the sense of an action research approach (cf., Braa and Vidgen 1995;
Checkland 1991; Mathiassen 1998; Vidgen and Braa 1997); researchers embodied multiple
roles as system developers, evaluators, and support staff for users, always in close contact with
their clients to make the system fit their needs better. Two of the authors of this paper were
directly involved in system development; the third was not member of the development team,
but helped to analyze the project practices. Thus, in this paper we combine an inside view of the
development process with an external analysis and reflection to answer the following research
questions:
• How can PD approaches be carried out in a distributed setting, when one system is
developed for and within several different contexts of use?
• Which traditional PD methods prove difficult in distributed context and what new
problems arise, respectively?
Inter-Contextual Distributed Participatory Design • 53

• What kind of new approaches are necessary to carry out participatory design processes
in distributed settings?
The paper is structured as follows: In section 2, we will introduce the case study in detail.
In section 3, we sum up the resulting challenges and introduce the new techniques that were
developed to meet these challenges. We reflect on different levels of (D)PD and corresponding
techniques in section 4 and finally sketch perspectives for the future development of both the
methodology and (D)PD as a discipline in section 5.
2 Case study: Inter-contextual groupware

development
CommSy is an open source web-based groupware system. In contrast to other CSCW sys
tems, community support was always a top priority, hence the name that stands for community
system. It supports communication (e.g., with discussion forums and a bulletin board) and
the exchange of working materials (with, e.g., file uploads and a group editor) as well as the
organization of work (aided by a shared calendar, to-do lists, etc.). CommSy was originally in-
troduced as an educational tool for the post-secondary level which was used in a variety of fields
including History, Language studies, Education, Economics, and Informatics. Today, CommSy
is no longer bound to the university: It is used in the public as well as private sector, ranging
from industry to secondary schools. A recent poll of the main hosting site resulted in more than
60,000 users (Table 1).
Use context University Secondary schools Industry Other

No. of users 32,800 29,500 2,700 500
Table 1: Number of users in different contexts using the CommSy installation hosted by the
primary application service provider, effective webwork GmbH (April 2009).
Two key features of CommSy are:

• Project Workspaces designed for closed groups of approximately 10 to 30 members (e.g.,
student groups), offering typical groupware functionalities for asynchronous communi-
cation, project management, and information storage and retrieval.
• Community Workspaces incorporate project workspaces into a larger structure support-
ing not only small groups but a community of users (e.g., all members of a school or
university) over a period of time, similar to an intranet structure (cf., Pape et al. 2002).
The user interface design of CommSy is based on abstract design principles which communicate
the design philosophy behind the development process: (1) Ease of use, (2) socially translu-
cent cooperation (cf., Erickson et al. 2002), and (3) simple integration into existing technical
infrastructures are the current design principles describing the shared vision and development
goals.

CommSy—with the system originating in a University context—has always been the subject
matter of various research projects. The development of CommSy started in May 1999 at the
Department of Informatics at the University of Hamburg. Initially, the system was designed
to support the communication and coordination in learning groups in the department. About
half of the initial development team were non-programming domain experts; this founded their
explicit view of “users as the experts – the ones with the most knowledge about what they do
and what they need—and the designers as technical consultants” (cf., Schuler and Namioka
1993, p. xiii).
  1: The CommSy development process

Figure
In the spring of 2001, some of the CommSy developers joined a research project dealing
with computer supported cooperative learning (CSCL) and participatory software development
(cf., Pape et al. 2002). As a consequence, the CommSy development was no longer a volun-
tary, unpaid activity, but grew into a professional software development project, following the
STEPS (Software Technology for Evolutionary and Participative System Development) software
engineering framework (cf., Floyd et al. 1989). STEPS is a methodological framework for evo-
lutionary development in the PD tradition, stressing the importance of close cooperation with
users as well as an extensive use of prototyping to facilitate the embedding of software systems
in meaningful work processes in the user organization.
As a result of the three-year research project, CommSy had evolved into a sophisticated
groupware tool that had been extensively field-tested and refined in the context of university
education. As the CSCL research project ended in 2003, development of CommSy continued
in form of an Open Source project. However, hosting and support services could no longer be
offered free of charge for the universities using CommSy, as this had been the case while develop-
ment was funded by the research project. Instead, a spin-off company began to provide applica-
tion services (ASP) for anyone wishing to use CommSy without hosting their own webserver.
This new commercial perspective led to an exploration of new contexts of use.
One use context that seemed promising commercially were secondary schools. During its
development, CommSy had been used to a small extent in secondary education by schoolteach-
ers who had been introduced to CommSy during their studies at the University of Hamburg.

As existing relations were intensified and CommSy use spread to more and more schools, new
requirements arose to adapt to the specific teaching situation and the needs of especially younger
pupils.
Later in 2004, another research project was launched that focused on software support of
highly networked virtual organizations (cf., Finck and Janneck 2005; Janneck and Finck 2006).
In this project, CommSy was taken as a prototype that was continually adapted and custom-
ized due to the networks’ needs. In this context, a number of requirements regarding specific
project management-related features emerged. The networks wanted to use CommSy to support
acquisition processes and also—in case of successful bids—to coordinate their joint business
projects.
To sum up, during the now ten years of development, CommSy turned Open Source and
new use contexts appeared, ranging from support for learning processes in higher education
and in secondary schools, to the support of virtual organizations. Use in these different contexts
lead to a variety of different and sometimes outright conflicting requirements. The participatory
design process had to meet the challenge of increased physical, in some cases also temporal, but
especially organizational distribution (cf., Gumm et al. 2006).
3 Bringing them all together: techniques for inter-

contextual participatory design
Rooted in Floyd’s STEPS software development model (cf., Floyd et al. 1989), CommSy was
developed from the start using an evolutionary participatory design process (cf., Pape et al.
2002).
To get started a close cooperation with users, a mix of established PD methods such as paper
prototyping (cf., Snyder 2003), user workshops (cf., Greenbaum and Kyng 1991,; Braa 1995),
interviews and scenarios (cf., Greenbaum and Kyng 1991; Kyng 1995; Rosson and Carrol 2002)
were used, typically including a smaller number of especially active users in the process. Fur-
thermore, various feedback channels were established, such as telephone and e-mail support,
and a large user survey was carried out regularly to collect feedback on system use from a larger
number of users (cf., Finck et al. 2004a).
User workshops were a particularly valuable PD instrument in all development phases to
exchange experiences, discuss different uses, and elaborate requirements for the future devel-
opment of CommSy. As a result, users involved in these processes started to develop a shared
understanding of software support in their respective context. Typical characteristics of commu
nities of practice (Fischer 2004; Wenger 1998) emerged—such as the negotiation of meaning
among the members, mutual engagement in joint enterprises, and a shared repertoire of activi-
ties, symbols, and artifacts.
However, some of the specific requirements developed within these closed communities of
practice turned out to be of little or no significance in other contexts. Also, users from different
contexts had little in common regarding their ways of working, in their daily tasks, in the strate-
gies for applying the software to challenges in their work practice, and in the language they used

to describe their work. They were rooted in increasingly diverging backgrounds; naturally, this
limited the number of shared practices. Consequently, the ability of users from different com-
munities of practice to discuss system use on the level of their actual work tasks decreased. There
were still more abstract topics that were of interest to different practices; the interest for specific
ways of working, however, was limited as users shared nothing beyond the software, and their
practice was certain to be independent to that of other users.
This grew into a problem for the development team. On the one hand, exploring new con
texts of use—and thus, new customers—was vital to the commercial interests of the spin-off
company. On the other hand, the development team needed to bundle resources and tried to
avoid parallel implementations that would increase the complexity of the software as well as
administration and maintenance. Another goal was to avoid the fragmentation of the develop-
ment team and process, called forking in open source projects (cf., Stalder and Hirsh 2002).
Also, the different requirements had to be aligned with the original design philosophy.
To meet this new challenge of pooling the interests of different distributed communities,
new ways of bringing users from different contexts together had to be established to balance
their respective needs and emerging requirements. The goal was to enable communication of
members of different communities of practice, both with each other and with the development
team.
This challenge has been described by Arias et al. (2000) and Fischer (2001, 2004) as creating
a community of interest (CoI). A community of interest spans across different communities of
practice (CoP). Unlike CoPs, a CoI is not characterized by common practices of its members in
their respective work context, but rather by a shared interest—in this case study, the object of
shared interest is CommSy. For a community of interest, greater communication problems must
be expected than within communities of practice. Fischer (2001, p. 4) writes, “fundamental
challenges facing communities of interest are found in building a shared understanding of the
task at hand, which often does not exist upfront, but is evolved incrementally and collabora-
tively […] Members of communities of interest must learn to communicate with and learn from
others […] who have a different perspective and perhaps a different vocabulary for describing
their ideas.” In other words, they need to establish a common ground and a shared understand-
ing (see Table 2).
Building such a community of interest would allow—at least to some extent—a joint devel-
opment without overly neglecting specific use contexts as users from different contexts can be
included in the participatory design effort; this meets the challenge created through the greater
dissemination of the CommSy software that was initially developed for a well-defined and fo-
cused community of practice.
In doing so, however, new needs for participation emerged. Existing PD methodology need-
ed to be extended as it traditionally focuses on a single use context and the working practices in
this context in order to fit the software to the task. Balancing the needs of users from different
contexts is inherently more difficult as shared work practices are harder to identify and may
not even exist. Thus, it became necessary to create participation on both the concrete use level
and an abstract design level that can be shared by different groups of participants in the design
process.

Community members
Example tasks and topics
User communities Problems and artifacts in the
in the case study
case study
Different tasks in the Introduction of the Users from an individual
same domain and a community platform for group sharing common
uniform representation e-learning in different practices, e.g., users
schools and subjects from different secondary
CoPs
Project management for schools
virtual companies, use
conventions in virtual
teams
Multiple domains and Software support All participating
CoIs different representational Document management CommSy users
schemes
Table 2: Comparison of Communities of Practice and Interest (cf., Fischer 2001, p. 9).
To this end, two new techniques were introduced in the design process: inter-contextual user
workshops and commented case studies. Inter-contextual user workshops bring users from dif-
ferent contexts and developers together to reflect on the usage and design of the software and its
future development. Commented case studies make this type of face-to-face interaction persist-
ent – providing a written documentation of use experiences and design decisions.
3.1 Inter-contextual user workshops

Starting in 2003, the CommSy development team has been organizing inter-contextual user
workshops regularly about once a year. Their goal is bringing users from different contexts to-
gether in face-to-face interaction to discuss experiences and requirements, thus possibly finding
a minimal consensus regarding future developments.
Participants
Usually, about ten to fifteen users and three to five developers or members of current research
teams attend the one-day workshops. Over the past years, more than 50 users took part in the
workshops.
All moderators of CommSy workspaces—typically teachers or leaders of project groups –
receive individual invitations. Workshops are also announced publicly, e.g., on the CommSy
website, to reach a wider audience.
Participants are typically ‘heavy users’, representing a certain institution or group, who will
communicate the workshop results back to their colleagues. Their motivation is to partake in the
future development of CommSy and to benefit from the experience of others’ ‘lessons learned’.

Setting
Generally, the workshops are framed as one-day events, starting in the late morning. When plan-
ning the workshops, different working conditions, time schedules and time constraints (e.g.,
term breaks, holidays, freelancers’ high workload phases) need to be considered. Also, workshop
locations (in the past, workshops have been hosted by universities and schools) are changed in a
rotational system to allow people from different regions to participate, as the development team
cannot compensate for travel expenses.
Workshops are prepared and moderated by two or three members of the interdisciplinary
development and research team, including software developers and evaluation staff. Classical
moderation and visualization (Metaplan) techniques are used to facilitate discussion. A detailed
record is kept and sent to all participants afterwards together with the photominutes.
Figure  2: Workshop participants discussing mindmaps for future development.
Typically, the workshops start with an extensive round of introductions to help participants
to get to know each other and their respective contexts of use. Then, developers report about
upcoming developments. This is usually the starting point for an extensive discussion of use
experiences, problems, and requirements. The moderators visualize the contributions for clus-
tering later on. Sometimes the participants wish to work on different issues emerging from the
discussion in smaller groups. Topics often address use problems or phenomena that are shared
by many participants, e.g., how to increase active participation in online communication. A
feedback round concludes the workshop.
Since these workshops are not primarily research activities and need to provide an informal
and productive working atmosphere, they are not audio- or videotaped for detailed analysis. The
workshops are evaluated using the records and documentation (posters, etc.). A lot of evaluation
and documentation is actually carried out during the workshops, as participants, e.g., cluster

and prioritize requirements. The results are discussed within the development team afterwards
and consolidated with other evaluation measures such as questionnaires (as described before).
Example
In the following section, a concrete example of an inter-contextual user workshop is described.
The workshop took place from 10.30 A.M. to 5 P.M. at a secondary school. It was organized
by two schoolteachers who had introduced CommSy to their school and provided support for
their colleagues regarding CommSy use. They had already attended earlier workshops and had
offered to host the workshop when a location was needed.
Ten users and three members of the development team attended the workshop. Five us-
ers were also secondary school teachers, three were university lecturers and two were from an
Employers’ Liability Insurance Association which used CommSy as an intranet. Two of the
participants hosted their own CommSy servers while the others made use of the application
services provided by the CommSy development team. Figure 3 reproduces an excerpt from the
workshop record.
Reflection
So far, the development team received thoroughly positive feedback by the workshop partici-
pants regarding the benefits and usefulness of inter-contextual exchange. Being confronted with
perspectives originating in different contexts makes users reflect their own usage; this is pos-
sible from an angle not present within their own community of practice. Challenged by other
participants to explain why certain features are important to them, and contrasting this with
experiences from other backgrounds, they start to think through and sometimes question their
use routines. An example: The demand for highly differentiated access rights, which was voiced
by several users, turned out to be grounded in the transfer of practices participants knew from
other software products. In the workshop, this boiled down to a discussion of trust, hierarchies,
and authority and the value of equality that is inherent in the design of CommSy.
Nevertheless, establishing a dialogue between users from different ‘cultures’ (in the sense of
different working contexts, disciplines, values), speaking different ‘languages’ (conveying very
different experiences and beliefs) is not always easy and requires time. For example, at one oc-
casion entrepreneurs, teachers and university lecturers all spoke of ‘projects’—while addressing
completely different settings. A lengthy argument arose on the correct usage of the term, which
in the end had to be settled by the moderators.
Other conflicts arise when more ‘powerful’ user groups (who are able and willing to pay
for specific developments, for example) threaten to dominate the development of the software.
Inter-contextual user workshops help to set these conflicts at rest, as a compromise is worked out
in direct interaction with other communities of practice. However, sometimes the compromise
does not work out in practice and the problem surfaces again.
Furthermore, developers use the workshops to validate their own design decisions and ideas:
If features are approved by users from different contexts and with different backgrounds, they
are more likely to be helpful for the user community as a whole.

Excerpt From Workshop Record
1. Introduction
The workshop started with the participants introducing themselves, their involvement with CommSy, their
motivation for participation and their expectancies for the workshop. Four clusters emerged for discussion:
Exchange of experiences; problems and queries; feature requests; and participation in the development
process.
Afterwards, the developers gave a short overview of the latest software developments and possibilities for
application service providing.
2. Exchange of experiences
The discussion generally circled around the issue of lurking and motivating users (especially students) to
participate actively within the workspace, respectively. The workshop participants shared experiences and “best
practices”, e.g. methods to encourage online discussions and feedback or useful netiquette rules. Even though
different approaches were regarded as appropriate in the different use contexts (e.g. when dealing with pupils
rather than adults) the participants expressed that they had learned from each other. One participant, who had
produced short videos as “guided tours”, offered to make them available for the other particpants as well.
3. Problems and queries
Afterwards, the participants discussed difficulties they encountered when using CommSy. e.g. problems with
access rights, copyright concerns when distributing digital material, data import/export, or breaking usage rules.
A lengthy discussion arose on the topic of individualization, as participants missed opportunities to configure
their workspaces according to their respective use contexts (e.g. by advocating different terms for workspace
features). So far, the CommSy team refrained from offering individualization possibilities because of the increased
complexity and also possibly inconsistencies that might arise from it, which might in turn impair usability.
Again, even though the participants faced different challenges in their respective contexts of use, they were
able to share some useful advice. For example, one participant working for an Employers’ Liability Insurance
Association offered to ask his legal department for standard form contracts which might be used by the
other particpants as well. The discussion on different terms and their meaning in diverse contexts raised the
participants’ consciousness to watch out for different viewpoints and understandings within their contexts as
well.
4. Feature requests
Besides the issue of individualization, especially school teachers asked for the possibility to trace individual use
paths (e.g. to check whether certain pupils read their assignments). On the opposite, participants from other
contexts pointed out that several users might share a computer and work cooperatively, with only one person
actually logged in. Others argued that students might download a document but not necessarily read it. They
suggested to call for feedback explicitly, for example by requesting written responses/comments to material
posted in the workspace. After the discussion all participants agreed that individual read access information
should not be provided by the system.
Two feature requests were brought forward congruently by all user groups: Workspace templates to save
standard configurations and also content for reuse, and easy integration of other software tools (e.g tools for
online assessment).
5. User participation in the development process
The participants discussed further possibilities for particpation in the development process. They suggested
using blogs and a newsletter in addition to the existing websites and workspaces.
Figure 3: Excerpt from example workshop record.

Inter-contextual user workshops: a novel technique?
In contrast to more traditional PD workshops, inter-contextual user workshops focus on reflect-
ing and enhancing use practices within the current system design rather than designing new or
revised features. Therefore, traditional PD techniques such as paper prototyping or developing
scenarios together with users have not been applied. While traditional PD workshops focus on
support for the respective work context, it’s not the primary purpose of inter-contextual work-
shops to analyze all of the participants’ different backgrounds in detail. Instead, participants re-
fer to the shared knowledge about an existing piece of software to highlight how its design relates
to their tasks. This reflection will always be partial or anecdotal and cannot cover all aspects of
neither the software design nor the respective use context. To some extent, this is a drawback.
However, with users from backgrounds as diverse as in our case study, the immediate challenge
was to enable basic communication. Compared to design discussions related to work tasks, as
in traditional PD workshops, design discussions in inter-contextual workshops address a more
abstract level (cf., section 4).
Inter-contextual user workshops are similar to focus groups – moderated group discussions
(e.g., Krueger and Casey 2000)—as they also make use of discussions among participants. How-
ever, focus groups are mainly a method for data collection, used in qualitative or market research
to measure attitudes and experiences. They are typically not oriented towards compromise, con-
sensus and mutual learning, concepts central to the inter-contextual user workshops described
here. Furthermore, group discussions are only one method incorporated in the workshops. If
appropriate, workshop moderators also make use of a variety of other moderation techniques;
these can be more focused on an outcome as, e.g., some specific design decision and its trade-offs
need to be decided.
3.2 Commented case studies: inter-contextual communication

made persistent
Participants of user workshops often wished to find a more permanent form for the workshop
results and to establish an ongoing exchange between users from different contexts. To facilitate
this, the CommSy development team set up an online discussion forum and invited users to
participate. Yet this platform never came to life; participation was almost nonexistent due to the
lesser immediate value compared to workshops.
Therefore, a new, more compact and more moderated form of documenting and distribut-
ing user experiences was established, called commented case studies. Commented case studies
are similar to a small book or journal. They describe the use of CommSy in different contexts in
original user voices, as users from different contexts are invited to reflect and report on their in-
dividual appropriation of CommSy and share their practices with the user community. Just like
inter-contextual user workshops, they aim at communicating design decisions among a larger
group of people involved in the development process and at enabling interaction between users
of different communities of practice and also between users and developers.

Participants
The experiences reported here are based on two editions of commented case studies from 2004
and 2006, containing 23 case studies from 30 authors (Finck et al. 2004b, Finck and Janneck
2006).
To produce a collection of case studies, the editors—two or three members of the develop-
ment team—contact moderators of CommSy workspaces and ask them to contribute. Natu-
rally, a higher motivation is necessary for contributing a written text than for spending some
hours in a workshop. Typically, contributors have made personal contact with the development
team before (e.g., as workshop participants) and have experienced particpation in the develop-
ment process as fruitful.
The authors are typically ‘power users’, representing a certain institution or group. Their
motivation is to influence the future development of CommSy by giving a voice to their respec-
tive user group and to benefit from the experiences of other users. The second edition of case
collections was initiated and encouraged by users themselves, who wanted to read more about
different use contexts.
Setting
Commented case studies require some time for completion as they loosely follow the publishing
process of a scientific journal.
Invitations for contributions were made by e-mail and included a description of the current
CommSy version that the reports would refer to. Also a common format and structure was
proposed to make it easier to both comprehend and compare the user reports. As a side effect,
the development team used the thematic structure to moderate the writing process and guide
attention to topics they needed information on, much as in the workshop format. In the first
edition, the structure included section titles such as initial configuration, introduction into use
context, use conventions and so forth. This structure was not directly imposed on the authors,
but the template document that was sent out included the section subtitles. As a result, the
structure was accepted and used by almost all authors (some omitted sections they felt were not
relevant for their report).
Contributed reports were subjected to a review cycle that was mainly focused on under-
standibility and spelling, while care was taken not to influence the content of the article. These
reports formed the main body of the resulting case study volume. To preserve the value of the
contributions over time, a software description was prepended to the report collection.
The development team then grouped the different reports and created several classifications
of the practices and problems described in the case studies. An excerpt of the classifications used
in the case study are listed in Table 3: Apart from descriptive and quantitative data character-
izing the individual use case, the reports were analyzed regarding similarities with other use
cases and typical activities such as configuration of the workspace, documentation and support
given to users, etc. This offered pointers to similarities and differences, providing the basis for
developing a common language and the negotiation of requirements that also takes place in the
workshops.

Form of organization (e.g. research group, company)
Characterization of CoP
Reasons for using the software
Number of Users
Quantitative data on use
Number of Instances of the system in use
Configuration Focus of preparatory administrative work
Forms of existing documentation used
Documentation
Additional user-created documentation
Different support channels
Support
Roles and models for support
Main topic the case reports deal with (e.g. information
Focus of use description
management, appropriation etc.)
Features used Software features that were used in the respective context
Table 3: Examples for how the case studies were classified.
Finally, the last section of a volume consists of comments by the development team, discuss-
ing the different use experiences reported in the case studies and the feature requests resulting
from them, and explaining software features and design decisions on the background of these
use experiences. Also, an outlook on the upcoming version of CommSy was added to the vol-
ume to indicate how the development would react to the requirements voiced by the different
user communities.
Example
In the following section, one edition of commented case studies is described in detail.
In the 2006 edition, contributions were made by users from VIRKON, a research project
on virtual networks, Mikropolis, an information management research group, Branta, JUMP
and Consulting Netzwerk, three networks of independent consultants, HR-Verband, a human
resources association and C1 WPS, an IT-consulting firm. All these organizations use CommSy
to support their internal communication and documentation needs. Furthermore, MQ21, a
European management development network, reported on their use of CommSy for organizing
a conference, and a member of the Branta network described his use of CommSy as a commu-
nication platform for a single customer-related project. Finally, the eCampus report detailed how
CommSy was introduced and supported in a University department.
Three different categories were created to group the case studies: preparation of use, appro-
priation of use, and facilitation of use. Furthermore, the case studies were categorized according
to the functionalities of the software that were used and described (Figure 4). Based on these
classifications, the case studies were indexed to enable readers to find case descriptions that
match their use context and practices.

 
Figure 4: Different classifications for indexing the case studies
The case descriptions are about three to four pages in length. In this volume, they are typi-
cally structured along the following paragraphs: Description of use context (name of group/
organization using CommSy, number of members, reason for using CommSy), types of us-
age (objectives and purpose of use, frequency and participation etc.), moderation of use (e.g.,
netiquettes and rules), and evaluation and conclusion (e.g., problems, lessons learned, feature
requests etc.). Figure 5 shows an excerpt of an actual case description.
The case descriptions are followed by an extensive comment (about 10 pages) by the devel-
opers, reflecting their past design decisions, answering questions and commenting on feature
requests raised in the case descriptions. For example, they describe how the CommSy design
principles were slightly altered to meet the requirements of heterogenous use contexts better,
and if and how (or why not, respectively) specific feature requests would be addressed in the
future. Furthermore, they make suggestions how specific use problems might be targeted within
the current software design.

  5: Example of a case description
Figure
Reflection
To sum up, commented case studies consist of indexed and annotated typical descriptions of use
written by real users, and a reflection and explanation of design decisions on the basis of these
case studies. They follow an informally proposed structure, including a short description of the
use context and purpose, the participants, the way the software was introduced and adopted
in the respective setting, and an extensive report of use experiences and ‘lessons learned’. An
extensive introduction written by the editors gives an overview and classification of the cases
presented and helps readers to select the ones that are most relevant or interesting for their pur-
poses (cf., Finck et al. 2004b, Finck and Janneck 2006).
Commented case studies satisfy several possible uses:
For developers, they bundle authentic, unedited reports from different use contexts in a
comparable way to help them develop a more thorough understanding of their requirements
beyond the observable feature requests. Furthermore, the commented case studies serve as a col-
lection of use cases for evaluation or documentation purposes.
Users are given access to the experiences of other users. If desired, they can get in contact
with them to exchange their experiences. Furthermore, by commenting the case descriptions,
developers document design decisions in a transparent way, enabling users to gain insights about
how and why CommSy is developed in the way it is, or—possibly—why design proposals us-
ers made were changed or could not been taken into account. This also makes it possible to

anticipate future design decisions, especially when they do not primarily address the respective
use context.
One major drawback of commented case studies is that the production of a volume is quite
time-consuming. Thus, by the time it is published the case descriptions might already be out-
dated, especially regarding the software version that they correspond to. However, the underly-
ing use scenarios described in the case studies seem to be subject to somewhat lesser changes.
Commented case studies: a novel technique?

Commented case studies can be contrasted to use cases or scenario techniques employed in
software engineering (cf., Hertzum 2003). While use cases are highly formalized and detailed
descriptions relating to the way concrete implementations are planned or carried out, com-
mented case studies describe existing experiences of use in a more anecdotal, less formalized way.
Scenarios, on the other hand, which are similar to commented case studies in their narrative
descriptions of tasks and ways of use, are meant to capture typical, representative descriptions of
use, while commented case studies are highly individual.
4 Reflections on inter-contextual participatory design

To summarize experiences, we found that traditional PD methods are well adapted to work
within a singular context of use, or community of practice. By developing methods for inter-
contextual PD to be applied in project settings where stakeholders are physically and, even more
important, organizationally distributed, we were able to establish a diverse community of inter-
est. The methods we presented here reflect on system usage and design both on a concrete level
of use practices and tasks within specific communities and an abstract level of design philosophy
and underlying viewpoints and values within the different contexts. By addressing values on an
abstract level, users from different contexts succeed in clarifying their requirements without hav-
ing to share details of their daily work routines and practices. Furthermore, both methods aim
at establishing a deepened understanding between users and developers, especially when some of
the developers are almost exclusively involved in just one context. Ideally, users and developers
succeed in speaking a common language.
Applying PD to different communities of practice with their special needs presents a number
of challenges that have not been central to PD research so far. Traditional PD workshops and
future workshops (Greenbaum and Kyng 1991) are useful to criticize and develop existing work
practice in a single community of practice with similar daily work practices and tasks. Even
when different parties participated, they were involved in one context as stakeholders with dif-
ferent interests (e.g., O’Day et al. 1996; Henderson and Kyng 1991). The same is true for other
design techniques, such as probes (see, e.g., Kanstrup and Christiansen 2006 for an example of a
postcard probe or Lindquist et al. 2007 for examples of cultural communication and technology
probes), design games (e.g., Brandt et al. 2008) or various forms of prototyping: They focus on
the design task to support users and designers in finding an optimal solution for their concrete
everyday tasks—“designing for skill and work practices in context” (Binder et al. 2008, p. 2).

Some of these methods have also been applied in distributed contexts before (cf., Lindquist
et al. 2007; Björgvinsson 2008). Hoewever, while dealing with e.g., physical distribution as in
the case of Lindquist et al., who explored design activities with distributed family members,
they were still designing for shared practices and tasks, such as communication between family
members. Björgvinsson (2008), who emphasizes the vital interaction between co-workers rather
than between workers and designers, still explicitly states that this was carried out within a single
community of practice.
For multiple distributed contexts, it is necessary to first establish a common ground for ex-
periences. When DPD aims to build a community with shared interests, it is in our experience
very useful to consequently develop sensitivity for different perspectives among the users. We
introduced techniques to foster a shared understanding among users from different contexts and
gain insights into each others’ perspectives, values, and norms: Both the inter-contextual user
workshops and the commented case studies were successfully employed to foster a community
of interest (although the software developers might have lost some prospective customers who
did not share the common values). While other design techniques might also be adapted to work
towards this goal, this has yet to be investigated in future research.
In contrast to communities of practice, a community of interest requires no shared rep-
ertoire of activities and little mutual engagement between its members. It still needs a shared
repertoire of symbols—a common language—and of artifacts—software being a central one for
participatory design. The community of interest thus has no need for codified knowledge, in-
stead focusing on shared understanding and trying to “make all voices heard” (cf., Fischer 2004).
This matches well with the personal motivation of the users – naturally, not all members of all
communities of practice were interested in learning about software use in unrelated contexts,
but the common language used for inter-contextual community building revealed shared in-
terests, and seemingly unrelated activities successfully served as stimuli for self-reflection. Users
explicitly stated that they were surprised how differently CommSy was used in other contexts,
but nevertheless they learned from the practices employed in these “exotic” settings and adopted
and generated new ideas for their own software use.
The experiences from this case study can be generalized based on the distinction between
the different levels of involvement that coexisted during the development process (see Figure 6):
communication with individual user representatives, with members from a single community
of practice, and with representatives from several different communities of practice that should
form a community of interest.
• Single users / user representatives: A permanent form of participation existed through the
one-to-one link of key users and individual members of the development team. The
nature of this participation can be described as feature-based – design requests often
concerned single features and were based on immediate tasks.
• Community of practice: Development workshops with members of a single community
of practice were scheduled to involve stakeholders without immediate contact with the
development team in the process, to broaden the view on the application context and to
provide a forum for reflection and redesign of organizational practices. While key users
and development team members already cooperating with each other typically initiate
the discussion, active participation of other end users —and developers—is encouraged

in this format. The participatory design process touches both concrete features/tasks and
use practices/organizational structures.
• Community of interest: Inter-contextual user workshops and commented case studies
draw upon experience from different communities of practice. User-user and user-de-
veloper interaction is expanded to include different contexts. This distributed form of
participatory design covers anecdotal aspects of the software where interests meet. To
find a common language, the more abstract discussion is centered on values and their
application within different contexts to form a use vision appropriate to individual com-
munities of practice. Shared values allow formulating a common product vision and
building a community of interest. Both the interested users and all members of the de-
velopment team can improve their understanding of the different contexts the software
tries to serve and thus better understand the direction that design will be taking.
 
Figure 6: Extension of participation scope in user and developer domain and increasingly
abstract level of discussion.

The reported shift from targeting a single community of practice to trying to support hetero-
geneous, distributed contexts changed the aims for participatory design: development could no
longer pursue optimal software quality for a well-defined user audience or cater for a coherent,
finite number of tasks and redesign work practice within a single organization. Rather, com-
promises had to be formed. On a concrete level, limited resources and conflicting requirements
had to be moderated: “what is (already) possible to do with the software?”—with appropriation
(cf., Pipek 2005) taking a more prominent role. On an abstract level, the nature of these com-
promises needs to be reflected. A discourse about the values inherent in the software design and
their application and applicability to different contexts allows an indirect and abstract dialogue
about concrete features. While discussions with users are often centered on feature requests
and new ways of tailoring the software to existing processes, awareness of how the same ‘tool’ is
used in completely unconnected contexts made it possible for the discussion to shift to a more
abstract level. Both users and developers are encouraged to take a step back from their respective
standpoints and their daily routines to gain an understanding of other forms of usage.
Thus, the development of values and a design philosophy for the software acquired a central
role. Values were used to communicate the reason for taking one design direction (and con
sciously ignoring other options) to users (e.g., for maintaining non-hierarchical access rights in
spite of CommSy use in more hierarchical contexts). The design philosophy acquired a commu-
nication function, and changes in the direction of design became visible as the values underlying
the design discussions shifted. Features proposed or demanded by users could be assessed not
only by asking what consequences they would have in respect to certain work practices, but also
how they would affect central design principles.
Creating a shared language for discussing design and understanding what the software is
about enabled users from dissimilar contexts to begin an exchange about their perspectives on
how and why to use this particular groupware system. In our view, this is another facet of ‘user
empowerment’: In addition to the ability to influence organizational structures, users acquire
skills for technology selection, adoption and use. However, sometimes the level of abstraction
lead to misunderstandings, e.g., when the users propagated values of their work that were actu-
ally not so visible in their daily routines (cf., Greenbaum et al. 1994).
Inter-contextual user workshops were a useful tool for fostering a community of interest as
they focus on commonalities between different use contexts in their dual discussion of specific
software aspects and abstract software values. As differences between the communities of prac-
tice mainly reside on the organizational level that is skipped in communal discussion and only
surfaces in personal reflection, workshop communication successfully manages to focus on simi-
larities. Lifting the discussion to the abstract level of values allows an indirect exchange regarding
work practices and organizational structure that would not be possible otherwise.
The commented case studies add to this a persistent form and greater depth. Authors are
given more room to detail their experiences and to reflect on their personal or organizational
work practices. While this is associated with a slower feedback cycle, we see the written format
as a very useful complement to user support, traditional workshops and inter-contextual use
workshops, and as an important link to participatory use documentation. As with all written
documentation, however, the audience must be considered carefully. While the format used
here may be well suited for users with an academic background, target groups who are less ac-
customed to abstract, long-winded discussions may prefer other media.

Users Development Team
Increased acceptance Legitimation for design
Empowerment Domain understanding
Use vision Software vision
Table 4: Benefits of inter-contextual community building.
Based on the different levels of abstraction that we find in both the workshops and the case
studies, we distinguish three levels where the software development process benefited from the
methods described here (Table 4): on the first level, users’ acceptance of development decisions
is increased as they learn about the rationale behind design. Reciprocally, developers obtain
legitimation as their underlying values are being confirmed by users. On the second level, em-
powerment of users comes along with a deepened understanding of the respective domain on
the developers’ side. Building upon this, on the third level, users are enabled to integrate the
software into their work practice, while developers can form a consistent vision spanning the
different use contexts.
5 Summary and future work

In this paper, we analyzed changes in focus and aims of participartory design based on the Com-
mSy case study: In increasingly distributed contexts, the focus shifted from custom software
development to empowering users in assessing their own practice and technology use. While
discussions between stakeholders and developers remained detailed concerning the application
of certain features, they became more abstract in communicating values and design philosophy
instead of analyzing daily work practice. The resulting participation was essential for this project
to foster acceptance for design decisions, enriching user experience, and building an inter-con-
textual community across several distributed use contexts.
With distributed participatory design across several contexts, the repertoire of PD tech-
niques was adapted and changed. New methods for DPD were developed—inter-contextual
user workshops and, as a persistent alternative, commented case studies – and integrated into the
software development process. While the decision to extend PD methodology was at the time
due to practical necessity rather than theoretical reflection, the changes in the user-developer-
interaction can be interpreted as emergence of a new level of abstraction. Introducing values into
discussions with users provided a common language for different communities of practice as a
basis for collective design.
Furthermore, this type of inter-contextual interaction turned out to actually enrich user ex-
perience by encouraging users to experiment with new, creative, and unthought-of ways of use
and also to rethink the very foundations of their work practices.
There is, however, further need to expand the repertoire of distributed participatory de-
sign methods. Especially the issue of the up-to-dateness of the information documented in the

commented case studies might need to be addressed. As sustainability was a main goal for the
research project during which the work described here was performed, its end presents a prom-
ising test case for measuring success. Finally, we seek to explore even further use contexts for
CommSy development and will continue to put inter-contextual PD to the test.
In this paper, we described an alternative PD approach for distributed software development,
where new use contexts were sought that share not the concrete work practices, but common
use values—a community of interest (cf., Fischer 2004) formed by distributed participatory
design. As a discipline, PD will make a decision whether the challenge of building software that
spans individual communities of use is a worthwhile goal. Current PD methodology focuses on
optimizing support for existing and future work practices for different groups within a single
context. Compromises will have to be made if that focus is extended: commercial products often
successfully apply an approach that modularizes functionality and provides different perspec-
tives to different target groups.
6 Acknowledgement
First and foremost, we would like to thank all our users, especially the workshop participants
and the case study authors. We have all learned much from you, and continue to do so. This
work would not have been possible without our colleagues in the WissPro project and at the
University of Hamburg. We would like to thank Christiane Floyd and Horst Oberquelle for
laying the roots for this work, and for continuous support during the last decade. We thank
Susanne Bødker for her comments on an earlier draft, and for helping us to find our own words
for the issues described in this paper. Finally, we would like to thank the anonymous reviewers
for their patient and helpful insistence on numerous improvements to this paper.
7 References
Amaury, D., Charlier, B., France, H. and Esnault, L., “Participatory Design for Developing
Instruments for and with Communities of Practice: a Case Study,” 2nd Workshop on Dis-
tributed Participatory Design, CHI 2008, Italy, Florence (2008).
D’Andrea, V., Baskin, A., and Reinke, R., “A Distributed Software Development Team Meets a
Distributed Community of Practice: Participatory Meetings,” 2nd Workshop on Distrib-
uted Participatory Design, CHI 2008, Italy, Florence (2008).
Arias, E., Eden, H., Fischer, G., Gorman, A. and Scharff, E., “Transcending the Individual Hu-
man Mind. Creating Shared Understanding through Collaborative Design,” ACM Transac-
tion on Computer-Human Interaction, (7), 2000, pp. 84-113.
Barcellini, F., Détienne, F., and Burkhardt, J.-M., “Requirements for design participation in
Open Source Software communities,” 2nd Workshop on Distributed Participatory Design,
CHI 2008, Italy, Florence (2008).
Binder, T., Brandt, E., and Gregory, J., “Design Participation(-s),” CoDesign, (4: 1), 2008, 1-3.

Braa, K., “Priority Workshops: Springboard for User Participation in Redesign Activities,” in
Proceedings of COOCS 1995, ACM Press, 1995, pp. 246-255.
Braa, K. and Vidgen, R., “Action Case: Exploring The Middle Kingdom in IS Research Meth-
ods,” in Proceedings of Third Decenial Conference Computers in Context: Joining Forces in
Design, 1995, Aarhus, Denmark, pp. 50-60.
Brandt, E., Messeter, J. and Binder, T., “Formatting design dialogues – games and participa-
tion,” CoDesign, (4: 1), 2008, pp. 51-64.
Bjerkness G. and Bratteteig, T., “A strategy for work life democracy?” in Proceedings of PDC
1994, CPSR, 1994, pp. 3-12.
Björgvinsson, E. B., “Open-ended participatory design as prototypical practice,” CoDesign, (4:
2), 2008, pp. 85-99.
Bødker, S., Grønbæk, K. and Kyng, M., “Cooperative Design: Techniques and Experiences
from the Scandinavian Scene,” In Human-computer interaction: toward the year 2000, R.
M. Baecker, J. Grudin, W. A. Buxton and S. Greenberg (eds.), Morgan Kaufmann, 1995,
pp. 215 - 224.
Checkland, P., “From Framework through Experience to Learning: the Essential Nature of Ac-
tion Research,” in Information Systems Research: Contemporary Approaches and Emergent Tra-
ditions, H.-E. Nissen, H. K. Klein and R. Hirschheim (eds), North Holland, Amsterdam,
1991, pp. 397-403.
Cole, J., Foster, H., Using Moodle, O’Reilly, 2007.
Czyzewski, P., Johnson, J. and Roberts, E., “Introduction – Purpose of PDC ’90,” in Proceedings
of PDC 1990, CPSR, 1990, pp. i-ii.
O’Day, V.L., Bobrow, D.G., Hughes, B., Bobrow, K.B., Saraswat, V.A., Talazus, J., Walters,
J. and Welbes, C., “Community designers,” in Proceedings of PDC’96, CPSR, 1996, pp.
3-13.
DePaula, R., “Lost in translation: a critical analysis of actors, artifacts, agendas, and arenas in
participatory design,” in Proceedings of PDC 2004, ACM, 2004, pp. 162-172.
Elovaara, P., Igira, F. T., and Mörtberg, C., “Whose participation? whose knowledge? exploring
PD in Tanzania-Zanzibar and Sweden,” in Proceedings of PDC 2006, ACM, 2006, pp.
105-114.
Erickson, T., Halverson, C., Kellogg, W.A., Laff, M. and Wolf, T., “Social translucence: design-
ing social infrastructures that make collective activity visible,” Communications of the ACM,
(45: 4), 2002, pp. 40-44.
Floyd, C., Reisin, F.-M. and Schmidt, G., “STEPS to software development with users,” in
Proceedings of ESEC 1989, Springer, 1989, pp. 48 - 64.
Finck, M., Gumm, D. and Pape, B., “Using Groupware for Mediated Feedback,” in Proceed-
ings of PDC 2004, ACM, 2004a, pp. 45-48.
Finck, M. and Janneck, M., “Hospitality in Hosting Web-Based Communities: Two Case Stud-
ies,” in Proceedings of WBC 2005, IADIS, 2005, pp. 327 - 330.
Finck, M., Obendorf, H. and Pape, B. (eds.), Fallbeispiele der CommSy-Nutzung – Eine Samm-
lung von Nutzungsberichten [Case Studies of Using CommSy], University of Hamburg, De-
partment of Informatics, 2004b.

Finck, M. and Janneck, M. (eds.), Fallbeispiele der CommSy-Nutzung im Kontext virtueller Orga-
nisationen und Netzwerke - Eine Sammlung von Nutzungsberichten. [Case Studies of Using
CommSy in Virtual Organizations], University of Hamburg, Department of Informatics,
2006.
Fischer, G., “Communities of interest: Learning through the interaction of multiple knowledge
systems,” in Proceedings of the 24th Annual Information Systems Research Seminar in Scandi-
navia, Ulvik, Norway, 2001, pp. 1-14.
Fischer, G., “Social creativity: turning barriers into opportunities for collaborative design,” in
Proceedings of PDC 2004, ACM, 2004, pp. 152-161.
Franssila, H. and Pehkonen, M., “Feasibility of iterative design as a challenge in distributed par-
ticipatory design lifecycle,” Nordichi workshop on distributed participatory design, Oslo,
Norway, 2006.
Greenbaum, J. and Kyng, M. (eds.), Design at Work: Cooperative Design of Computers Systems,
Lawrence Erlbaum Ass, New Jersey, 1991.
Greenbaum, J., Snelling, L., Jolly, C., and On’, J., “The limits of PD? Contingent jobs and work
reorganization,” in Proceedings of PDC 1994, CPSR, 1994, pp. 173-174.
Gumm, D.C., Janneck, M. and Finck, M., “Distributed participatory design – a case study,” Nor-
dichi workshop on distributed participatory design, Oslo, Norway, 2006.
Gumm, D. C., “Distribution Dimensions in Software Development Projects: A Taxonomy,” in
IEEE Software, September/October 2006, pp. 45-51.
Henderson A., Kyng M. There’s no place like home: Continuing design in use. In: J. Greenbaum
and M. Kyng (eds) Design at work: Cooperative Design of Computer Systems. Hillsdale, NJ,
Lawrence Erlbaum Ass. (1991), 219 - 240.
Herrmann, T., Kunau, G., Loser, K. and Menold, N., “Socio-technical walkthrough: designing
technology along work processes,” in Proceedings of PDC 2004, ACM, 2004, pp. 132-
141.
Hertzum, M., “Making use of scenarios: a field study of conceptual design,” Int. J. Hum.-Com-
put. Stud, (58: 2), 2003, pp. 215-239.
Irestig, M., Eriksson, H., and Timpka, T., “The impact of participation in information system
design: a comparison of contextual placements,” in Proceedings of PDC 2004, ACM,
2004, pp. 102-111.
Janneck, M. and Finck, M., “Appropriation and Mediation of Technology Use in Stable Self-
organized Online Communities,” in Proceedings of WBC 2006, IADIS Press, 2006, pp.
149-156.
Janneck, M., Finck, M. and Obendorf, H., “Participatory Design: An Issue for Web-Based
Community Development?!” in Proceedings of WBC 2006, IADIS Press, 2006, pp. 274-
277.
Kanstrup, A. M. and Christiansen, E., “Selecting and evoking innovators: combining democ-
racy and creativity,” in Proceedings of NordiCHI 2006, ACM, 2006, pp. 321-330.
Karasti, H. and Syrjänen, A., “Artful infrastructuring in two cases of community PD,” in Pro-
ceedings of PDC 2004, ACM, 2004, pp. 20-30.
Kensing, F., “Participatory design in commercial settings: A conceptual framework,“ in Proceed-
ings of PDC 2000, CPSR, 2000, pp. 116-126.

Korpela, M., Soriyan, H.A., Olufokunbi, K.C., Onayade, A.A., Davies-Adetugbo, A. and Ade-
sanmi, D., “Community Participation in Health Informatics in Africa: An Experiment in
Tripartite Partnership in Ile-Ife, Nigeria,” Computer Supported Cooperative Work (7), 1998,
pp. 339-358.
Krueger, R. A. and Casey, M. A., Focus groups: A practical guide for applied research. Sage Publica-
tions, Newbury Park, CA, 2000.
Kyng, M., “Creating Contexts for Design,” in Scenario-Based Design, J. Carroll (ed.), John
Wiley, 1995, pp. 1-24.
Lindquist, S., Westerlund, B., Sundblad, Y., Tobiasson, H., Beaudouin-Lafon, M., and Mackay,
W., “Co-designing Communication Technology with and for Families – Methods, Experi-
ence, Results and Impact,” in The Disappearing Computer, Interaction Design, System Infra-
structures and Applications for Smart Environments, N. Streitz, A. Kameas, I. Mavrommati
(eds.), Springer, Berlin, 2007, pp. 99-119.
Mathiassen, L., “Reflective Systems Development,” Scandinavian Journal of Information Systems
(10: 1 & 2), 1998, pp. 67-118.
Merkel, C. B., Xiao, L., Farooq, U., Ganoe, C. H., Lee, R., Carroll, J. M., and Rosson, M. B.,
“Participatory design in community computing contexts: tales from the field,” in Proceed-
ings of PDC 2004, ACM Press, 2004, pp. 1-10.
Pape, B., Bleek, W.-G., Jackewitz, I. and Janneck, M., “Software Requirements for Project-
Based Learning – CommSy as an Exemplary Approach,” in Proceedings of HICSS 2002,
IEEE, 2002.
Pipek, V., From tailoring to appropriation support: Negotiating groupware usage, PhD Thesis, Fa
culty of Science, Department of Information Processing Science (ACTA UNIVERSITATIS
OULUENSIS A 430), University of Oulu, Oulu, Finland, 2005.
Rosson, M. B. and Carroll, J., Usability Engineering: Scenario-based Development of Human-
Computer Interaction, Morgan-Kaufman, 2002.
Schuler, D. and Namioka, A., Participartory Design: Perspectives of Systems Design, Lawrence
Erlbaum Assoc., 1993.
Schümmer, T. and Lukosch, S., Patterns for Computer-Mediated Interaction, Wiley & Sons,
2007.
Stalder, F. and Hirsh, J., “Open Source Intelligence,” First Monday, (7: 6), 2002. http://first-
monday.org/issues/issue7_6/stalder/index.html
Snyder, C., Paper Prototyping: The Fast and Easy Way to Design and Refine User Interfaces, The
Morgan Kaufmann Series in Interactive Technologies, 2003.
Stefanov, S., Rogers, J. and Lothar, M., Building Online Communities with Phpbb 2, Packt Pub,
2005.
Vidgen, R. and Braa, K., „Balancing Interpretation and Intervention in Information System
Research: The Action Case Approach,“ in Information Systems and Qualitative Research, A.S.
Lee, J. Liebenau and J.I. DeGross (eds.), London, 1997, pp. 524 - 541.
Wenger, E., Communities of Practice. Learning, Meaning, and Identity, Cambridge University
Press, 1998.

Furthering Distributed Participative Design
Unlocking the walled gardens
Claudia Loebbecke
University of Cologne, Germany
claudia.loebbecke@uni-koeln.de
Philip Powell
Birbeck College, University of London, UK and University of Groningen,
Netherlands
p.powell@bath.ac.uk
Abstract. Participatory design (PD) and its derivative distributed participatory design (DPD)
are examples of collaborative research methods that have been successfully applied to in-
formation systems problems. Yet, there are other collaborative research methods such as
action research and design science that have also been used in the same context. This paper
argues that this trifurcation in collaborative methods is unhelpful and that the ‘walled gar-
dens’ in which these methods exist inhibit learning and the methods’ development. As PD
moves to tackle the problems that arise in distributed projects, it becomes more necessary
to look outside its own domain for solutions. This paper investigates whether collaborative
research projects that are categorized under one method also match the characteristics of
the other methods. It finds that research projects using different methods demonstrate re-
markable similarities concerning research contributions, roots, and methodological guide-
lines, but use different terminologies, and also maintain method-specific publication out-
lets and communities. Thus, insight into some of the issues raised by participatory design in
distributed contexts may arise if PD looks outside its walled garden.
Keywords: Participatory design, action research, design science, text analysis.

1 Introduction
In its initial form, participatory design (PD) was developed in locations where users and de-
signers physically met. While this was common for most IS projects at the time, technological
developments now allow the use of virtual teams that may be distributed across time and space.
Distributed PD poses new problems. However, while there are suggestions that PD could gain
from research in computer supported co-operative work, this paper argues that there are other
collaborative research methods that may also offer useful experiences. Distributed PD needs to
look to distributed action research (AR) and distributed design science (DS) as its antecedents as
well as from within itself. As a start, this paper investigates the similarities exhibited by PD, AR,
and DS. The notion of ‘walled gardens’ is well established in information systems with debates
as to the nature and extent of the ‘problem’ (Holmstrom 2005). Walled gardens are domains that
have rich internal inter-linkages but lack links to and from the rest of the network (Labrogere
2008; Foros 2007). Information flows within walled gardens are constrained and thus learning
from other gardens is limited. Clearly, any domain has the potential to learn from any other.
Niehaves (2008), for example, questions if there is a dominance of positivistic thought or just
that journals tend to publish it, and asks if design-oriented research prefers other outlets. How-
ever, if it can be established that approaches have significant commonalities then this points to
fertile soil where exploration might commence.
Many scholars have called for more collaborative research approaches and methods to in-
crease practical relevance (Daft and Lewin 1990; Deetz 1996; Avison et al. 1998; Benbasat
and Zmud 1999; Lee 1999; Dennis 2001; Hirschheim and Klein 2003) and the theoretical
contributions of research efforts (Daft and Lewin 1990; Davenport and Markus 1999; Lee
1999; Kock et al. 2002; Van de Ven and Johnson 2006). AR is probably the most prominent
and widely-quoted collaborative research method (Orlikowski and Baroudi 1991; Baskerville
and Wood-Harper 1998; Baskerville and Meyers 2004; Davison et al. 2004). More recently, PD
and DS have been applied as methodological grounds for involving practitioners and fostering
relevance for practice (Hevner et al. 2004; Kensing et al. 1998a; Baskerville 2008). The three
methods are frequently applied to understand system development, change, adoption, use, and
acceptance by practitioners (Avison et al. 1999; Baskerville and Meyers 2004; Hevner et al.
2004).
Even though PD, AR and DS all imply collaboration between researchers and practitioners
and partially share philosophical grounds, and despite “a ��
convergence on a set of shared con-
cerns, response to problems, similar practices, and identifiable commonalities among groups”
(Howcroft and Wilson 2003a, p. 4), the methods are often described as unique concerning
their specific characteristics (Clement and Van den Besselaar 1993; Avison et al. 1999; Spinuzzi
2005a).
While Cole et al. (2005) and others have compared these methods, and they feature promi-
nently in the unresolved debate polarizing ‘hard’ and ‘soft’ research approaches (Fitzgerald and
Howcroft 1998), this paper takes a different approach. It investigates whether selected IS re-
search projects not only match the characteristics of the research method by which they are
categorized, but also whether they share characteristics of the other two methods. This is im-
portant, as similar methods, using similar approaches, are rich grounds for cross-fertilization
78 • Loebbecke & Powell

and knowledge creation can be constrained by splitting ontologically hairs in ways that reduce
practitioner relevance or limit IS research productivity. The paper argues that the walled gardens
in which methods currently appear to live need to be opened up.
The data set employed here consists of IS research projects carried out in collaboration with
practice in a variety of organizational units that have subsequently been published in leading IS
research outlets. Text analysis of fifteen IS publications is conducted, each of which focuses on
an IS research project undertaken collaboratively between researchers and practitioners. Finally,
findings, implications and limitations of the investigation are presented.
2 Collaborative research methods

This section analyses the three approaches, PD and distributed PD, AR and DS with the aim of
working towards a comparative synthesis to allow a textual analysis. Each approach is described,
along with its research contribution and discussion of its roots and methodological guidelines.
The comparative synthesis provides the basis for the text analysis that follows.
2.1 Participatory design and distributed participatory design

PD began by striving for increased democracy in the workplace and worker empowerment, and
so challenged the organizational structures of power (Clement and Van den Besselaar 1993;
Kuhn and Muller 1993; Asaro 2000). It has now developed to encompass issues such as better
acceptance of software, too.
Research contribution. PD regards possible design improvements as objects of study (Clement

and Van den Besselaar 1993) and stresses the worker/user perspective during the search for bet-
ter technology fit in organizations. It permits workers to decide how to implement and integrate
technologies in their work practices and aims at system design that integrates workers’ tacit
knowledge of work processes (Floyd et al. 1989; Spinuzzi 2005a; Spinuzzi 2005b).
Roots. PD developed as an architectural discipline responding to organizational specificities

(Kensing et al., 1998a), advocating the difference between theoretical reflection and practical
involvement in design (Ehn 1988). Conceptually, PD is based on the Scandinavian practice
of computing and process design, on Marxist ideology, and on socialist philosophies (Spinuzzi
2005b). It started in the 1970s in Scandinavia from collaborations between computer specialists
and trade unions (Floyd et al. 1989). The exemplary project commonly cited is UTOPIA (from
the Swedish standing for ‘education, technique and product in a perspective of work quality’)
launched in 1981 by the Royal Institute of Technology in Stockholm, the University of Aarhus
in Denmark, and a typesetting union (Ehn et al. 1983). The large trade-unionist tradition char-
acterizing Scandinavian work relations is often seen as having laid the ground for PD (Asaro
2000). In addition, the legacy of the Tavistock Group in London contributed to the founding
Furthering Distributed Participative Design • 79

of PD, which later influenced the American movement ‘Joint Application Design’ (JAD), con-
ceptualized by IBM in the 1980s (Carmel et al. 1993).
PD implicitly integrates the Human Relations School concerned with informal organiza-
tional life (Mayo 1933) and the Motivation School dealing with the needs and expectations
of people at work (Maslow 1954; Herzberg 1966). Due to those roots, PD is positioned at the
intersection of computer sciences, organization studies, and humanities (Floyd et al. 1989).
Aiming to resolve rational problems without forgetting their context, Howcroft and Wilson
(2003b) link participatory methods to the functionalist paradigm which considers IS design as
a contribution to specific needs, and therefore rational.
PD deals with organizational dynamics such as routines, language, and tacit knowledge
(Clement and Van den Besselaar 1993; Asaro 2000). It implies learning by doing and results in
practice-driven knowledge creation (Carmel et al. 1993; Kuhn and Muller 1993). PD empha-
sizes permanent learning inspired by the interaction between researchers and workers. It expects
additional learning cycles to emerge from these interactions (Argyris and Schön 1978).
Methodological Guidelines. PD instructs researchers to enter the organization under investi-

gation in ‘frameworks of cooperation’ (Kensing et al. 1998a, p. 173) comprised of researchers,
designers, and users (Greenbaum and Kyng 1991; Bødker et al. 1993). It requires involvement
of the users in the design activities (Floyd et al. 1989). The degree of participation depends on
the participants’ goals, aspects of the design, and the overall relevance of the input (Damodaran
1996; Asaro 2000). PD recommends excluding managers in order to involve workers and non-
technical staff along the entire system development life cycle (Carmel et al. 1993). Hence, the
PD researchers’ practical concerns for satisfying both workers’ and management’s needs require
particular attention, due to underlying worker-management conflicts in projects (Howcroft and
Wilson 2003a).
According to Kensing et al. (1998a), PD research is based on six principles: (1) ‘Participa-
tion’, i.e., stressing the importance of the users’ involvement, (2) ‘close link to project manage-
ment’, i.e., suggesting the division of the team of designers and steering committee members,
in order to allow each to focus on its respective specializations, (3) ‘design as a communication
process’ (with users), i.e., allowing the tacit components of working conditions and tasks to be
included in the design, (4) ‘combining ethnography and intervention’, i.e., taking into account
organizational tacit aspects such as values, (5) ‘co-development of IT, work organizations and
users qualifications’, i.e., focusing on issues behind IT support in organizations, and finally (6)
‘sustainability’, i.e., demanding certain evidence for long-term usability and acceptance. These
principles have been extended, for example, by employing ethnography.
With the application of ethnography, PD research reflects values and tacit elements of the
workplace (Blomberg et al. 1993; Kensing et al. 1998a). The idea is to paint a comprehensive
image of the workers’ environments and thereby serve a better understanding of workers’ per-
ceptions of their activities (Kensing et al. 1998a). PD implies numerous iterations, punctuated
and initiated by collective evaluations (Carmel et al. 1993). It involves five recurring steps in
the research process: (1) Initial exploration, (2) need discovery, (3) ethnographic studies, (4)
collaborative prototyping, and (5) evaluation of concrete options (Carmel et al. 1993; Clement
and Van den Besselaar 1993; Spinuzzi 2005a).

More recent research in PD has extended the approach to encompass design that takes place
in multiple locations. Distributed PD recognizes that many contemporary design teams are often
not collocated. Naghsh et al. (2006) argue that a limitation of PD is that it primarily focuses on
stakeholders being collocated enabling designers to meet face-to-face. However, many software
projects involve distributed collaboration and many of the stakeholders have differing levels of
expertise and competence (Beynon and Chan 2006). Gumm et al. (2006) identify that many
projects are distributed in several ways, involving dispersed developers or distributed users, and
that this distribution may be physical, organizational or temporal. Physical distribution involves
different locations of people and resources, organizational distribution is related to work struc-
tures, while temporal distribution refers to synchronicity of working hours. The authors posit
two major challenges in distributed PD: (1) The concepts of real participation and of physical
distribution tend to be in conflict—most PD approaches are based on the possibility of face-
to-face meetings and (2) PD approaches do not address the organizational distribution within
the user group or between different user groups. That software development now has to deal
with dimensions such as physical, organizational, and temporal gives rise to challenges especially
concerning communication and knowledge sharing and the role of technology to support this
(Danielsson et al. 2006; Beynon and Chan 2006). Matz (2006) illustrates some of the prob-
lems. These include highly distributed teams having language problems and differing experience
in the technical domain language. The teams worked in different time zones which shortened
timeframes for on-line conferences, and any face-to-face meetings were time consuming and
expensive.
Distributed PD scholars are beginning to suggest solutions to the problems of these di-
mensions. For example, Gumm et al. (2006) offer five good practices. These involve mediated
two-directional feedback to address the organizational and temporal dimensions of distribution,
inter-contextual user workshops to reduce distance between users and developers and between
different user groups and to bridge organizational and physical distance between users from dif-
ferent domains. Further, they suggest commented case studies to enable an exchange between
users of different communities of practice and between users and developers which address the
physical and temporal dimensions of distribution and surveys that are used to bridge physical,
temporal, and organizational distance. Finally, user support helps to overcome physical distance.
Gumm et al. (2006) conclude that organizational distance proves to be the greatest challenge for
maintaining a participatory process.
Such good practice developments are the start of a tool kit of solutions to the problems
of distribution in PD. Components may come from within the PD community or they may
be found in research in similar approaches—such as AR and DS. Indeed, Warr (2006) takes
this further arguing that the distinction between situated or distributed design is false as most
projects are both. He maintains that design practice in reality occurs across all the cells of the
space-time matrix and calls for research to address how to effectively support such a process.
This suggests a need for distributed PD not just to look at research on distribution in other ap-
proaches but to the approaches themselves. With this in mind, this paper now investigates AR
and DS.

2.2 Action research (AR)
AR necessitates action, or, as Lee 2007 puts it, the “raison d’être of AR is, by definition, action”
(p. 44). AR is directed towards problem-solving “performed collaboratively in an immediate
social situation” (Hult and Lennung 1980, p. 247). Hence, it goes beyond isolated solutions
to invoke action and reflection along conceptual frameworks (McKay and Marshall 2001). AR
is a change-oriented method that assumes that complex social processes can best be studied by
introducing change to these processes and observing its effects (Baskerville 2001). Hence, it
focuses on practical problems with theoretical relevance (Clark 1972). The notion of distributed
AR (Adamides and Karacapilidis 2006) is established and may involve research with distributed
communities of practice.
Research contribution. AR contributes to both practice and theory (Baskerville 1999; Basker-
ville and Meyers 2004; Cole et al. 2005). It aims at triggering change and investigates the results
of organizational development in a constructed social system (Lewin 1946; Rapoport 1970;
Avison et al. 1999; Gummesson 2000; Davison et al. 2004). AR allows researchers to enter the
social context under consideration and conduct change in collaborative and mutually nurturing
relations with practitioners (Susman and Evered 1978; Avison et al. 1999). It targets practical
consequences and is usually organized so that each iteration of an AR process adds to the theory
(Vidgen 2002).
Roots. Until World War II, most social science research employed partial and sense data under
positivist methods. As scientists started to study behavioral issues during wartime, a call for re-
search methods respecting the socio-psychological aspects of social reality emerged. AR marked
an appropriate approach (Foster 1972; Susman and Evered 1978). The roots of AR lie in the
works of the Research Center for Group Dynamics in the US on social change and social con-
flicts (Lewin 1946; 1947; 1948) and of the Tavistock Institute of Human Relation in the UK
on socio-technical theory (Emery and Trist 1960), and in Checkland’s (1981) view of human
activity systems. The Tavistock pioneers believed that research projects should improve work
situations that were unsatisfactory in human terms and therefore fostered AR and developed
the so-called socio-technical approach (Mumford 2006). Following its genesis in this post-war
research, AR established itself as a research method suitable for academic fields as diverse as
organization studies and medicine (Baskerville 1999). An early example in the field of IS lies in
the Multiview Methodology (Avison and Wood-Harper 1990).
Generally, the action researcher’s viewpoint is an interventionist one (Baskerville 1999).
Thus, AR differs from positivist epistemological considerations and outcomes due to the impli-
cations of the researcher in the field—“prediction versus making things happen” (Susman and
Evered 1978, p. 597)—even though AR may develop positivist, critical, and interpretive forms
(Klein and Myers 1999). AR follows the main conceptual and epistemological foundations of
pragmatism and social constructivism (Baskerville and Myers 2004). Pragmatism understands
reality as based on the interactions among social selves (James 1890; Mead 1913; Haack 1976).
It demands efforts to understand how people learn and create the structures of their social sys-
tem (Dewey 1938) and paved the way for later constructivism, which stresses the importance of

practice for developing more context-based knowledge (Bourdieu 1977). Other contributions
link AR to post-positivist epistemology (Baskerville and Wood-Harper 1996; Baskerville 1999),
which considers knowledge as a contextual artifact, making positivist tools of evaluation obso-
lete (Susman and Evered 1978).
AR is grounded in the cognitive foundation of learning by acting (Hult and Lennung 1980).
It wants to overcome the “juxtaposition of action and research…of practice and theory” (McKay
and Marshall 2001, p. 47). Hence, it is closely linked to the double loop-learning concept,
stressing iterations in cognitive mechanisms for knowledge creation (Argyris and Schön 1978).
The acceptance of participation and reflection by the researcher in the field under study as a
research method implies that his/her experience and pre-understanding is a valuable asset to the
research (Gummesson 2000), enriched by iterative reflection, from action to concepts and vice
versa (Argyris et al. 1985), reflecting Gadamer’s (1976) hermeneutic circle.
Methodological Guidelines. AR requires the researcher’s participation along the entire change
process, stretching from initial reflections on the social context to implementing change in the
social system (Davison et al. 2004). AR instructs researchers to determine the requirements
for change in dialogue with the actors of the observed social systems. It expects researchers to
interpret the inter-subjective meaning of the observations (Baskerville 1999) and to contribute
to the organizational change processes and reflect upon the scientific knowledge created in the
process.
When conducting AR in IS, it is strongly suggested that the researchers start by stating the
“purpose of any action” from a conceptual and practical/contextual perspective (Baskerville and
Myers 2004, p. 333) and to follow five principles (Davison et al. 2004): (1) The ‘researcher-
client agreement’ should give clarity to both practitioners and researchers concerning the proc-
ess and the conditions of work for both parts. (2) ‘A model of cyclical process’ respects the AR
iterative tradition of research steps along a cycle of activities. (3) ‘The principle of theory’ recalls
the importance of conceptualizing the intervention. (4) ‘The principle of change through action’
puts the focus on change, claiming that any action taken should be an attempt to provide change
and therefore should not be hampered. Finally, (5) ‘the principle of learning through reflection’
assures that the research efforts lead to relevant new insights for other researchers.
As to AR guidelines, Lewin (1947) initially suggested (1) analysis, (2) fact finding, (3) con-
ceptualization, (4) planning, (5) implementation of action, and (6) evaluation. Later refine-
ments have resulted in five iterative steps: (1) Understanding and diagnosis of the situation and
its underlying dynamics, (2) action planning, (3) intervention, (4) evaluation, and (5) reflec-
tion (Susman and Evered 1978; Hult and Lennung 1980; Baskerville and Wood-Harper 1996;
Baskerville and Wood-Harper 1998; McKay and Marshall 2001; Baskerville and Myers 2004;
Davison et al. 2004; Järvinen 2005). Concerning the creation of knowledge, AR briefs research-
ers to ground interpretations in pre-existing knowledge to develop new knowledge following the
hermeneutic circle (Gadamer 1976; Gummesson 2000).

2.3 Design science (DS)
DS—also termed Design Research (DR)—is concerned with creating new and innovative ar-
tifacts. It constructs and evaluates artifacts of technology to meet organizational needs and to
investigate associated theories. That is, it aims at developing theoretically and practically relevant
innovative technologies (Walls et al. 1992; March and Smith 1995; Hevner et al. 2004; Cole
et al. 2005; Williams et al. 2008; Winter 2008). It “addresses research through the building
and evaluation of artifacts designed to meet the identified business need” (Hevner et al. 2004,
p. 79-80). DS fosters system development subject to observation and theorizing (Nunamaker
et al. 1991). It investigates the creation of artifacts in a specific organizational context and calls
for user participation in the project to facilitate a better understanding of the context and thus
adequate development, prototyping and evaluation (Simon 1969; Checkland 1981; March and
Smith 1995; Orlikowski and Iacono 2001; Hevner et al. 2004). Again, distributed DS has
been used for government interactions (Karacapilidis et al. 2005) and in e-business research
(D’Aubeterre et al. 2008).
Research contribution. Answering how to questions (Walls et al. 1992), DS follows the objec-
tive “to create things that serve human purposes” (March and Smith 1995, p. 253). It aims at
designing artifacts that enhance the efficiency of the interaction between humans and technol-
ogy (March and Smith 1995; Orlikowski and Iacono 2001) and, for that purpose, applies design
theories to guide developers and reduce their uncertainties in design. DS intends to enhance IT
use and performance in organizations, striving for organizational acceptance of its outcomes
(Markus et al. 2002).
Roots. The birth of DS as a research field lies in the early efforts of governments and mili-
tary institutions to apply computing technology in the 1950s and 1960s (Banker and Kauff-
man 2004). DS is grounded in the seminal work ‘Sciences of the Artificial’ by Simon (1969)
and draws on other disciplines such as engineering, architecture, or art where problem solving
through design plays a key role (Hevner et al. 2004). “The intellectual activity that produces
material artifacts is no different fundamentally from the one that prescribes remedies for a sick
patient or the one that devises a new sales plan for a company or a social welfare policy for a
state” (Simon 1969, p. 130).
Differing from natural sciences that are based on positivism and foster insights towards
generalization, DS seeks prescriptions. It aims at improving performance through the use of IT
in a given institution, in order “to create things that serve human purposes” (March and Smith
1995, p. 253). Hence, DS enables researchers to theorize about the IT artifact itself, instead
of building variable-driven theories around the artifact (Orlikowski and Iacono 2001). With
its problem-driven orientation, it applies an engineering approach to IT research (Orlikowski
and Barley 2001). Being rooted in pragmatism (e.g., Haack 1976), where “truth lies in utility”
(Cole et al. 2005, p. 326), DS supports a situated and practice-driven vision of research, where
“truth (justified theory) and utility (artifacts that are effective) are two sides of the same coin”
and where “scientific research should be evaluated in light of its practical implications” (Hevner
et al. 2004, p. 77). Considering the interactions between technology and the members of an

organization as parts of the social construct, DS is linked to the constructivist tradition (Hevner
et al. 2004).
DS applies design theory to social reality (March and Smith 1995). It challenges the existing
body of knowledge with regard to a kernel theory that underlies the respective IS design theory
(Walls et al. 1992; Markus et al. 2002). DS implies understanding of the social construction,
its elements, technologies, and organizational members, and their interaction as applied to con-
ceptual and technical development (Iivari 1991). It should be evaluated in light of its practical
implications (Hevner et al. 2004). DS builds on double-loop learning schemes that use feedback
to trigger further research, which in turn enables additional learning (Argyris and Schön 1978).
It considers science as the activity by which theories are not only generated but also tested (Walls
et al. 1992).
Methodological Guidelines. Considering the research process as such, DS relies on an iterative

IS development process which distinguishes ‘build’ and ‘evaluate’ as recurring activities (Keen
and Scott Morton 1978; March and Smith 1995; Markus et al. 2002). To better integrate re-
lated theory in these iterations, an extended DS process comprises five phases: (1) Identification
of needs, (2) grounding in practice, (3) grounding in theory, (4) creation of artifact, and (5)
evaluation and theorizing (e.g., Nunamaker et al. 1991; Markus et al. 2002; Hevner et al. 2004;
Arnott 2006).
March and Smith (1995) and Hevner et al. (2004) recommend that users participate in the
design process in order to understand the context and specific needs, and adequately build and
evaluate the artifact with and by users. The integration of user requirements, system features,
and principles “deemed effective for guiding the process of development” (Markus et al. 2002,
p. 181) is core to successfully conducting DS (Walls et al. 1992). In more detail, Hevner et al.
(2004) suggest seven DS principles: (1) ‘Design as an artifact’, i.e., producing a viable artifact
(construct, model, method, instantiation), (2) ‘problem relevance’, i.e., searching for important
solutions for the business world, (3) ‘design evaluation’, i.e. assuring quality and utility of an
artifact, (4) ‘research contributions’, i.e., reflecting upon the design (how did she/he contribute
to the body of knowledge he used?), (5) ‘research rigor’, i.e., rigorously applying methods along
the process, (6) ‘design as a search process’, i.e. stressing the dual imperative between solutions
and environmental constraints, and finally (7) ‘communication of the research’ underlining the
double audience of stakeholders and research community.1
2.4 Comparative synthesis

PD, AR, and DS clearly exhibit similarities. However, there is disagreement within and between
each of the communities as to how they relate to one another. Järvinen (2007) argues that AR
is similar to DS. He compares seven aspects of AR and DS: concrete results, knowledge pro-
duced, activities, intent, nature of the study, division of labour, and generation, use and testing
of knowledge. He finds the fit between the characteristics of AR and DS to be very high and
concludes that AR and DS can be considered as similar research approaches. In contrast, Bask-
erville (2008), while discussing what DS is not, states that “The community within information
systems with an interest in DS research is engaged in a discourse of discovery”, (p. 441). How-

ever, he feels DS is not AR. AR is focused on problem solving through social and organizational
change. DS is focused on problem solving by creating and positioning an artifact in a natural
setting. AR, he argues, is centered on discovery-through-action. In contrast, DS is centered on
discovery-through-design. Further, AR is a methodology and DS is a paradigm. Iivari (2007)
agrees. He argues that mainstream IS research has lost sight of its DS background but concludes
the opposite to Cole et al. (2005) who maintain that DS and AR share important assumptions
regarding ontology and epistemology. Iivari states that AR may be used to evaluate artifacts de-
veloped in DS and may provide information on how to improve those artifacts. Sein et al. (2007)
however disagree with Iivari. They claim that AR is anti-positivistic as it treats each context as
unique and that AR and DR, although exhibiting differences, are not incompatible and do not
have philosophical stances in conflict. They point to other research such as Järvinen (2007) and
Figueiredo and Cunha (2007) who claim that DR and AR are two faces of the same coin. Cole
et al. (2005) for example, suggest adding an AR cycle at the end of the DR cycle, enhancing both
by borrowing stages and processes or by an integrated approach combining the two.
There seems to be somewhat more agreement about the links between AR and PD. Foth and
Axup (2006) compare AR and PD and “would like the larger participatory design community to
continue the comparison and exploration we have begun here” (p. 1). They hope to promote the
usefulness of AR for PD. They see AR and PD as meta-methodologies or research frameworks
as they involve both qualitative and quantitative methods and tools. Both AR and PD entail
similar conceptualizations of participatory principles although they are quite different in their
intent and purpose. Foth and Axup argue that the key intention of PD is to find ways for people
to get involved in research and design activities that may impact on them. They identify various
dimensions to participation including pragmatic, theoretical, political for PD and political, epis-
temological, ecological and spiritual for AR. For them, the imperative of an AR project is not
only to understand and report on a given problem but also to provoke change through action
and they “think PD and AR frameworks have a similar interest in participation, but different
strategies for doing so and with different intent” (p. 4). One route is a dual approach that com-
bines AR and PD frameworks with PD understanding communication and interaction needs
from requirements and design prototypes, while AR contributes by ensuring that changes in
communication habits, interactions and power is captured by critical reflection, evaluation and
informed action. Clemensen et al. (2007) see PD as a research approach for the development of
technological solutions to real-world problems. Historically, they claim AR’s lack of a concept
of technological development led to the development of socio-technical systems design. Yet,
socio-technical design was mainly able to influence the organization of work not technological
development and it was this lack that was part of the background for the development of PD.
Clemensen et al. find that AR and PD both move in interactive cycles but, whereas AR studies
tend to focus more on current problems and structures, PD focuses on technological solutions
to practical problems. Cahill (2007) describes participatory AR which offers the potential for
challenging the normative production of knowledge by including excluded perspectives.
There is little that compares distributed PD with AR. However, Franssila and Pehkonen
(2006) use action researchers as intermediaries, “the idea of distributed tasks and roles in design
lifecycle for end-users, and for designers from diverse areas of expertise … makes sense but … is
rather hard to proceed time-efficiently and without serious interruptions” (p. 4). They claim that

the difficulties of IT designers being both organizationally and geographically distant are partly
the result of power asymmetries.
As a contribution to this debate, and as a starting point in assessing whether distributed
PD can learn from other approaches, Table 1 summarizes the PD, AR and DS concerning their
main (1) research contributions, (2) roots, and (3) methodological guidelines. This similarity
demands further attention to assess whether these methods, by existing side-by-side but not
interacting, are inhibiting fruitful knowledge exchange.
Research contribution. Despite different outcome foci (social change, artifact design, and
technology design), the three methods aim at problem-solving and influencing organizational
settings in given practical situations. Referring to change in social systems, AR remains more ab-
stract than DS, which prescribes the creation of an IT artifact and studies the consequences of it.
However, the two are similar insofar as “action in action research is itself an artifact” Lee (2007,
p. 49). Referring to technology design, PD fills the gap between the other two methods.
Roots. The three methods involve system development as prime theoretical concerns, but either
have a focus on social sciences (AR), a combination of social and technically-oriented sciences
(PD), or an emphasis on technically-oriented sciences (DS). They commonly refer to the con-
cept of learning in the context of the research project. Although their respective history and
‘birth’ differ, the Tavistock Institute has influenced both AR and PD. Also, all three share epis-
temological grounds in the concept of pragmatism.
Methodological Guidelines. The three methods imply participation of organizational members

in the research. They instruct researchers to intervene in the organization and collaborate with
the practitioner (and sometimes the sponsor) in AR, the user in DS, and the worker—de facto
excluding the manager or sponsor—in PD. All three imply subjective elements, challenging the
researcher’s classic objectified ‘observationist’ role. Further, they all demand iterative steps in a
cyclical process including (1) an initial understanding of the situation (diagnosis in AR, needs
identification in DS, initial exploration in PD), (2) researcher intervention in the organization
studied, and (3) reflection and evaluation of results before re-initiating the cycle.
This section has developed a comparative synthesis of the three research approaches. The
next section describes the research approach which asssesses whether there is common ground
that is unexplored that could assist in the development of PD and also distributed PD.
3 Research approach
This research now applies text analysis, a specific type of content analysis concerned with sys-
tematic reading of a body of texts, images, and symbolic matters. By analyzing textual data,
text analysis helps to retrieve pre-defined structures in texts and allows inferences on the basis
of retrieved structures (Krippendorf 2004). Text analysis is a widely applied qualitative method
in the social sciences and has been frequently used in the fields of anthropology and cultural

Method
AR DS PD
Research contribution
Change of (social) IT artifact creation, Technology design,
system, organizational improved human/ user empowerment to
Objectives
development IT interaction, contribute to design
technology acceptance
Practice / Theoretical Practice / Theoretical Practice / Theoretical
Focus
development work consideration
Roots
Susman and Evered Walls et al. ‘92; Greenbaum and Kyng
Selected core ‘78; Baskerville and March and Smith ‘95; ‘91, Clement and Van
references Wood-Harper ‘98; Hevner et al. ‘04 den Besselaar ‘93,
Avison et al. ‘99 Kensing et al. ‘98a
Social Sciences Engineering, Social Science,
Theoretical grounds
Computer Science Computer Science
Learning approach Learning by acting Double-loop learning Learning by doing
Methodological guidelines
Researcher On-site in social setting On-site in On-site at work place
intervention organization
Practitioner User contribution to Worker contribution to
Org. members’
contribution to entire artifact design full-development cycle
involvement research
Determining change/ Dialogue Observation Ethnography
design requirements
Phased, continuous, Phased, continuous, Phased, continuous,
Research process
iterative iterative iterative
Table 1: Method characteristics
studies (Bernard and Ryan 2000), business and management history (O’Connor 1999), or-
ganization studies (Prasad and Prasad 2000; Munir and Phillips 2005), organizational behavior
(O’Connor 1995; Gibson and Zellmer-Bruhn 2001), and strategic management (Vaara and
Tienari 2002; Nørreklit 2003).
In IS research, text analysis has been utilized by Järvenpaa and Ives (1990) who investigated
firms’ involvement in IT based on letters to shareholders included in business reports, by Scar-
brough et al. (2005) who reported on the role of professional media in the diffusion of knowl-
edge management, and by Gallivan and Depledge (2003) who conducted a structured content
analysis of 16 published case studies on inter-organizational systems to understand the relation
between systems use and trust and control.
Within text analysis, Lacity and Janson (1994) distinguish three approaches—positivist, lin-
guistic, and interpretivist. The positivist approach, used here, considers texts as objective data rep-
resenting the reality without interferences. The positivist researcher works primarily on content

analysis, analyzing the text verbatim from a representative sample of textual data by means of
categories that she/he has identified from the literature. The researcher uses quantitative meas-
ures and validity criteria. The linguistic approach of text analysis considers texts as emergent and
influencing a shared definition of reality. It relies on discourse analysis theory (Lacity and Janson
1994) and focuses on the meanings given to concepts and issues by means of interpretation of
and confrontation among various points of views. In the linguistic approach, language is more
than a representation of reality; it is a part of the shaping of reality. “Through language, people
declare couples married, anoint ships, grant academic degrees, and arrest suspects” (Lacity and
Janson 1994, p. 145). The interpretivist approach regards texts as subjective data that necessitate
the researcher’s immersion in the group that produced the content in order to deeply understand
the meaning underlying the message. Whereas the interpretivist approach requires an insider’s
perspective on the data, both the positivist and the linguistic approach consider the researcher as
an external analyst, able to interpret by means of structured analytical approaches.
Conducting a positivist text analysis, this research probes 15 selected IS publications (see
also Loebbecke et al. 2007). Specific papers are chosen for this analysis from the population that
is available. Thus, the publications analyzed here are not a random sample but a purposefully
chosen one. Each publication reports on a system or software development project that occurred
in collaboration between researchers and practitioners and uses one of the candidate research
approaches. Then, the set of papers to be analyzed was chosen based on citation counts with
higher counts being preferred. The publications, five per method, were also selected to assure
heterogeneity with regard to authors and outlets. Table 2 identifies the author, publication outlet
and title of the research categorized by self-reported research approach. All the publication out-
lets are well established, some appearing high in many journal ranking indices. The publications
include MISQ, EJIS, ISJ, JMIS, DESRIST and CACM.
Some care needs to be taken as the terminology in different communities becomes estab-
lished though the terms may be describing the same phenomena. For example, terms such as
‘worker’, ‘user’, and ‘practitioner’ may all be used to describe a group of people that is affected
by, or will use, the software. Indeed, such terms may be used to describe the same person with
the same characteristics across different approaches.
From each publication quotes, i.e., sentences or set of words, were extracted which relate to
the applied method in the respective IS project (a similar research approach is used in Jarvenpaa
and Ives (1990) and Scarbrough et al. (2005)). These were coded and the respective content
schemes set against the characteristics of PD, AR, and DS from Table 1. Finally, the text analysis
was summarized to assess whether projects conducted under one method also meet the charac-
teristics of the other two methods. Table 3 provides an overview of these findings and the results
are described next.

Authors (Year) Outlet Title
Action research
Lindgren et al. ‘04 MISQ Design principles for competence management
Salmela et al. ‘00 EJIS systems: A synthesis of an action research
Information systems planning in a turbulent
environment
Vidgen ‘97 ISJ Stakeholders, soft systems and technology: Separation
and mediation in the analysis of information systems
requirements
Braa & Hedberg ‘02 TIS The struggle for district-based health information
systems in South Africa
Fruhling & De Vreede JMIS Field experiences with eXtreme programming:
‘06 Developing an emergency response system
Design science
Markus et al. ‘02 MISQ A design theory for systems that support emergent
knowledge processes
Miller et al. ‘06 DESRIST* Using a digital library of images for communications:
Comparison of a card-based system to PDA software
Peffers et al. ‘06 DESRIST* The design science research process: A model for
producing and presenting information systems
research
Haynes ‘06 DESRIST* Design knowledge as a learning resource
Jones & Gregor ‘06 DESRIST* The formulation of an information system design
theory for e-learning
Participatory design
Kensing et al. ‘98b CSCW Participatory design at a radio station
Anderson & Crocca ‘93 CACM Engineering practice and co development of product
prototypes
Bødker ‘96 HCI Creating conditions for participation: Conflicts and
resources in systems development
Clement ‘94 CACM Computing at work: Empowering action by ‘low-level
users’
Grønbæk et al. ‘93 CACM CSCW challenges cooperative design in engineering
projects
*International Conference on Design Science Research in IS Technology 2006, Claremont, CA
Table 2: Publications included in text analysis
4 Text analysis results

In Table 3, results are presented as follows: two check-marks represent a very strong match
where one or more quotes fully support the characteristic of the respective method. One check-

mark indicates that one or more quotes resemble the characteristic of the respective method,
and finally, a dash means that there is no match. In evaluating the overall fit of a project with a
method (the last column), three check-marks represent a very strong overall fit with more than
seven characteristics matching. Two check-marks indicate a strong overall fit with more than
five characteristics matching, while one check-mark signals a weaker overall fit with only three
characteristics matching.
Research Methodological Overall

Publiction Roots
contribution guidelines fit
OBJ FOC SCR THG LP RIN OMI CDR RP
Action research
AR √√ √√ √√ √√ √√ √√ √√ √√ √√ √√√
Lindgren
DS √√ √√ √√ √√ √√ √√ √√ √√ √√ √√√
et al. ‘04
PD √ √√ - √√ √√ √√ √ √ √√ √
AR √√ √√ √√ - √√ √√ √√ √√ √√ √√√
Salmela et
DS √√ √√ - √√ √√ √√ √√ √√ √√ √√√
al. ‘00
PD √√ √√ √ √√ √√ √√ √√ √ √√ √√
AR √√ √√ √√ √√ √√ √√ √√ √√ √√ √√√
Vidgen
DS √√ √√ - √√ √√ √√ √√ √√ √√ √√√
‘97
PD √ √√ - √√ √√ √√ √ √ √√ √
Braa & AR √√ √√ √√ √√ √√ √√ √√ √√ √√ √√√
Hedberg DS √√ √√ - √ √√ √√ √√ √√ √√ √√√
‘02 PD √√ √√ √√ √√ √√ √√ √√ √ √√ √√
Fruhling AR √ √√ √√ - √√ √√ √√ √√ √√ √√√
& De DS √√ √√ - √√ √√ √√ √√ √√ √√ √√√
Vreede
‘06 PD √ √√ - √ √√ √√ √√ √√ √√ √√

Design science
AR - √√ √ √√ √√ √√ √√ √√ √√ √√
Markus et
DS √√ √√ √√ √√ √√ √√ √√ √√ √√ √√√
al. ‘02
PD √√ √√ √ √ √√ √√ √√ - √√ √√
AR - √√ - - √√ √√ √√ √√ √√ √√
Miller et
DS √√ √√ - √√ √√ √√ √√ √√ √√ √√√
al. ‘06
PD √ √√ - √ √√ √√ √ √ √√ √
AR - √√ - √√ √√ √√ √√ √√ √√ √√
Peffers et
DS √√ √√ √√ √√ √√ √√ √√ √√ √√ √√√
al. ‘06
PD √ √√ - √√ √√ √√ √√ √ √√ √√
AR √√ √√ - √ √√ √√ √√ √√ √√ √√
Haynes
DS √√ √√ √√ √√ √√ √√ √√ √√ √√ √√√
‘06
PD √ √√ - √√ √√ √√ √√ √√ √√ √√
Jones & AR √√ √√ - √√ √√ √√ √√ √ √√ √√
Gregor DS √√ √√ √√ √√ √√ √√ √√ √√ √√ √√√
‘06 PD √ √√ - √√ √√ √√ √ - √√ √
Participatory design
AR √√ √√ - √√ √√ √√ √√ √√ √√ √√√
Kensing
DS √√ √√ √ √√ √√ √√ √√ √√ √ √√
et al. ‘98
PD √√ √√ √√ √√ √√ √√ √√ √√ √√ √√√
Ander- AR - √√ - - √√ √√ √√ √√ √√ √√
son & DS √√ √√ √ √√ √√ √√ √√ √√ √√ √√√
Crocca
‘93 PD √√ √√ √√ √√ √√ √√ √√ √√ √√ √√√
AR √√ √√ - - √√ √√ √√ √√ √√ √√
Bødker
DS √√ √√ - √√ √√ √√ √√ √√ √√ √√√
‘96
PD √√ √√ √√ √√ √√ √√ √√ √√ √√ √√√
AR √√ √√ - √√ √√ √√ √√ √√ √√ √√√
Clement
DS √√ √√ √ √ √√ √√ √√ √√ √√ √√√
‘94
PD √√ √√ √√ √√ √√ √√ √√ √√ √√ √√√
AR √ √√ - - √√ √√ √√ √√ √√ √√
Grønbæk
DS √√ √√ - √√ √√ √√ √√ √√ √√ √√√
et al. ‘93
PD √√ √√ √√ √√ √√ √√ √√ √√ √√ √√√

OBJ – Objectives RIN – Researcher intervention
FOC – Focus OMI – Org. members’ involvement
CR – Selected core references CDR – Change/Design requirements
THG – Theoretical grounds RP – Research process
LP – Learning approach
Table 3: Matching fifteen IS projects to characteristics of three methods
When analyzing the fifteen projects from the selected publications on the level of method
characteristics (see Table 1), 26 (of 45) combinations of project and method show a strong fit,
indicating a very strong match for most individual characteristics. Fifteen strong fit combina-
tions result from each publication following ‘its own’ method. The additional eleven strong fits
(out of 15 papers) between project and method indicate a match of a project with one of the
other two methods on the majority of individual characteristics.
Most AR projects show a strong similarity with DS characteristics; and some AR projects
also reflect PD characteristics. DS projects present some similarity with the other two methods.
Finally, PD projects indicate an almost perfect resemblance with DS characteristics and a very
strong similarity with AR characteristics (Tables 3, 4, and 5).
AR project DS project PD project

characteristics characteristics characteristics
AR DS PD AR DS PD AR DS PD
Objectives √√ √√ √ √ √√ √ √√ √√ √√
Focus √√ √√ √√ √√ √√ √√ √√ √√ √√
Selected core references √√ --- √ --- √√ --- --- --- √√
Theoretical grounds √√ √√ √√ --- √√ √√ √√ √√ √√
Learning approach √√ √√ √√ √√ √√ √√ √√ √√ √√
Researcher intervention √√ √√ √√ √√ √√ √√ √√ √√ √√
Org. members’
√√ √√ √ √√ √√ √ √√ √√ √√
involvement
Determining change/
√√ √√ √ √√ √√ √ √√ √√ √√
Design requirements
Research process √√ √√ √√ √√ √√ √√ √√ √√ √√
Table 4: Summary of projects meeting method characteristics
Four characteristics, ‘research focus’, ‘learning approach’, ‘researcher intervention’, and ‘re-
search process’ show a perfect match; at least 44 of the 45 selected projects show evidence of

all three methods. Five other characteristics show strong evidence. ‘Organizational member in-
volvement’ has 41 very strong matches. Also, ‘determining change/design requirements’, ‘theo-
retical grounds’, and ‘research objective’, have 36, 33 and 32 of 45 possible very strong matches
respectively, illustrating a good fit. Only ‘selected core references’, with 16 of 45 matches, does
not offer much evidence for projects also meeting the characteristic of the other two methods
(Table 5).
AR project DS project PD project
characteristics characteristics characteristics
AR DS PD AR DS PD AR DS PD
√√ Δ Organizational improvement vs. √√
Objectives
artifact design
Focus √√ √√ √√ √√ √√ √√ √√ √√ √√
Selected core √√ Δ Slightly different core references √√
references
Theoretical grounds √√ Δ Computer science vs. social science √√
Learning approach √√ √√ √√ √√ √√ √√ √√ √√ √√
Researcher intervention √√ √√ √√ √√ √√ √√ √√ √√ √√
Org. members’ √√ Δ Managers’ vs. workers’ involvement in PD √√
involvement
Determining change/ √√ √√ Δ Ethnography only in PD √√ √√
Design requirements
Research process √√ √√ √√ √√ √√ √√ √√ √√ √√
Table 5: Summary of method characteristics across projects
PD Publications. All five PD publications show that the respective projects also meet AR char-
acteristics. For instance, Bødker (1996, p. 220) aims at “general processes of organizational
development”. Kensing et al. (1998b, p. 244) “discuss coordination in a complex organization
with multiple, different and reconfigurable groups”. Supporting a learning-by-acting approach
typical for AR, Anderson and Crocca (1993, p. 50) apply “[o]pen and continuous feedback”,
and Bødker (1996, p. 226) points to “a structured brainstorming activity meant to emphasize
critique, fantasy, and realization in three phases”. Concerning users’ involvement, Anderson
and Crocca (1993, p. 50) argue that “[t]he team recognized the need to involve the users from
the beginning in all planning and development activities”. Clement (1994, p. 57-58) finds that
“[m]anagement…released the clerical staff to participate in project activities” involving the or-
ganizational members on an ongoing basis, where “the study circles were an impetus…generally
reshaping the design process as a whole so it was more in tune with the organizational realities”
(1994, p. 58). Kensing et al. (1998b, p. 248) report that the “analytic activities conducted by
the design team involved approximately one third of the total 140 employees”. By using inter-

views and workshops, Grønbaek et al. (1993) worked to foster an AR-typical dialogue including
observations, interviews, workshop, prototyping, and evaluation “in which users were actively
involved” (p. 69).
Finally, PD projects also meet DS characteristics. Bødker’s (1996) project “deals with user
participation in the design of computer applications” (p. 216). Her objective of designing an IT
artifact is clearly articulated; the “purpose of the project was to design a number of computer
applications” (p. 219). Clement’s (1994, p. 58) objective is ‘the specification of an IS’. Kensing
et al. (1998b, p. 244) “use the term “design” in the same way as architects do – focusing on the
analysis of needs and opportunities”. They explicitly appreciate that the “employees gave valu-
able feedback” (p. 253) to the design effort. As in DS, the authors focus on the actual design
and the typical DS learning cycle. Grønbaek et al. (1993, p. 67) cover the interaction between
technology and individuals, a classic DS domain. Their long-term project is centered around
approximately 10 developers and 20 users. Concerning the learning approach and research proc-
ess, Clement (1994, p. 58) claims that “the study circles were an impetus in structuring (and
restructuring) the design team, overcoming an impasse”. Anderson and Crocca (1993, p. 50)
wanted the research process “to support an iterative, evolutionary development, delivery, and
evaluation process”, hence clearly seeking one that also meets DS characteristics.
AR Publications. The five AR publications offer numerous quotes in support of DS character-

istics. Lindgren et al. (2004, p. 443), for instance, illustrate a typical DS objective to “identify
design principles for CMS” and also show substantial support for double-loop-learning by in-
dicating “a second action research cycle” (2004, p. 446). Vidgen’s (1997, p. 27) project is “con-
cerned with the development of software for wind tunnel operation”. Braa and Hedberg (2002,
p. 119) report on practitioner contribution to artifact design in a “period of active prototyping
and user interaction”. Salmela et al. (2000, p. 7) stress the user contribution in the IS projects
they studied over a longer period of time: “Data collection continued via personal visits to the
research sites, telephone interviews, and conversations. The last contacts with the client organi-
zations concluded seven to nine years after the initial project proposals”. Regarding the research
process and the change/design requirements, Fruhling and De Vreede (2006, p. 48) state that
“[a]fter the release of three key prototypes at different points in time, full-system usability evalu-
ation was done”. Vidgen (1997, p. 30) claims that his research process “consists of the following
activities: current situation analysis, systemic stakeholder analysis, requirements capture and
future analysis”—linking it closely to DS processes.
IS projects covered in AR publications show evidence of meeting PD characteristics. Salmela
et al. (2000, p. 8), when aiming at “a new control information system (CIS)” experienced that
“the difficulties in improving the CIS reflected a more general political tension”. Vidgen (1997,
p. 39) states that “[n]ew ways of working supported by new technologies were explored in the
future analysis allowing us to envisage different and exciting scenarios” signaling a will to amel-
iorate workers’ daily tasks typical of a PD learning approach. Similarly, Salmela et al. (2000,
p. 8) point to a learning approach that resembles the PD one and claim that “[t]he analyses
made during different planning stages were extensive”. Lindgren et al. (2004, p. 444) stress the
users’ involvement when describing how they “collected information about users’ experiences
with various kinds of IT-based competence management solutions…The data collected through

the technology review and the workshops were discussed and analyzed in collaborative sessions
involving both action researchers and practitioners”. Fruhling and De Vreede (2006, p. 48) also
stress user involvement; “[a]ll team members, developers and users, contributed to the online
workspace”. Braa and Hedberg (2002) refer to core references of the PD literature such as Ehn
and Kyng (1987), and Greenbaum and Kyng (1991), and report on a research process typical of
PD: “This is the start of the period of active prototyping and user interaction, which continues
at the time of writing” (Braa and Hedberg 2002, p. 119).
DS Publications. In the DS publications, we find a number of distinctive quotes serving as evi-

dence for also meeting AR characteristics. Markus et al. (2002, p. 191) follow the “objective of
improving overall organizational effectiveness”, resembling the typical organizational AR objec-
tive. Haynes (2006, p. 333) states that “the Rampart anti-terrorism planning system is the prod-
uct of a three-year collaboration between the United States Marine Corps and a research team
from academia.” In the project reported on by Miller et al. (2006, p. 457), “feedback from the
user study can be incorporated into the next iterative development cycle” pointing to a typical
AR learning approach. In their respective DS studies, Peffers et al. (2006, p. 99) select “30 par-
ticipants from different end-users segments to participate in the study” to show organizational
member participation, and Markus et al. (2002, p. 188) pursued a dialogue through “interviews
with representatives from the four sponsoring companies” and “over an 18-month period, the
development team repeatedly intervened into the organizational design activities of the involved
companies”. The closeness to AR is also obvious when Markus et al. (2002, p. 187-188) state
that “the TOP Modeler project followed this action research strategy ... deploying prototypes
that tested various assumptions about how organizational work is done, observing how users re-
sponded and iterating”. There is reference to the AR research process in Jones and Gregor (2006,
p. 357), who describe “[a]n iterative action research process [that] has been used to evolve the
system and the associated ISDT through three distinct generations over the last ten years”.
DS publications also show evidence of PD characteristics. For example, Markus et al. (2002,
p. 189) reflect that “there might be resistance to using the system by both potential hands-on
users and managers” and that they “learned that the IS design theory…was inapplicable to the
organization design process. As a result we were forced to reconceptualize” (p. 188), pointing to
learning efforts during project conduct. Miller et al. (2006, p. 457-458) find that “it is impera-
tive that we temper our desire to improve the software with the need to maintain consistency for
our users”. Markus et al. (2002, p. 191) quote typical PD work by Greenbaum and Kyng (1991)
in the context of claiming that “the Domain Team’s responsibility was to represent potential us-
ers and review prototypes in an iterative development methodology…using user-centered tech-
niques such as joint design meetings and cooperative prototyping.” Concerning the research
process applied, Peffers et al. (2006, p. 94) stress the planning along stages similar to classic PD
conduct: “The development of the DSRP model included six steps”. Haynes (2006, p. 333)
stress that “the project team carried out requirements elicitation, prototype design reviews, focus
groups, and structured walk-throughs with over 200 people from the Marines”, pointing to both
the research process and the user involvement typical of PD.

Overall, the IS projects covered in the fifteen publications not only meet the characteristics
of ‘their’ method, but to a large extent also meet the characteristics of the other two methods and
thus are potential, but unexploited, knowledge sources (Figure 1).
Projects Publications / People
AR DS AR DS
PD
PD
High similarity Different terminolo-
in research focus, gies, literature and
approach, outcomes, audience
and methods
Figure 1: Summary of observed phenomenon
5 Discussion and conclusions

The analysis contrasts the commonalities of IS research contributions published under PD, AR,
and DS in terms of research focus, outcomes, and research process, with their differences of
terminology, references, and audiences (Figure 1). The notion of distribution is gaining ground
in all three of these domains, but there is, as yet, an insufficient numbers of publications in each
focussing on distribution to allow such an analysis to be performed solely investigating the issue
of distribution. This analysis, however, lays the groundwork for such research.
In common with Fitzgerald and Howcroft (1998), who recognize the strengths within differ-
ent, seemingly competing research traditions, this paper can hold the tension between innovat-
ing at the edge and building robust research streams. Similarly, Baskerville and Pries-Heje (2001)
show that multiple-theory analysis of technology diffusion leads to different but complementary
insights; Fitzgerald (2003, p. 227) appreciate “the idea of competing and alternative research
methods in IS” and Avison (2003, p. 229) suggest that people use different methods “to address
different aspects” of the complex relations between technology and organizations. Hence, the
observed phenomenon may be interpreted as sign of increasing diversity and specialization in
the IS field, reflecting “IS as a fragmented adhocracy” (Banville and Landry 1989). Or, follow-
ing Simon (1969) who stresses the fertility of multiple points of view for a young discipline, the
phenomenon may be seen as having successfully developed separate research sub-communities

in spite of limited differentiation. Such separate sub-communities, also labelled communities of
research (Boehme 1975), scientific communities or intellectual communities (Mulkay 1979),
consist of the practitioners of a scientific specialty (Banville and Landry 1989).
Perhaps PD, AR and DS are better viewed as paradigms rather than methods; a “paradigm...
serves to define what should be studied, what questions should be asked, and what rules should
be followed in interpreting the answers obtained. The paradigm is the broadest unit of consensus
within a science and serves to differentiate one scientific community (or sub-community) from
another” (Ritzer 1975, p. 7). If so viewed, then this paper points to the underlying similarities
exhibited in PD, AR and DS and prompts those researching the development of distributed PD
to look to AR and DS as fruitful areas of inspiration and comparison. For example, the discus-
sion of distributed AR (Adamides and Karacapilidis 2006), that involves AR with distributed
communities of practice may prove useful or distributed DS used for government interactions
(Karacapilidis et al. 2005) or e-business (D’Aubeterre et al. 2008). Information systems re-
search has been criticized for its lack of impact upon practice. This research suggests that there
is a lack of cross-fertilization from approach to approach, and that walled gardens exist or are
starting to emerge. Thus, there is a need for the walled gardens of different methods to at least
have windows so that knowledge can be exchanged and ultimately the walls need to be removed.
However, as demonstrated by this research, the walls are high at present and, though there is
some evidence of cracks in the windows, distributed PD may miss potential sources of enrich-
ment. It may be that some approaches actively seek out external ideas while others are more
insular, or that there is a time dynamic which involves communities building the garden walls
as they develop their approach. Walled gardens may allow research communities to flourish
away from the gaze of the current dominant paradigms and the development of new, special-
ized research outlets can be beneficial, but set against this is the lack of evidence that some such
specialization is special at all and that redundancy fuelled by a lack of participation from other
communities may obtain.
The research reported here is, however, bounded in several ways. First, the sample of papers
is small, though no claim is made here for statistical significance. Rather the paper surfaces and
explores phenomena. However, the quality of the publication outlets suggests that these papers
have been through a rigorous peer review process and the reviewers must have accepted that the
reported research approach was appropriately described. In addition, the sample size of quotes
limits the persuasive power of the findings, even though replication is straight-forward based
on the selected fifteen publications and the method characteristics in Table 1. To substantiate
the findings, future research could extend the sample. Second, using publications as a unit of
analysis only considers the ‘finished product’ and not the process by which it emerged. Authors
may be aware of other literatures when they write but do not choose to reference them, or the
editing and publication process may result in the exclusion of references to some types of litera-
ture. Next, in most instances, following a linguistic approach to the text analysis in addition to
the positivist one, and interpreting the IS projects in the context of the complete publication
text, would have enriched the picture. This could provide further support for, or confound,
the overall finding. Third, one could have also looked at the actual research methods as unit of
analysis, instead of investigating the projects covered in papers published ‘under a method’. For
instance, Cole et al. (2005) investigate differences and similarities between methods and reveal

that AR and DS are remarkably similar and “share important assumptions regarding ontology,
epistemology, and, more importantly, axiology” (p. 332). If one avoids the idiosyncrasy of the
methods and stresses their similarities, an investigation into whether projects fit under more
than one method becomes almost tautological—without easing the underlying issue. Finally,
the research analyzed here may reflect different research scope and intentions though this does
not mean that the output could not be reported under a different label or that cross-fertilization
of ideas is precluded.
6 Notes
1. A slightly different set of iterative steps is provided by Dunne and Martin (2006) who
refer to testing artifacts in context, inducting generalizations, abducting ideas, deducting
consequences, and retesting the ameliorated artifact.
7 References
Adamides, E., and Karacapilidis, N., “A Knowledge Centred Framework for Collaborative Busi-
ness Process Modelling,” Business Process Management Journal, (12:5), 2006, pp. 557-575.
Anderson, W., and Crocca, W., “Engineering Practice and Co-Development of Product Proto-
types,” Communications of the ACM, (36:4), 1993, pp. 49-56.
Argyris, C., and Schön, D., Organizational Learning: A Theory of Action Perspective, Addison-
Wesley, Reading, 1978.
Argyris, C., Putnam, R., and Smith, D., Action science: Concepts, Methods and Skills for Research
and Intervention, Jossey-Bass, San Francisco, 1985.
Arnott, D., “Cognitive biases and decision support systems development: a design science ap-
proach,” Information Systems Journal, (16:1), 2006, pp. 55-78.
Asaro, P., “Transforming society by transforming technology: the science and politics of par-
ticipatory design,” Accounting, Management and Information Technologies, (10:4), 2000, pp.
257-290.
Avison, D., “Is IS an intellectual subject?,” European Journal of Information Systems, (12:3),
2003, pp. 229-230.
Avison, D., Lau, F., Myers, M., and Nielsen, P., “Action research,” Communications of the ACM,
(42:1), 1999, pp. 94-97.
Avison, D., Wood-Harper, A., Vidgen, R., and Wood, J., “A further exploration into infor-
mation systems development: The evolution of Multiview 2,” Information, Technology and
People, (11:2), 1998, pp. 124-139.
Avison, D., and Wood-Harper, A., Multiview: an Exploration in Information Systems Develop-
ment, McGraw-Hill, Maidenhead, UK, 1990.

Banker, R., and Kauffman, R., “The evolution of research on information systems: a fiftieth-
year survey of the literature in Management Science,” Management Science, (50:3), 2004,
pp. 281-298.
Banville, C., and Landry, M., “Can the field of MIS be disciplined?,” Communications of the
ACM, (32:1), 1989, pp. 48-60.
Baskerville, R., “What Design Science is Not,” European Journal of Information Systems, (17:5),
2008, pp. 441-443.
Baskerville, R., and Myers, M., “Special issue on action research in information systems: Making
IS research relevant to practice – Foreword,” MIS Quarterly, (28:3), 2004, pp. 329-335.
Baskerville, R., “Conducting action research: High risk and high reward in theory and practice,”
in Qualitative Research in IS: Issues and Trends, Trauth, E. (ed.), Idea Group Publishing,
Hershey, PA, 2001, pp. 192-218.
Baskerville, R., and Pries-Heje, J., “A multiple-theory analysis of a diffusion of information
technology case,” Information Systems Journal, (11:3), 2001, pp. 182-212.
Baskerville, R., “Investigating information systems with action research,” Communication of the
AIS, (2:19), 1999, pp. 1-32.
Baskerville, R., and Wood-Harper, T., “Diversity in action research,” European Journal of Infor-
Baskerville, R., and Wood-Harper, T., “A critical perspective on action research as a method for
information systems research,” Journal of Information Technology, (11:3), 1996, pp. 235-
246.
Benbasat, I., and Zmud, R., “Empirical Research in Information System: the Practice of Rel-
evance,” MIS Quarterly, (23:1), 1999, pp. 3-16.
Beynon, M., and Chan, Z., “A Conception of Computing Technology Better Suited to Distrib-
uted Participatory Design,” Proceedings of the NordiCHI Conference Workshop, Ghosh, G.,
and Svanaes, D. (eds.), Oslo, Norway, 2006, pp. 50-55.
Bernard, H,. and Ryan, G., “Text analysis: qualitative and quantitative methods,” in Handbook
of Methods in Cultural Anthropology, Bernard, H. (ed.), Alta Mira Press, Lanham, 2000, pp.
595-645.
Blomberg, J., Giacomi, J., Mosher, A., and Swenton-Wall, P., “Ethnographic field methods and
their relation to design,” in Participatory Design: Principles and Practices, Schuler, D., and
Namioka, A. (eds.), Lawrence Erlbaum Associates, Hillsdale, 1993, pp. 123-155.
Bødker, S., “Creating conditions for participation: conflicts and resources in systems develop-
ment,” Human-Computer Interaction, (11:3), 1996, pp. 215-236.
Bødker, S., Grønbaek, K., and Kyng, M., “Cooperative design: techniques and experiences from
the Scandinavian scene,” in Participatory Design: Principles and Practices, Schuler, D., and
Namioka, A. (eds.), Lawrence Erlbaum Associates, Hillsdale, 1993, pp. 157-175.
Boehme, C., “The social function of cognitive structures: A concept of the scientific community
within a theory of action,” in Determinants and Controls of Scientific Development, Knorr,
K., Strasser, H., and Zilian, H. (eds.), D. Reidel, Dordrecht, NL, 1975, pp. 205-225.
Bourdieu, P., Outline of a Theory of Practice, Cambridge University Press, New York, 1977.
Braa, J., and Hedberg, C., “The struggle for district-based health information systems in South
Africa,” Information Society, (18:2), 2002, pp. 113-127.

Cahill, C., “Including Excluded Perspectives in Participatory Action Research,” Design Studies,
(28:3), 2007, pp. 325-340.
Carmel, E., Whitaker, R., and George, J., “PD and Joint Application Design: a transatlantic
comparison,” Communications of the ACM, (36:4), 1993, pp. 40-48.
Checkland, P., Systems Thinking, Systems Practice, Wiley, Chichester, 1981.
Clark, P., Action Research and Organizational Change, Harper and Row Ltd., London, UK,
1972.
Clement, A., and Van den Besselaar, P., “A retrospective look at PD projects,” Communications
of the ACM, (36:4), 1993, pp. 29-37.
Clement, A., “Computing at work: empowering action by ‘low-level users’,” Communications of
the ACM, (37:1), 1994, pp. 52-105.
Clemensen, J., Larsen, S., Kyng, M., and Kirkevold, M., “Participatory Design in health Sci-
ences,” Qualitative Health Research, (17:1), 2007, pp. 122-130.
Cole, R., Purao, S., Rossi, M., and Sein, M., “Being proactive: Where action research meets
design research,” Proceedings of the Twenty-Sixth International Conference on Information Sys-
tems, King, W., and Torkzadeh, R. (eds.), Las Vegas, Nevada, USA, 2005, pp. 325-336.
D’Aubeterre, F., Singh, R., and Iyer, L., “A Semantic Approach to Secure Collaborative Inter-
Organizational eBusiness Processes (SSCIOBP),” Journal of the Association for Information
Systems, (9:3), 2008, pp. 231-266.
Daft, R., and Lewin, A., “Can organization studies begin to break out of the normal science
straitjacket?,” Organization Science, (1:1), 1990, pp. 1-9.
Damodaran, L., “User involvement in the systems design process-a practical guide for users,”
Behaviour and Information Technology, (15:6), 1996, pp. 363-377.
Danielsson, K., Gumm, D., Naghsh, A., and Warr, A., “Distributed Participatory Design,” Ex-
tended abstracts of the Conference on Human Factors in Computing Systems (CHI), Czerwin-
ski, M., and Lund, A. (eds.), Florence, Italy, 2008, pp. 3953-3956.
Davenport, T., and Markus, M. L., “Rigor vs. relevance revisited: response to Benbasat and
Zmud,” MIS Quarterly, (23:1), 1999, pp. 19-23.
Davison, R., Martinsons, M., and Kock, N., “Principles of canonical action research,” Informa-
tion Systems Journal, (14:1), 2004, pp. 65-86.
Deetz, S., “Describing differences in approaches to organization science: Rethinking Burrell and
Morgan and their legacy,” Organization Science, (7:2), 1996, pp. 191-207.
Dennis, A., “Relevance in Information Systems Research,” Communications of the AIS, (6:1),
2001, pp. 1-7.
Dewey, J., Logic: The Theory of Inquiry, Henry Holt and Co., New York, 1938.
Dunne, D., and Martin, R., “Design thinking and how it will change management education:
an interview and discussion,” Academy of Management Learning and Education, (5:4), 2006,
pp. 512-523.
Ehn, P., Work Oriented Design of Computer Artifacts, Swedish Center for Working Life, Stock-
holm, 1988.
Ehn, P., and Kyng, M., “The collective resource approach to system design,” in Computers and
democracy-A Scandinavian challenge, Bjerknes, G., Ehn, P., and Kyng, M. (eds.), Avebury,
Aldershot, 1987, pp. 17-57.

Ehn, P., Kyng, M., and Sundblad, Y., “The UTOPIA project,” in Systems Design for, by, and
with the Users, Briefs, U., and Ciborra, C. (eds.), North-Holland, New York, 1983, pp.
439-450.
Emery, F., and Trist, E., “Socio-technical systems,” in Management Sciences, Models and Tech-
niques, Churchman, C., Verhulst, M. (eds.), Pergamon, Oxford, UK, 1960, pp. 83-97.
Fitzgerald, B., and Howcroft, D., “Towards dissolution of the IS research debate: From Polariza-
tion to Polarity,” Journal of Information Technology, (13:4), 1998, pp. 313-326.
Fitzgerald, G., “Information systems: a subject with a particular perspective, no more, no less,”
European Journal of Information Systems, (12:3), 2003, pp. 225-228.
Floyd, C., Mehl, W., Reisin, F., Schmidt, G., and Wolf, G., “Out of Scandinavia: alternative ap-
proaches to software design and system development,” Human-Computer Interaction, (4:4),
1989, pp. 253-350.
Foros, O., “Price Strategies and Compatibility in Digital Networks,” International Journal of
Economics of Business, (14:1), 2007, pp. 85-97.
Foster, M., “An introduction to the theory and practice of action research in work organiza-
tions,” Human Relations, (25:6), 1972, pp. 529-556.
Foth, M., and Axup, J., “Participatory Design and Action Research: Identical Twins or Syn-
ergetic Pair?,” Participatory Design Conference (PDC), Jacucci, G., and Kensing, F. (eds.),
Trento, Italy, 2006.
Franssila, H., and Pehkonen, M., “Feasibility of Iterative Design as a Challenge in Distributed
Participatory Design Lifecycle,“ Proceedings of the NordiCHI Conference Workshop, Ghosh,
G., and Svanaes, D. (eds.), Oslo, Norway, 2006.
Figueiredo, A.D., and Cunha, P. D., “Action Research and Design in Information Systems: Two
Faces of a Single Coin,” in Information Systems Action Research: An Applied View of Emerging
Concepts and Methods, N. Kock (ed.), Springer, Berlin, 2007, pp.61-96.
Fruhling, A., and De Vreede, G., “Field experiences with eXtreme programming: developing
an emergency response system,” Journal of Management Information Systems, (22:4), 2006,
pp. 39-68
Gadamer, H., Philosophical Hermeneutics, University of California Press, Berkeley, 1976.
Gallivan, M., and Depledge, G., “Trust, control and the role of inter-organizational systems in
electronic partnerships,” Information Systems Journal, (13:2), 2003, pp. 159-190.
Gibson, C., and Zellmer-Bruhn, M., “Metaphors and meaning: an intercultural analysis of the
concept of teamwork,” Administrative Science Quarterly, (46:2), 2001, pp. 274-303.
Greenbaum, J., and Kyng, M. (eds.), Design at Work: Cooperative Design of Computer Systems,
Lawrence Erlbaum, Hillsdale, 1991.
Grønbæk, K., Kyng, M., and Morgensen, P., “CSCW challenges cooperative design in engineer-
ing projects,” Communications of the ACM, (36:6), 1993, pp. 67-77.
Gumm, D., Janneck, M., and Finck, M., “Distributed Participatory Design,” Proceedings of the
NordiCHI Conference Workshop, Ghosh, G., and Svanaes, D. (eds.), Oslo, Norway, 2006.
Gummesson, E., Qualitative Methods in Management Research, Sage Publishing, Thousand
Oaks, 2000.
Haack, S., “The pragmatist theory of truth,” British Journal of Philosophical Science, (27:3),
1976, pp. 231-249.

Haynes, R., “Design knowledge as a learning resource,” First International Conference on Design
Science Research in Information Systems and Technology, Chen, H., Olfman, L., Hevner, A.,
and Chatterjee, S. (eds.), Claremont, CA, 2006, pp. 329-342.
Herzberg, F., Work and the Nature of Man, World Publishing Co, Cleveland, 1966.
Hevner, A., March, S., Park, J., and Ram, S., “Design science in information systems research,”
MIS Quarterly, (28:1), 2004, pp. 75-105.
Hirschheim, R., and Klein, H., “Crisis in the IS field? A critical reflection on the state of the
discipline,” Journal of the Association for Information Systems, (4:5), 2003, pp. 237-293.
Holmstrom, J., “Theorizing in IS Research,” Scandinavian Journal of Information Systems, (17:1),
2005, pp. 167-174.
Howcroft, D., and Wilson, M., “Paradoxes of participatory practices: the Janus role of the sys-
tems developer,” Information and Organization, (13:1), 2003a, pp. 1-24.
Howcroft, D., and Wilson, M., “Participation: ‘bounded freedom’ or hidden constraints on user
involvement,” New Technology, Work and Employment, (18:1), 2003b, pp. 2-19.
Hult, M., and Lennung, S., “Towards a definition of action research: a note and bibliography,”
Journal of Management Studies, (17:2), 1980, pp. 241-250.
Iivari J., “A Paradigmatic Analysis of IS as a Design Science,” Scandinavian Journal of Informa-
tion Systems, (19:2), 2007, pp. 39-64.
Iivari, J., “A paradigmatic analysis of contemporary schools of IS development,” European Jour-
nal of Information Systems, (1:4), 1991, pp. 249-272.
Järvinen, P., “Action Research is Similar to Design Science,” Quality and Quantity, (41:1), 2007,
pp. 37-54.
Järvinen, P., “Action research as an approach in design science,” European Academy of Manage-
ment Conference, Reichwald, R., and Huff, A. (eds.), Munich, Germany, 2005, pp. 1-16.
James, W., The Principles of Psychology, Henry Holt and Co., New York, 1890.
Järvenpaa, S., and Ives, B., “Information technology and corporate strategy: a view from the
top,” Information Systems Research, (1:4), 1990, pp. 351-376.
Jones, D., and Gregor, S., “The formulation of an information system design theory for e-
learning,” First International Conference on Design Science Research in Information Systems
and Technology, Chen, H., Olfman, L., Hevner, A., and Chatterjee, S. (eds.), Claremont,
CA, 2006, pp. 356-373.
Karacapilidis, N., Loukis, E., and Dimopolous, S., “Computer-Supported G2G Collaboration
for Public Policy and Decision Making,” Journal of Enterprise Information Management,
(18:5), 2005, pp. 602-624.
Keen, P., and Scott Morton, M., Decision Support Systems: An Organizational Perspective. Addi-
son-Wesley, Reading, 1978.
Kensing, F., Simonsen, J., and Bødker, K., “MUST: a method for participatory design,” Human-
Computer Interaction, (13:2), 1998a, pp. 167-198.
Kensing, F., Simonsen, J., and Bødker, K., “Participatory design at a radio station,” Computer
Supported Cooperative Work, (7:3-4), 1998b, pp. 243-271.
Klein, H., and Myers, M., “A set of principles for conducting and evaluating interpretive field
studies in information systems,” MIS Quarterly, (23:1), 1999, pp. 67-94.

Kock, N., Gray, P., Hoving, R., Klein, H., Myers, M., and Rockart, J., “IS research relevance
revisited: Subtle accomplishment, unfulfilled promise, or serial hypocrisy,” Communications
of the AIS, (8:23), 2002, pp. 330-346.
Krippendorf, K., Content Analysis: An Introduction to its Methodology, Sage Publications Inc.,
Thousand Oaks, 2004.
Kuhn, S., and Muller, M., “Participatory design (Guest editors foreword),” Communications of
the ACM, (36:4), 1993, pp. 24-28.
Labrogere, P., “Com 2.0: A Path Towards Web Communicating Applications,” Bell Labs Techni-
cal Journal, (13:2), 2008, pp. 19-24.
Lacity, M., and Janson, M., “Understanding qualitative data: a framework of text analysis meth-
ods,” Journal of Management Information Systems, (11:2), 1994, pp. 137-155.
Lee, A., “Action is an artifact: what action research and design science offer to each other,” in
Collaboration in Modern Organizations: Initiating and Managing Distributed Projects, Ad-
vances in e-Collaboration Series, Kock, N. (ed.), Idea Group Publishing, Hershey, 2007, pp.
43-60.
Lee, A., “Rigor and relevance in MIS research: beyond the approach of positivism alone,” MIS
Quarterly, (23:1), 1999, pp. 29-34.
Lewin, K., Resolving Social Conflicts, Harper, New York, NY, 1948.
Lewin, K., “Frontiers in group dynamics II,” Human Relations, (1:2), 1947, pp. 143-153.
Lewin, K., “Action research and minority problems,” Journal of Social Issues, (2:4), 1946, pp.
34-46.
Lindgren, R., Henfridsson, O., and Schultze, U., “Design principles for competence manage-
ment systems: a synthesis of an action research study,” MIS Quarterly, (28:3), 2004, pp.
435-472.
Matz, A., “Scattered Locations and Different Stakeholder Views,“ Proceedings of the NordiCHI
Conference Workshop, Ghosh, G., and Svanaes, D. (eds.), Oslo, Norway, 2006.
March, S., and Smith, G., “Design and natural science research on information technology,”
Decision Support Systems, (15:4), 1995, pp. 251-266.
Markus, L., Majchrzak, A., and Gasser, L., “A design theory for systems that support emergent
knowledge processes,” MIS Quarterly, (26:3), 2002, pp. 179-212.
Maslow, A., Motivations and Personality, Harper and Row, New York, 1954.
Mayo, E., The Human Problems of an Industrial Civilization, MacMillan, New York, 1933.
McKay, J., and Marshall, P., “The dual imperatives of action research,” Information Technology
and People, (14:1), 2001, pp. 46-59.
Mead, G., “The social self,” Journal of Philosophy, Psychology, and Scientific Methods, (10:14),
1913, pp. 374-380.
Miller, T., Leroy, G., Huang, J., Chuang, S., and Charlop-Christie, M., “Using a digital library
of images for communications: comparison of a card-based system to PDA software,” First
International Conference on Design Science Research in Information Systems and Technology,
Chen, H., Olfman, L., Hevner, A., and Chatterjee, S. (eds.), Claremont, CA, 2006, pp.
454-460.
Mulkay, M., Science and the Sociology of Knowledge, George Allen and Unwin, London, UK,
1979.

Mumford, E., “The story of socio-technical design: Reflections on its successes, failures and
potential,” Information Systems Journal, (16:4), 2006, pp. 317-342.
Munir, K., and Phillips, N., “The birth of the ‘Kodak moment’: institutional entrepreneurship
and the adoption of new technologies,” Organization Studies, (26:11), 2005, pp. 1665-
1687.
Naghsh, A., Danielsson, K., Fischer, G., Bratteteig, T., Blomberg, J., and Nocera, J., “Distribut-
ed-PD: Challenges and Opportunities,” Proceedings of the NordiCHI Conference Workshop,
Ghosh, G., and Svanaes, D. (eds.), Oslo, Norway, 2006.
Niehaves, B., “On Epistemological Pluralism in Design Science,” Scandinavian Journal of Infor-
Nørreklit, H., “The balanced scorecard: what is the score? A rhetorical analysis of the balanced
scorecard,” Accounting, Organizations and Society, (28:6), 2003, pp. 591-619.
Nunamaker, J., Chen, M., and Purdin, T., “Systems development in information systems re-
search,” Information Systems Research, (7:3), 1991, pp. 89-106.
O’Connor, E., “The politics of management thought: A case study of Harvard Business School
and the Human Relation school,” Academy of Management Review, (24:1), 1999, pp. 117-
131.
O’Connor, E., “Paradoxes of participation: Textual analysis and organizational change,” Organi-
zation Studies, (16:5), 1995, pp. 769-803.
Orlikowski, W., and Barley, S., “Technology and institutions: what can research on information
technology and research on organizations learn from each other?,” MIS Quarterly, (25:2),
2001, pp. 145-165.
Orlikowski, W., and Iacono, C., “Research commentary: desperately seeking the ‘IT’ in IT
research-a call to theorizing the IT artifact,” Information Systems Research, (12:2), 2001, pp.
121-134.
Orlikowski, W., and Baroudi, J., “Studying information technology in organizations: research
approaches and assumptions,” Information Systems Research, (2:1), 1991, pp. 1-28.
Peffers, K., Tuunanen, T., Gengler, C., Rossi, M., Hui, W., Virtanen, V., and Bragge, J., “The
design science research process: a model for producing and presenting information systems
research,” First International Conference on Design Science Research in Information Systems
and Technology, Chen, H., Olfman, L., Hevner, A., and Chatterjee, S. (eds.), Claremont,
CA, 2006, pp. 83-106.
Prasad, P., and Prasad, A., “Stretching the iron cage: the constitution and implications of routine
workplace resistance,” Organization Science, (11:4), 2000, pp. 387-403.
Rapoport, R., “Three dilemmas in action research,” Human Relations, (23:4), 1970, pp. 499-
513.
Ritzer, G., Sociology: A Multiple Paradigm Science, Allyn and Bacon, Boston, MA, 1975.
Salmela, H., Lederer, A., and Reponen, T., “Information systems planning in a turbulent envi-
ronment,” European Journal of Information Systems, (9:1), 2000, pp. 3-15.
Scarbrough, H., Robertson, M., and Swan, J., “Professional media and management fashion:
The case of knowledge management,” Scandinavian Journal of Management, (21:2), 2005,
pp. 197-208.
Sein, M., Rossi, M., and Purao, S., “Exploring the Limits of the Possible,” Scandinavian Journal
of Information Systems, (19:2), 2005, pp. 105-110.

Simon, H., The Sciences of the Artificial, MIT Press, Cambridge, MA, 1969.
Spinuzzi, C., “The methodology of participatory design,” Technical Communication Quarterly,
(52:2), 2005a, pp. 163-174.
Spinuzzi, C., “Lost in the translation: shifting claims in the migration of a research technique,”
Technical Communication Quarterly, (14:4), 2005b, pp. 411-446.
Susman, G., and Evered, R., “An assessment of the merits of action research,” Administrative
Science Quarterly, (23:4), 1978, pp. 582-603.
Vaara, E., and Tienari, J., “Justification, legitimization and naturalization of mergers and acquisi-
tions: A critical discourse analysis of media texts,” Organization, (9:2), 2002, pp. 275-304.
Van de Ven, A., and Johnson, P., “Knowledge for theory and practice,” Academy of Management
Review, (31:4), 2006, pp. 802-821.
Vidgen, R., “Constructing a web information system development methodology,” Information
Systems Journal, (12:3), 2002, pp. 247-261.
Vidgen, R., “Stakeholders, soft systems and technology: Separation and mediation in the analysis
of information system requirements,” Information Systems Journal, (7:1), 1997, pp. 21-46.
Walls, J., Widmeyer, G., and Sawy, O., “Building an information system design theory for vigi-
lant EIS,” Information Systems Research, (3:1), 1992, pp. 36-59.
Warr, A., “Situated and Distributed Design,” Proceedings of the NordiCHI Conference Workshop,
Ghosh, G., and Svanaes, D. (eds.), Oslo, Norway, 2006, pp. 1-3.
Williams K., Chatterjee S., and Rossi M., “Design of Emerging Digital Services: A Taxonomy,”
European Journal of Information Systems, (17:5), 2008, pp. 505-517.
Winter R., “Design Science Research in Europe,” European Journal of Information Systems,
(17:5), 2008, pp. 470-475.

Special Issue On Distributed Participatory Design, SJIS 2009, Issue 1

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Special Issue On Distributed Participatory Design, SJIS 2009, Issue 1

Încărcat de

Drepturi de autor:

Formate disponibile

C o n t en t s :

Scandinavian Journal of Information Systems 2009 Vol. 21, No. 1

Consumers' Views on Privacy in E-Commerce 3

A Special Issue Editorial 23

Distributed Development to Enable User Participation: Multilevel design in the 27

Inter-Contextual Distributed Participatory Design: Communicating design 51

Furthering Distributed Participative Design: Unlocking the walled 77

Publication date: 30th June, 2009.

Electronic journal: www.e-sjis.org; available free of charge.

Bjørn Erik Munkvold

© Scandinavian Journal of Information Systems, 2009, 21(1), 1–2

Key words: information privacy, consumer, e-commerce, phenomenography.

© Scandinavian Journal of Information Systems, 2009, 21(1), 3–22

4 • Kaapu & Tiainen

Consumers’ Views on Privacy in E-Commerce • 5

6 • Kaapu & Tiainen

Areas of information privacy studies Example references

Table 1: Focal areas of information privacy studies

Consumers’ Views on Privacy in E-Commerce • 7

8 • Kaapu & Tiainen

Interviewees Female Male Total

Table 2: Consumers interviewed

Consumers’ Views on Privacy in E-Commerce • 9

4 Result: Categorization of consumers’ views

Table 3: Summary of the categorization of consumers’ information privacy conceptions in

10 • Kaapu & Tiainen

4.1 Referential objects of information privacy

Consumers’ Views on Privacy in E-Commerce • 11

Table 4: The contents of the conceptions illustrated with interview examples

12 • Kaapu & Tiainen

Consumers’ Views on Privacy in E-Commerce • 13

4.2 Structural components of information privacy

14 • Kaapu & Tiainen

Component 2: Technology. In the second component, privacy is understood through technolo-

Consumers’ Views on Privacy in E-Commerce • 15

16 • Kaapu & Tiainen

Consumers’ Views on Privacy in E-Commerce • 17

18 • Kaapu & Tiainen

Consumers’ Views on Privacy in E-Commerce • 19

20 • Kaapu & Tiainen

Consumers’ Views on Privacy in E-Commerce • 21

22 • Kaapu & Tiainen

Karin Danielsson Öberg

© Scandinavian Journal of Information Systems, 2009, 21(1), 23–26

24 • A Special Issue Editorial

A Special Issue Editorial • 25

26 • A Special Issue Editorial

Ola Hodne Titlestad

Key words: Distributed participatory design, boundary spanning, scaffolding, innofusion,

© Scandinavian Journal of Information Systems, 2009, 21(1), 27–50

28 • Titlestad, Staring & Braa

Distributed Development to Enable User Participation • 29

2.1 DPD, DSD and standards

1 Have Clear Distribution Rationale 6 Manage Processes

Table 1: Ten strategies for DSD, from Lings et al. (2006)

The distribution of developers geographically introduces problems of mutual understanding

30 • Titlestad, Staring & Braa

2.2 PD in the Global South and learning

Distributed Development to Enable User Participation • 31

2.3 Boundary spanning and scaffolding

32 • Titlestad, Staring & Braa

Distributed Development to Enable User Participation • 33

Academic Theses, dissertations, papers, conference presentations

  1: The CommSy development process

3 Bringing them all together: techniques for inter-

Figure  2: Workshop participants discussing mindmaps for future development.