The University of Sheffield CITY Liberal Studies

MSc in Information Systems DISSERTATION

FREE/OPEN SOURCE SOFTWARE AND BUSINESS: FORMING A

MIGRATION POLICY

This report is submitted in partial fulfillment of the requirement for the degree of Master of Science in Information Systems by

Kristijan Spirov

August 2006

Approved

Dr. Iraklis Paraskakis

1

Abstract

This paper explores the non-technical issues that arise with migration to

Free/Libre and Open Source Software (FLOSS) in Large Scaled Enterprises (LSEs),

with special emphasis on human and social issues. The result of the study is a

conceptual framework for effective addressing of non-technical issues in a migration

policy. The study is of exploratory nature and uses secondary research findings from

multiple disciplines: free and open source software, management of human

resources, power and conflict, system migration, information systems development,

motivation, change management, organizational psychology and social informatics;

including anecdotal evidence from FLOSS migrations in practice. A holistic

approach has been used during the analysis of the migration issues, based on the key

principles of social informatics that IT technology and social contexts are

interdependent. The relationship between FLOSS migrations and different

organizational contexts is elaborated and it has been found out how lack of vendor

support for FLOSS motivates LSEs to improve their self-support of own IT

infrastructures. In-depth analysis is performed of the individual and social impact

that FLOSS migrations have on enterprise IT personnel, as well as how different

individual factors such as age, gender, personal motivations and beliefs may impact

on individual migration efforts. The social differences between typical enterprise IT

personnel and the FLOSS community are elaborated, as well as how to overcome the

communication gap between them. The role of end users in FLOSS migration and

relevant issues are discussed. It is proved why staged migration is the most viable

option and it is shown how it is related to human and social factors. At the end, a

conceptual framework for description of non-technical FLOSS migration issues is

discussed, which can be used as a helpful tool in a migration policy.

2

Table of contents

Table of figures ............................................................................................................ 4

Chapter 1. Introduction ................................................................................................ 5

1.1. Free/Libre and Open Source Software (FLOSS) .............................................. 5

1.2. Factors of FLOSS adoption among enterprises ................................................ 7

1.3. Large Scaled Enterprises (LSEs) ...................................................................... 9

1.4. FLOSS in European LSEs............................................................................... 11

1.5. Aims and objectives ........................................................................................ 12

1.6. Research questions .......................................................................................... 13

1.7. Structure of the dissertation ............................................................................ 13

Chapter 2. An insight into migration to FLOSS in organizational context of LSEs.. 15

2.1. Importance of organizational contexts in which migrations take place.......... 15

2.2. Attributes of FLOSS migrations ..................................................................... 18

2.3. The community perspective ............................................................................ 20

2.4. Importance of addressing non-technical issues............................................... 22

2.5. Summary ......................................................................................................... 23

Chapter 3. Methodological issues regarding the study .............................................. 24

3.1. Issues related to research of FLOSS migrations ............................................. 24

3.2. Description of the research methodology ....................................................... 26

3.3. Summary ......................................................................................................... 29

Chapter 4. Critical evaluation of non-technical FLOSS migration issues ................. 30

4.1. Challenges and opportunities for IT departments ........................................... 31

4.1.1. The impact of different vendor support for FLOSS and COTS............... 31

4.1.2. Impact of FLOSS on the internal culture ................................................. 34

4.1.3. Coping with differences between IT personnel ........................................ 36

4.1.4. Influence of personal motivation.............................................................. 39

4.1.5. Influence of age and gender factors......................................................... 42

4.1.6. Improvement of skills and knowledge sharing ......................................... 45

4.1.7. Approaching the FLOSS community through acceptance of new

collaborative practices....................................................................................... 47

4.1.8. Issues with personal and shared ideology and beliefs ............................. 51

3

4.1.9. Issues with online communication and usage of online resources .......... 54

4.1.10. Process of changing the internal culture ............................................... 59

4.1.11. Establishing bonds with the community and self identity of the IT

personnel ............................................................................................................ 64

4.2. End users and FLOSS migration..................................................................... 68

4.2.1. Role and profile of end users ................................................................... 68

4.2.2. Importance of user training ..................................................................... 70

4.2.3. Challenges of end users participation...................................................... 72

4.2.4. Impact of conflicts and internal power distribution on the migration ..... 74

4.2.5. Resistance to change ................................................................................ 77

4.3. FLOSS migration steps ................................................................................... 79

4.3.1. The benefits of staged migration .............................................................. 79

4.3.2. The impact of staged migration on end users .......................................... 82

4.3.3. End users and desktop migration............................................................. 84

4.4. Summary ......................................................................................................... 87

Chapter 5. Conclusion................................................................................................ 90

5.1. Discussion of the findings............................................................................... 90

5.2. Recommendations for future research ............................................................ 94

References .................................................................................................................. 96

4

Table of figures

Figure 1: The FLOSS community triangle .................................................................. 5

Figure 2: OSS project community................................................................................ 9

Figure 3: The method of alternating point of view in exploratory research .............. 27

Figure 4: Support options for FLOSS ........................................................................ 33

Figure 5: Communities built around common interests............................................. 34

Figure 6: Age of FLOSS developers.......................................................................... 43

Figure 7: Reasons to join and stay in the FLOSS community ................................... 45

Figure 8: Unofficial experimentation with Linux in enterprises................................ 52

Figure 9: The classification of FLOSS users and developers .................................... 57

Figure 10: Employment status of FLOSS developers................................................ 59

Figure 11: Comparison between traditional and emerging organizational systems .. 61

Figure 12: Example of community structure of a big FLOSS project ....................... 62

Figure 13: Example of social network of FLOSS developers ................................... 62

Figure 14: The layered behavioural model of software development........................ 65

Figure 15: Single community built around source code ............................................ 66

Figure 16: Knowledge domains involved in system building.................................... 77

Figure 17: Impact of changes..................................................................................... 78

Figure 18: Example of a Linux adoption roadmap .................................................... 81

Figure 19: Survey on Linux usability: Level of enjoyment ....................................... 86

Figure 20: Survey on Linux usability: Time needed for achieving competence ....... 86

Figure 21: Influence of extrinsic and intrinsic factors on migration outcomes ......... 91

5

Chapter 1. Introduction

1.1. Free/Libre and Open Source Software (FLOSS)

This is a brief introduction to the concept of free and open source software,

its historical perspective and the international organizations which support it.

Sharing source code between software developers is not a new concept and it

exists since the invention of software and programming languages. When computers

were still a new technology and were used merely as research tools, people and

organizations started to share software source code for practical reasons (Brunelle

and Bruce, 2002). Ever since that period the practical side of this phenomenon has

been the exchange of different views and ideas for the purpose of more efficient

problem solving, while its underlying generator have been the social factors which

have sustained the gradual progress towards community formation (Figure 1). As

Stallman (1999) explains, the software-sharing community existed for many years.

But, the development of the software industry, thus commercialization of software,

has contributed the social aspect of software development to become tightly

controlled within closer boundaries, on institutional level.

Figure 1: The FLOSS community triangle

(Source: Bergquist and Ljungberg, 2001; p.2)

6

When computers reached the business world (Perens, 1999; Weerawarana

and Weeratunge, 2003), the rights to software became restricted and fees were

charged for each copy. In order to protect their investment and income, enterprises

have been forbidding sharing of source code with the outside world. As a reaction,

this has triggered the appearance of a large community of developers that support the

concept of free exchange of ideas for software development and source code, like in

the old times when computers were used as research tools and software was freely

shared (Perens, 1999).

There are two major movements within the software-sharing community: one

supports the “free software” and the other one supports the “open source software”

concept. The first is represented by the Free Software Foundation

(http://www.fsf.org), an organization which defends the principle that users should

be given a complete freedom and rights to exchange or adapt any software code,

without imposing any restrictions on these rights. The supporters of “free software”

insist that all software code should be freely available and users should be given full

rights to use, adapt, change and distribute it. The movement represented by the Open

Source Initiative Organization (http://www.osi.org) supports the “open source

software” concept. This concept is more flexible from a business point of view than

“free software” because it allows different models of using freely open source code

which may include some conceptual restrictions for users. For example, open source

software can be used in combination with commercial software. Although the

supporters of “open source” strongly support the above stated freedoms, they do not

insist that all software code should be freely open for the users, thus express higher

consideration for the investments made by the software creators. Elliott and Scacchi

(2003a) show how FLOSS developers actually split into these two subcultures.

Several abbreviations have been used to describe free and open source

software concepts. The European community has widely accepted the term FLOSS,

where “L” stands for “Libre” and symbolically denotes the freedom. Even though the

concepts of “free” and “open-source” software obviously have intrinsic differences,

hereinafter they will be discussed as a single concept under the term FLOSS, because

they basically share the same ideas that it should be allowed to freely exchange

7

source code. In contrast, the term “commercial software” will be referred to further

in the text as COTS (abbreviation of Commercial Off-The-Shelf).

In the above text it was presented that the FLOSS community has a long

history and has eventually split into two major movements: free and open source.

Within this study they are referred to as FLOSS, abbreviation that is commonly used

in EU. The difference between the two movements is not relevant to this study.

1.2. Factors of FLOSS adoption among enterprises

A brief discussion follows about the factors that influence the adoption of

FLOSS among business enterprises.

At present, FLOSS is an emerging software concept driven by the rapid

progress of information technology, especially by the development of open standards

and Internet as a global distribution and marketing channel. By enriching the

software market with new products, FLOSS contributes to the commoditization of

software (Koch, 2003), a process which inherently pulls down the market prices in

the software industry. Lower cost of software is also one side effect of the unique

licensing rights given to the users of FLOSS to freely use, copy, adapt and distribute

changes of the software without any restrictions. The growing popularity of FLOSS

brings many challenges to free enterprises.

Cost is considered (Dedrick and West, 2004; Dedrick et al, cited in Holck et

al, 2004; Wheeler, 2002; QNB Intelligence, 2004; The Dravis Group, 2003) as one of

the most important drivers for FLOSS adoption among enterprises. However, Ghosh

(2003) and Meehan (2005) found out that enterprises consider other reasons for

migration to FLOSS as even more important than cost, like stability and security.

These attributes of FLOSS are another side effect from the freedom given to software

users, because the public availability of the source code for review provides high

probability of maintaining high quality of code in general. That is perceived as a

8

huge incentive for migration to FLOSS by enterprises. IT market research (Evans

Data, cited in Kuchinskas, 2005) also confirms the finding that lower cost is an

important, but not a deciding factor for adoption of FLOSS among enterprises.

The previous findings actually describe a push-pull system established

between the FLOSS concept on one side and what enterprises need on the other. This

mechanism strengthens the foundation for higher adoption of FLOSS inside the

business world, which in feedback generates new development initiatives and further

emergence of the concept. It may also be one of the key reasons why more and more

IT vendors enrol in these kinds of projects. In other words, FLOSS is becoming a

well-grounded choice for enterprise software users and that also drives more vendors

to support it.

On the other side, there are various barriers for migration (Dedrick et al, cited

in Holck et al, 2004) like: lack of compatibility with existing technologies, skills and

organizational resources, low availability of external technological resources, as well

as potential legal barriers. The existence (Barton and Nissanka, 2001) of many types

of open source licenses and complicated requirements related to them exposes the

potential of FLOSS to cause difficulties. In contrast to this, through a comparative

study of EULA (used by Microsoft and other vendors) and GPL (used for most

FLOSS) licensing models, Zymaris (2003) points out that even in case of commercial

software the user is not legally protected, which is opposite of what many

commercial software vendors claim. The Swedish Agency for Public Management

Publication Service (2003) also emphasizes that most user licenses for proprietary

software products exclude responsibility for flaws.

To summarize, the adoption of FLOSS is emerging among enterprises, and

according to some cost may not be the primary factor for this. Still, there are many

different barriers for the adoption process, such as technological, human, legal, etc.,

that enterprises have to overcome.

9

1.3. Large Scaled Enterprises (LSEs)

The scope of this study are large enterprises that migrate to FLOSS systems.

In this section it is discussed why large enterprises are important for the IT industry

and how the size of the enterprise may affect the decision for FLOSS migration.

Large scaled enterprises are very important to the IT industry, because they

are a segment that invests a lot in technology. The level of adoption of forthcoming

IT trends among large enterprises usually gives course to the overall IT industry.

According to Gartner (2003), the strength of this sector is crucial for sustaining

growth in demand for IT products and services. Spanos et al (2001) point out that

enterprise size is an important factor in the adoption of information technology,

because large firms are able to allocate considerable resources on new technologies.

In that sense, it is important to understand how large enterprises deal with FLOSS,

because it is an emerging concept which gradually changes the shape of the IT

market, by introducing a different viewpoint on how software should be developed.

Adoption of FLOSS among large enterprises has always had strong influence on the

public opinion and the predictions for the future of FLOSS. (Figure 2)

Figure 2: OSS project community

(Source: Seidel and Niedermeier, 2003, p.8)

10

Large enterprises are traditionally organized with a strong vertical hierarchy

(Sharma et al, 2002; Dinkelacker and Garg, 2001; Galup et al, 1998). The size and

complexity of a large enterprise simply dictates the need for granulation of the

organizational scheme. The bigger the enterprise is, the more obvious is the difficulty

to flatten the organizational layers. Large enterprises usually have many departments,

each managed by a single manager who either reports directly to the CEO, or maybe,

in case of a very big enterprise, to a senior middle manager that is responsible for

several departments. Large Scaled Enterprises (LSEs) typically have their own

internal IT personnel that provide the following functions: in-house software

development, system administration and support. In contrast to this, Small and

Medium Enterprises (SMEs) hardly have IT teams with all previously defined

functions, due to their size.

The role of the IT department in this traditional setting of a LSE is to support

the business activities of the enterprise and the employees from the other

departments are end users for the IT department. In most large enterprises whose

primary business is not IT related, the IT department is expected to support the

business and follow an already established company strategy, instead of being a

crucial player in creating one. In some top positioned enterprises, IT may have

stronger role in the development of a business strategy for reaching competitive

advantage, but usually most IT departments have been facing problems of alignment

with the business and predominantly cope with every-day operational issues.

When a large enterprise decides to migrate to FLOSS, it should expect a

number of complex issues. Most of them a large enterprise will face primarily due to

its size. These issues can be of technological or non-technological nature

(organizational, human, social) and can influence on the planned process of change.

But migration to FLOSS may be different from other migrations, especially because

of the capability of introducing new paradigms in the enterprise. There is an apparent

need for an enterprise migration policy for FLOSS, in order to guide transition, but

also mitigate potential risks. A migration policy should cover not only technical

aspects of the migration, but also organizational, human and social aspects in order to

prevent from project failures and potential future disappointments from FLOSS

within the enterprise.

11

To summarize this section, it was explained why large enterprises are

important for the IT industry, that they typically have complicated organizational

structures and use internal IT personnel to support deployment of information

technology. It was discussed that there might be specific migration issues due to

enterprise size, but also that FLOSS might create new paradigms.

1.4. FLOSS in European LSEs

This section briefly presents how LSEs are defined in Europe and some

statistical data in order to provide a clear picture about the actual size of the LSE

sector. The current state of FLOSS adoption in Europe is also presented.

LSEs are defined by the European Commission (recommendation 96/280/EC)

as independent enterprises that have more than 250 employees and annual turnover

over �40 million. According to the European Observatory for SMEs (2000), there are

around 40,000 LSEs in Europe, which represents only 0.2% of the total number of

European enterprises. Only around 2,000 of these enterprises are located in the non-

EU countries, a fact which is closely related to the economic differences between EU

and the rest of the countries. Furthermore, an average LSE in the EU has 1010

employees, while in the non-EU countries their average size is around 780

employees. The average size of the IT team in LSEs is around 10-40 people.

Empirical studies of large enterprises show (e-Business Watch, 2002b) that the size

of IT departments may dramatically vary between business sectors.

There is extensive experience in the use of FLOSS in the public sector in

Europe (The Swedish Agency for Public Management Publication Service, 2003).

This means that there is support from national governments. This motivates the

adoption of FLOSS among enterprises. FLOSS has long tradition in Europe. For

example, Dempsey et al (2002) conclude that the Linux project has deep European

roots. According to the research of International Institute of Infonomics (2002), the

12

numbers of companies that use FLOSS are not the same in different EU regions and

that provides some information about the actual popularity of FLOSS within these

establishments, but also about the maturity of the software markets.

In some EU countries (International Institute of Infonomics, 2002; QNB

Intelligence, 2004), like Germany for example, the adoption rate of FLOSS in LSEs

is very high (around half of all large enterprises have performed formal evaluation of

some FLOSS products), while in others, it is substantially lower. This may be a result

of cultural differences. The adoption of FLOSS in LSEs in the non-EU regions is

lower than in the EU. Cosovanu (2003) explores the reasons behind this issue in the

peculiarities of the local software markets in Eastern Europe, which still experience a

lot of fear and inertia within the enterprises.

To recapitulate the above, the number of LSEs is very low compared to the

total number of enterprises and there is a big difference between the size of the LSE

sector in the EU and non-EU countries. Regarding FLOSS, it has higher level of

adoption among EU countries. This may be partially due to the strong support for

FLOSS from national governments in EU.

1.5. Aims and objectives

The aims of this study are to identify and elaborate the non technical issues

that are experienced by large commercial enterprises (LSEs) when they carry out

partial or full migrations to FLOSS, as well as to discuss the implications derived

from them. The primary focus of the study is on identification of human and social

issues with such migrations, through appropriate reflection on the organizational

contexts of large enterprises in general.

In pursuit of the aims of this study, the following objectives are set: to

identify the FLOSS migration issues that are specifically related to IT personnel in

large enterprises and also those that are common for end users. Other objectives are

13

to explore the dependency between various human and social issues and the level of

FLOSS migration (partial or full), as well as to explore and discuss the advantages of

a staged migration in favour of a full roll-out to FLOSS systems.

1.6. Research questions

The main focus of the study are the following research questions that address

human and social issues with FLOSS migrations in LSEs:

� What are the human and social issues with migration to FLOSS in LSEs?

� Are these issues different from those in other types of migration projects?

� Do these issues vary across different types of organizations?

� What best practices can be defined for dealing with human and social

issues with migration to FLOSS in LSEs?

� Can FLOSS change the corporate culture in LSEs?

The main benefit of this study, through detailed discussion and analysis in

relation to the above research questions, is expected to be a better understanding of

the human and social issues that are common during migrations to FLOSS in LSEs,

with purpose of developing an outline of a conceptual framework for dealing with

those issues as a tool for addressing them effectively in a corporate FLOSS migration

policy.

1.7. Structure of the dissertation

The content of this paper is organized with the following structure: Chapter 1

is an introductory chapter to the topic. Basic terms are defined and brief history of

FLOSS is presented in section 1.1, while the adoption of FLOSS among the business

14

is described in 1.2 In 1.3 the main aspects of large enterprises are introduced and the

situation with large enterprises in Europe is briefly discussed later in 1.4 The aims

and objectives of the study are revealed in sections 1.5 and 1.6 respectively. Chapter

2 presents an insight into how organizational context of LSE relates to FLOSS

migration. In 2.1 it is discussed why the context is important for the analysis, in 2.2 it

is discussed what distinguishes FLOSS migration from other kinds of migrations, in

2.3 it is explained why the community perspective is important for LSEs, while in

2.4 the importance of non-technical issues with FLOSS migrations is stressed.

Chapter 3 presents the research methodology which is used for finding the relevant

answers to the research questions. Section 3.1 reveals the gap in existing academic

knowledge which also defined the problem domain of the study, while 3.2 presents

and justifies the chosen research methodology. Chapter 4 deals with the critical

evaluation of non-technical FLOSS migration issues. The analysis performed in this

chapter is divided into three main parts. In section 4.1 the human and social issues

faced by the IT department with FLOSS migration are discussed. Both individual and

social aspects are analyzed in this section. The role of the end users in FLOSS

migration and the issues related to them are discussed in section 4.2 Separate

analysis of the human and social issues that urge the need for staged migration, is

performed in section 4.3 After the in-depth analysis of the human and social issues,

finally in Chapter 5 in section 5.1, the conclusions of this paper are presented and

answers provided to the research questions of the study, while in 5.2

recommendations are given for future research.

15

Chapter 2. An insight into migration to FLOSS in

organizational context of LSEs

This chapter provides an insight into FLOSS migrations within a broader

organizational context of LSE. In the first section it is discussed how the

organizational context of LSE is related to the different issues with migration

projects, how organizations from inside perceive the role of internal IT personnel and

what are the typical problems with human resources with these kinds of projects in

LSEs. In the next section it is elaborated what makes FLOSS migration similar or

different from other kinds of system migrations, regarding the human and social

issues. It is discussed about various approaches of LSEs towards the perspective of

belonging to a wider community and why this community perspective is important

for FLOSS migrations. In the last section a discussion is made about the importance

of addressing human and social issues with FLOSS migrations in LSEs.

2.1. Importance of organizational contexts in which

migrations take place

In the text that follows it is discussed how the organizational context of LSE

is related to the issues with migration projects. It is presented how IT is perceived

internally in LSEs and what kinds of problems with human resources are typical for

these kinds of projects.

When a large-scale IT project like FLOSS migration is undertaken in a large

enterprise, it usually affects many segments of the enterprise, especially the core

departments which are perceived from inside as the main generators of profit. The

active involvement of end users in IT projects, especially from core departments, is

16

very important for avoiding potential failures and disappointments. But for end users

in LSEs, IT is just a service which provides the tools to do the job or improve

efficiency. The major expectation (Hirschheim and Lacity, 2000; Sumner and

Klepper, 1987) that users have for IT performance is high service quality and this

imposes external pressure to the IT personnel. In organizations where IT is involved

in making company-wide strategic decisions, this pressure is even higher.

Employees of a large organization share the organizational culture in the

environment they work in, but also have personal motives. Differences between

motives may induce conflicts between IT and non-IT personnel inside the

organization during the FLOSS migration project, because non-IT personnel could

treat the migration as a trouble that nobody needs at all. The experience of managers

in LSEs (Dedrick and West, 2003; McHugh and Donnelly, 2004; Parker, 2000) is

that users do not have the knowledge and do not care. Non-savvy users (Vaden,

2003) may need an extra effort to justify the migration. Things can become more

complicated if the relationship between the IT department and the business units has

been prone to conflicts, especially if such conflicts have longer history, spanning

over more than one generation of managers.

If a decision for migration to FLOSS is put in a general context of a large

enterprise, the whole migration effort may seem to be devalued compared to the high

business strategic enterprise goals and objectives. As it was previously described, the

decision to migrate to FLOSS will typically not have the same meaning for end users

and IT personnel. While for IT personnel the migration may mean professional

challenge or chance for improvement of professional careers, for end users it may

mean unnecessary trouble. Or even worse may happen, the management to see it as

an opportunity, while to be perceived as a cumbersome, not needed change by the IT

personnel. But regardless of the specifics of the particular setting, it is very important

that if a LSE decides to migrate its information system on FLOSS platform,

ultimately this will be done only with a management decision.

FLOSS can not be adopted at any significant scale (Holck et al, 2004; Larsen

et al, 2000) if it conceptually does not fit into the IT strategic framework that has

been set up by the top management, except maybe in certain areas with low

17

visibility, like server platforms. All individual initiatives, even those engaged in

shadow, regardless of the level of influence, cannot force major change of this kind

because of lack of authority. Major change is possible only if the management

decides to make it. Furthermore, the motives for FLOSS migration of management

and IT personnel could be very different, especially if the migration initiative did not

originate from IT personnel. And some individuals may favour FLOSS simply

because of their own beliefs that software should be free. These differences should

be treated as risk factors during migration projects. In fact, any serious dissents

among the IT personnel could potentially cause disruption of migration projects, or

even failure.

From a technical perspective, migration is the actual conversion of a

computer system from one state to another. From a holistic perspective, migration is

represented with a set of procedures and steps that are aimed to navigate the safe

change of all constituting parts of the information system of the enterprise from the

commencing state to the desired one, i.e. not only computer systems, but also human

resources and organizational frameworks. The migration should be well defined with

a migration policy that should cover not only technical issues, anticipate potential

problems and offer practical solutions for resolution of such problems.

To summarize the above, there are differences between how IT personnel and

end users view IT projects, since their motives are distinct. Since migrations have

power to change people and organization as well, it is very important to understand

the organizational contexts in which projects are performed and to properly address

the human and social issues in a migration policy.

18

2.2. Attributes of FLOSS migrations

The following question can be framed regarding FLOSS migration: what is

similar and what is different between FLOSS migration and other kinds of system

migrations, concerning the associated human and social factors? This is discussed in

the paragraphs that follow.

As with other types of migrations, migration to FLOSS is not an easy process

and there are many hidden obstacles and risks on the way. Internal users may face

many difficulties with migration. Kleintop et al (1994) point out that users will better

accept a technology if there is: full support from top management for the usage of

technology; training; open communication with employees and their involvement in

the decision making process. Yet participation alone (Butler and Fitzgerald, 1997)

does not guarantee that the information system will be efficiently operated and used,

because there is a tendency in organizations to view the development process as a

mechanism for resolution of internal power battles. Therefore it is very important to

provide a trustful organizational framework.

The inter-group conflict (Wynekoop and Walz, 1999) which is proportional

to the knowledge and expectation gaps between internal groups may be suppressed

through improvement of communication. Though it has to be considered that (Nilsen

and Sein, 2004) neither knowledge nor communication helps when the support

function is not properly organised and more importantly, when users are not

informed about how it is organised. If the organization (Holck et al, 2004) has little

competence in FLOSS, the migration will be perceived as a high risk operation with

little or no support from the IT department. This could produce conflicting situations

between IT and their business counterparts, or activate dangerous residues if such

conflict existed in the past.

One of the difficult problems that an IT manager may face in a LSE that

decides to migrate to FLOSS, is that not everybody would necessarily agree with the

FLOSS concept at the first place. Even though programming is (Nuvolari, 2003) a

19

creative problem-solving activity, there are cultural differences between IT

professionals in how computer playfulness and personal innovativeness are

expressed (Fardal and Tollefsen, 2004). Therefore, while other types of software

migration would be expected to cause shallower distress to an IT professional of

more temporary character, inflicting migration to FLOSS is about introducing and

accepting a whole new framework of ideas and attitudes, i.e. creating (Dubie, 2005) a

cultural difference. The concept of availability of source code and the rights to freely

modify and adapt it, as well as freely copy and distribute the software, are rights

which the users of COTS are not familiar with. FLOSS promotes the culture of

knowledge sharing and level of collaboration that IT personnel may not be used to.

The individual attitudes towards FLOSS inside a large enterprise will most

likely resemble patterns that surround this concept. Some IT workers will like it,

some will oppose it, while others will be ignorant. As in the online communities, it is

very likely that the enterprise will have its own zealots and heretics regarding this

topic. Like in the flame wars inside the online community, similar disagreements

may occur between the participants of the migration project. Some people may even

dislike the concept so much that they would be ready to leave their jobs hoping to

save their own careers. Staged migration to FLOSS could support solving these kinds

of problems since it would provide time for people to accommodate to change.

The important question is if the intrinsic values of FLOSS can be used to

overcome at least some of the above mentioned problems, in the pre-migration

period. Gallivan (2000) found out that the perspective of community of practice is

considerably stronger than applying of traditional organizational models among

knowledge workers. Replicating (Melian et al, 2002) collaboration styles of open

source in an enterprise positively influences on creating diverse networked groups

within the enterprise. Strong organizational cultural beliefs (Elliott and Scacchi,

2003a) tie networked groups together so that conflict is more easily mitigated and

resolved. The sharing culture behind FLOSS is obviously very compatible with these

propositions for better internal functioning of organizations. The community

perspective is thus very important to be understood, because it can be sponsored

during FLOSS migration projects for improvement of internal collaboration and

communication.

20

To conclude the previous discussion, like any other IT project, FLOSS

migration may be perceived differently within the organization, therefore a full

support from top management is needed. It also requires strong commitment from the

IT personnel. FLOSS promotes the culture of knowledge sharing and collaboration

and this is the outstanding attribute of FLOSS migrations. Everyone should be

carefully guided towards adoption of these values for a successful migration.

2.3. The community perspective

In the following paragraphs the different approaches of enterprises towards

the community perspective are discussed and it is explained why the community

perspective is important for FLOSS migrations.

When enterprises migrate to FLOSS platforms, their contribution to the

community is not uniform. The majority of enterprises are passive users of FLOSS

and communicate with the community primarily when they need support, but they

don’t share source code. Rossi (2004) claims that not all enterprises contribute back

to the community. Some enterprises decide to release some of their in-house source

code in order to reduce the cost of its development and maintenance (Hawkins, 2004;

Bac et al, 2005). Still, many organisations (Fitzgerald and Kenny, 2003) are not

comfortable with making changes to the source code of the FLOSS products they

use. Other enterprises frame software development in their own organizational

boundaries in order to enforce their intellectual property (Kogut and Metiu, 2001).

These differences reflect how different enterprises react on various IT

challenges. Egri (2001) believes that companies are reacting on IT challenges in

different ways. The level of IT demand has significant impact on this. Kwan and

West (2004) point out that there is considerable variation of IT demand within

organizations and even organizational subunits. Enterprises will more likely be open

to new technologies (Dedrick and West, 2004) when IT is strategic to the business.

21

Ciborra and Andreu (2001) point out that strategic use of knowledge and its

management depends on particular organizational settings. Therefore it is

explainable why there is no single pattern in how LSEs collaborate with the FLOSS

community. However, the results of the survey conducted by CIO Insight (Agostino,

2005) shows that adoption of FLOSS has fostered innovation in almost two thirds of

the enterprises. Adopting FLOSS concepts may advance the attitude towards

creativeness and innovation within enterprises and the relationship with the FLOSS

community is an important catalyst in this process.

One possible effort to become closer to the FLOSS community is to embrace

its development model in the enterprise, thus provide similar conceptual

environment. Researches claim that deploying an open source model internally can

improve work effectiveness and collaboration (Melian et al, 2002; VaSoftware,

2004; Scacchi, 2002a). But this model may not work for all enterprises (Sharma et al,

2002; Ven and Verelst, 2004) and management appears to be the greatest barrier

(VaSoftware, 2004). IT managers have probably the most important role in the

decisions that shape enterprise involvement in the works of the community.

Bonaccorsi and Rossi (2004b) suggest that the level of contribution to FLOSS

projects much depends on the social and technological motivations within

enterprises. They also found out that in general enterprises join FLOSS projects less

then individuals. One possible explanation is that the motivational frameworks which

reflect the settings in majority of enterprises are still adjusting to the global

community.

Enterprises with a weak commitment are not likely to adopt community-

oriented behaviours (Bonaccorsi and Rossi, 2005). Existing research shows that

(Wagner, 2005) enterprises which are accustomed to collaborative work with other

organizations seem to gain more from FLOSS than those with an internal focus who

value technology through cost savings. Lagace (2003) points out that successful

enterprises actively foster sense of inclusion among their employees. Still majority of

enterprises emphasise economic and technological reasons for starting their

contribution to FLOSS (Bonaccorsi and Rossi, 2003), which is not common for

individual contributors.

22

In the previous text it was explained that there are different patterns in how

LSEs collaborate with the FLOSS community. The relationship with the FLOSS

community eases the adoption of FLOSS, a process which seems to foster

innovation. It was explained that this is the reason why the community perspective is

important and that strong commitment from the whole organization is needed for

adopting behaviours that are community oriented.

2.4. Importance of addressing non-technical issues

A brief discussion follows regarding the importance of addressing non-

technical issues with FLOSS migrations.

Human resources supported and guided by an appropriate organizational

framework actually operate, use and change computer systems and a system can be

replaced with a new one only if the personnel are well prepared to perform and adopt

the change. While the number of failed IT transformation initiatives in enterprises is

very high (Sirkin et al, 2005), researchers point out that the main reasons behind

these failures are attributed to non-technological issues, (Dravis, 2003; Nelson and

Joshi, 1993; Batler Group, 2005) like organizational and social. LSEs have to learn

from these negative experiences when they plan to migrate to FLOSS. Otherwise,

space may be left for unreasonable interpretations of migration failures on expense of

technology. Research indicates (Sauer et al, 1997; Batler Group, 2005) that IT

managers in enterprises are not able to manage satisfactory many of the most

common causes of failure.

It is very important to address these findings in FLOSS migration projects by

implementing them in a migration policy. If the IT management is not competent and

skilled to control the project objectives in FLOSS migration projects where there are

many non-technological factors that may affect the project outcome, then it is

impossible to expect clear direction for the action of the IT personnel and much of

the project phases may be left to spontaneous flow. This is very dangerous for the

23

integrity of all enterprise resources involved during and after the migration. Scacchi

(2004a) points out that enterprises risk problems if people issues are slighted or

ignored in any of the project phases. In these cases the success could be

compromised even if the technical side of the proposed change was flawless.

To summarize what was presented above, addressing non-technical issues

with FLOSS migrations through building migration policies may reduce the number

of failed migration projects.

2.5. Summary

In this chapter it was presented that IT personnel and end users view IT

projects differently due to different motives. It was argued that it is important to

understand the organizational contexts in which FLOSS migration projects are

performed and to address the human and social issues in a migration policy, because

beside technology, migrations as well change people and organization. In the second

section it was explained that the perception of FLOSS migration may vary internally

in LSEs, due to which full support from management is needed and strong

commitment from the IT personnel. A prerequisite for a successful migration is the

guidance of employees to adopt the culture of knowledge sharing and collaboration

which is promoted by FLOSS. In the third section the different patterns of

collaboration between LSEs and the FLOSS community was explained. The fact that

relationship with the community eases adoption of FLOSS and triggers innovation,

leaded to conclusion that the community perspective is very important and that

adopting the principles of the community requires strong commitment, especially

from IT managers. At the end it was discussed that addressing human and social

issues FLOSS migration policies may reduce the number of unsuccessful migrations.

24

Chapter 3. Methodological issues regarding the

study

As elaborated in the introductory chapter, the purpose of this study is to

identify and discuss the non technical issues with FLOSS migrations in LSEs. The

study focuses primarily on human and social issues with FLOSS migrations, through

appropriate reflection on the general attributes that characterize the organizational

contexts of LSEs. The analysis of these issues requires familiarity with a range of

research disciplines, because human and social aspects in the area of information

systems research are inherently considered as very complex. In this chapter the

methodological issues regarding the research area of the study are discussed. In the

first section the gap in the existing academic research regarding FLOSS migrations

will be presented, while the section that follows the methodology used in this study

will be elaborated, which is exploratory research with analysis of literature from

multiple disciplines and secondary data. Justification is offered for the chosen

approach.

3.1. Issues related to research of FLOSS migrations

The discussion in this section is about the gap in existing academic research

that may be identified as lack of knowledge regarding non-technical issues with

FLOSS migrations and FLOSS adoption in LSEs in general.

The main gap in the existing academic research related to FLOSS migrations

is that existing literature covers predominantly technical aspects of migration to

FLOSS. IDA Open Source Migration Guidelines (European Communities, 2003) for

example, one of the most detailed FLOSS migration guidelines, only modestly

highlights the human issues that have to be considered and is focused predominantly

on the technical aspects of FLOSS migrations. The same shortcoming is found out in

25

other well known migration guidelines (KBSt, 2003; Treasury Board of Canada

Secretariat, 2005; Almond et al, 2004). All of the existing guidelines concentrate

mainly on the technical side of FLOSS migrations. Nevertheless, the number of

published guidelines for migration to FLOSS is growing. At the time of preparation

of this paper, no previous research could be found that would offer systematic and

structured analysis of the complex social phenomena present with FLOSS

migrations.

Restrictive access to research data may be one serious constraint that gives

some explanation to why non-technical issues have not been systematically treated

by researchers. Seidel and Niedermeier (2003) claim that most of the enterprises who

are involved in FLOSS activities do not disclose information about the

transformation of their internal processes and culture towards FLOSS. Spiller and

Wichmann (2002) point out that measuring the professional use of FLOSS in

organizations is more difficult than measuring the use of COTS. Another issue is the

domain size, i.e. the number of LSEs is very low indeed, which makes it even more

difficult to understand the non-technical issues that are experienced during migration

projects and to draw general conclusions about them.

Although the research area (Holck et al, 2004) of FLOSS is relatively young,

there is a growing number of research contributions that are investigating the

phenomena related to FLOSS from different perspectives. However much less

attention has been devoted (Dedrick and West, 2004; Ghosh et al, 2002; International

Institute of Infonomics, 2002) to FLOSS activity of enterprises than to FLOSS

activity of individuals. FLOSS adoption in the enterprise domain and its social and

organisational impact has not received so much attention from researchers (Conlon

and Carew, 2005). Gasser and Scacchi (2003) state that one of the topics needing

further research is identifying what policies should guide the acquisition, adoption, or

use of FLOSS in enterprises. Dedrick and West (2003) point out that comparatively

little research has been done to determine how the organizational adoption of FLOSS

is different from adoption of other technologies. It is very important to expand the

academic knowledge in this area, not only for filling the obvious theoretical gap, but

also because of the practical implications it could have on enterprises who undertake

FLOSS migration projects.

26

The fact that FLOSS is still an emerging area explains the lack of academic

knowledge about the behaviour of LSEs regarding this topic. Nikula and Jantunen

(2005) point out that FLOSS adoption is still in early phases. Therefore there is lack

of availability of clear directions and procedures. Furthermore according to Briggs

and Peck (2003) many migrations are performed without a recognizable migration

methodology. It is a real challenge for an IT manager to incorporate human and

social issues in a clear and phased FLOSS migration strategy that would form basis

for many positive changes in the whole enterprise. This study tries to fill the gap in

the existing research area and provide a more comprehensive view of how migration

to FLOSS and its collaboration and knowledge sharing principles affect enterprise

employees.

To summarize what was discussed in this section, there is a knowledge gap in

the area of FLOSS migration in LSEs and the weakness of existing migration

guidelines is that they do not offer framework for dealing with non-technical issues.

This study tries to fill this gap.

3.2. Description of the research methodology

In this section the research methodology chosen for this study is presented

and justified in respect to the previously described knowledge gap in the observed

problem domain. It is also explained why and how the research framework of the

relatively new research area called social informatics has been deployed as a helpful

tool in the analysis.

In order to pursue the goals of this study most effectively while considering

the lack of academic knowledge in respect of the area of FLOSS migrations, which

was described in the previous section, it is chosen the research in this study to be of

exploratory nature. Exploratory studies are used for exploring new areas (Saunders,

2000) from which conceptual frameworks are built and form basis for some future

27

studies (Lynn, 1991; Tellis, 1997). The exploratory study starts with a holistic

approach (Routio, 2006) and by saturating data from multiple viewpoints, a deeper

understanding of the problem is reached and new theoretical framework finally

emerges (Figure 3).

Figure 3: The method of alternating point of view in exploratory research

(Source: Routio, 2006)

According to Saunders et al (2000), one of the principal research strategies of

conducting exploratory research is through a review of existing literature. For the

purpose of gathering research information for this study, an extended literature

review was conducted and diverse qualitative secondary data was collected from

prior research findings from the following research disciplines: free and open source

software, management of human resources, power and conflict, system migration,

information systems development, motivation, change management, organizational

psychology and social informatics. The multidisciplinary approach was necessary for

effective analysis of the human and social phenomena that exist with FLOSS

migrations in LSEs. Changes of information systems affect also non-technical

settings, thus these implications do not fit completely into any particular and distinct

research discipline and instead they have to be analyzed in parallel from different

research angles.

28

The discussions throughout this study moderately reflect the personal

experience of its author as a systems manager in a large Eastern European enterprise

in the area of financing, which heavily relies on proprietary software products and

development platforms and IT personnel which is not much familiar with FLOSS

products and the underlying principles of the FLOSS community. The results will be

integrated and discussed through a presented outline of a conceptual model for

understanding human and social issues with migration to FLOSS. Any effective

migration must address these issues (Paul, 2005).

Information systems are constituted of technology, people and organizational

settings. One research framework that provides a scheme for examining social and

technical processes in networks of people who work together, are socio-technical

interaction networks (Scacchi, 2004a). The socio-technical approach is aimed for

technology oriented environments. Lyytinen (1987) points out though that socio-

technical approaches may suggest too limited change strategy since their image of an

information system is limited to a technical system. The dynamics of information

systems can not be explained through information technology observed separately

from social phenomena. Therefore a holistic approach is needed for better

understanding of issues related to technological changes in complex environments.

A holistic view of technology, processes and social structures is offered with

the framework suggested by a relatively new research field, known as social

informatics, which is complementary to the principles of exploratory research

explained in Routio (2006). Kling (2000) points out that a key idea of the social

informatics research is that the social context of information technology has a

significant role in influencing the ways in which people use information and

technologies, thus affects the consequences for organizations. Social behaviours can

not be treated separately of technologies. The discussion that follows in this study

will be guided with the key principles of social informatics. The rationale behind this

is, as explained previously in the analysis of FLOSS migrations in organizational

contexts of LSEs, the great diversity of behaviours in LSEs that can be found with

large IT projects. This proves that social phenomena cannot be isolated from the

specifics of the organizational environments and technological settings, instead they

emerge from them.

29

The greatest dangers in exploratory studies are premature releasing of

conclusions and overextending the exploratory phase by inadequate level of diversity

of viewpoints (Lynn, 1991). These problems have been addressed in the study by

strictly focusing on several areas where non-technical issues with FLOSS migration

fit into and these are: the impact that FLOSS migration may have distinctively on IT

personnel and end users; as well as the impact of staged FLOSS migration on the

people involved and on the social setting. At the end it is also worth to note that this

work represents only a preliminary exploratory study which offers an elementary

outline of a conceptual framework that can be extended in some future research.

In this section it was discussed that the deployed research methodology is

exploratory study. Justification of why this methodology had been used was

elaborated, as well as how the potential weaknesses had been addressed. It was also

explained that principles of social informatics were used in the analysis because of

their holistic approach.

3.3. Summary

In this chapter the methodological issues regarding the study were discussed.

In the first section a gap in the existing knowledge about the area of FLOSS

migrations in LSEs was identified. The weakness of existing migration guidelines

were described in sense that they do not deal with non-technical migration issues. It

was pointed out that this study tries to fill this gap. In the last section the exploratory

research methodology is elaborated and justified. The potential weaknesses of the

methodology had been addressed as well. It was also discussed that principles of

social informatics were reflected throughout the paper, since social informatics offers

a holistic view on various phenomena in information systems.

30

Chapter 4. Critical evaluation of non-technical

FLOSS migration issues

In this chapter a critical evaluation of non-technical FLOSS migration issues

is performed. These issues have been divided into three categories, which are the

main defined viewpoints on the problem domain. Each of them is presented in a

separate section.

The first and largest category of explored issues contains the issues related to

IT personnel. It is first presented how lack of external support impacts on internal

processes of change. The next topic is the impact of FLOSS on internal culture of

LSEs. It is explained why it is important for enterprises to contribute to the

community and how this involvement impacts on the internal social behaviour. Later

in the text a wider analysis is performed about attitudes and motivation regarding

FLOSS and about impact of factors like age and gender on FLOSS migration

outcomes. The importance of skills and training is also explored. Ideological issues

related to bridging the gap between internal culture of LSEs and FLOSS community

follows next. While exploring the importance of online communication, at the end

there is a discussion about establishing links with the community and identity issues

in that environment.

The second category of explored issues includes the ones related to end users,

their profile and role in the FLOSS migration project. Further in the text there is a

discussion about importance of training and challenges associates with it. The

importance and challenges of participation of end users in FLOSS migration projects

is also observed in detail, especially in relation to internal conflicts and power

distribution. In the last part it is discussed how to fight resistance to changes induced

with FLOSS migrations.

The last presented viewpoint on the problem domain is analysis of the impact

of staged migration. It is explored how staged migration affects end users and

31

internal groups and at the end it is explored how desktop migration to FLOSS

platforms affects end users.

4.1. Challenges and opportunities for IT departments

4.1.1. The impact of different vendor support for FLOSS and

COTS

In the following text it is discussed how external support models for FLOSS

differ from support models for COTS. The impact of this issue on enterprises is also

analyzed.

IT departments in large enterprises are typically segregated into

organizational units that are responsible for in-house software development, system

administration and end-user support (helpdesk, technicians, etc). IT departments also

communicate with multiple external vendors who provide different support and

services. Adopted FLOSS solutions certainly have to be supported during and after

migration is performed. The lack of vendor support has been claimed to be one of the

most serious barriers for adoption of FLOSS (Gustafson and Koff, 2004; Dedrick and

West, 2003; Vile, 2005; Koch, 2003; Meehan, 2005; QNB Intelligence, 2004;

Forrester Research, 2003, cited in Williams et al, 2005; The Swedish Agency for

Public Management Publication Service, 2003; Scherler, 2004). Wang and

Whitehead (2001) explain that it is even more difficult to receive global support for

FLOSS, especially in non-English speaking areas due to many contributing factors.

McHugh and Donnelly (2004) point out that without the support from commercial

enterprises, FLOSS would not have sustained growth in popularity.

32

There are several scenarios to support FLOSS in a large enterprise: to rely on

external vendor as a gateway between the organization and the FLOSS community,

to rely on internal resources, or to deploy some mixed approach. Gustafson and Koff

(2004) point out that the last scenario is a bridge to the open source world because

both the vendor and the organization have access to the source code. Holck et al

(2004) have done extensive research on these scenarios and found out several ways

in which organizations can participate and contribute to FLOSS projects: in-house

source code improvements, documentation writing, error reporting, bug finding and

fixing, suggesting improvements, supplying and maintaining technical infrastructure,

participating in management of the community organization, responding to help

requests and supporting local FLOSS communities.

The relationship between customer and vendor in FLOSS is fundamentally

different from that in COTS, more obscure and more robust (FFIEC, 2004; Holck et

al, 2004), because there is a wider range of choice for FLOSS users. Dedrick and

West (2004) claim that vendor support is more important to LSEs because they have

the financial means to afford it, while small firms are more likely to rely on their own

skills and free support from the FLOSS community. The findings of QNB

Intelligence (2004) suggest that promotion of Linux by large mainstream vendors

such as IBM and Oracle has increased the formal evaluation of Linux for 75% in

large European enterprises. McDaniel (2000) confirms that announcements of

FLOSS support by large IT companies have changed the perception of IT managers

in large organizations. There are even FLOSS products (Wagner, 2005) for which

there is no support available from the community and enterprises should find vendors

willing to provide that support.

Williams et al (2005) suggest that the above concerns are becoming weaker

partly because companies understand the FLOSS development model better and

learn how to support themselves through time. Self support allows greater flexibility

for IT personnel (Figure 4). There are also a diverse and growing number of support

options (The Dravis Group, 2003) and as demand extends, larger vendors will likely

increase their service offerings in FLOSS support. Support for FLOSS projects

(David et al, 2003) from external organizations has already increased significantly in

the last decade. Therefore, if vendors deliver and support FLOSS solutions

33

(Gustafson and Koff, 2004), from the organization’s perspective the support issue

can be actually nonexistent.

Figure 4: Support options for FLOSS

(Source: Australian Government Information Management Office, 2005; p.15)

But problems (Davis et al, 2000) can often be beyond the capacity of the

support of the commercial vendor, and the enterprise can resolve them in the same

manner as most FLOSS issues, because in general FLOSS authors are more

accessible than developers for commercial vendors. Internal support (Lassen and

Sorensen, 2001) can also be organized. But this task is not easy. Sirkin et al (2005)

explain how organizations often underestimate their ability to build internal support.

Organizations are just starting to learn how to overcome this gap. Actually, FLOSS is

now transforming the way how software is supported in enterprises (Koch, 2003;

Fitzgerald and Kenny, 2003; WatchIT, 2004). Organizations begin to learn how to

support themselves by involving in the FLOSS community.

In the previous text it was discussed that support models for FLOSS differ

from support models for COTS in sense that they are more obscure and that

enterprises begin to learn how to provide self support. The impact of this on

enterprises is that they gradually involve in the FLOSS community in different ways.

34

4.1.2. Impact of FLOSS on the internal culture

The following discussion is about the impact of FLOSS on the internal

culture of LSEs. It is discussed how internal groups in LSEs distinguish from the

FLOSS community, why it is important for enterprises to contribute to the

community and how this involvement impacts on the behaviour of internal groups.

Active involvement helps in fostering a sense of community (Steward et al,

2001), and the community is often first line technical support when problems occur

later. Of course, every type of involvement in FLOSS (GITOC, 2003) means

different requirements, commitments and benefits, and correlates to some particular

implementation approach. Communities exist around common interests (Figure 5).

IT personnel and especially IT management, have central role in discovering the real

values of FLOSS, not only for the economic benefit of the enterprise, but also for

building up new internal social relationships. It is because the impact of FLOSS on

the organizational culture is evidently made by deployment of information

technology, a resource which is managed by IT departments. Melian et al (2002), as

well as Dinkelacker and Garg (2001), explore how communities of practice within a

large enterprise can be encouraged by software technologies, maintaining knowledge

networks, adopting third party engagement and promoting contribution.

Figure 5: Communities built around common interests

(Source: Gabriel and Goldman, 2002; p.7)

35

Considering the above, migration to FLOSS does not change only technology

in the organization. It also creates conditions for establishing a dynamic connection

between the internal organizational culture and the FLOSS community. The above

analysis shows that with migration to FLOSS enterprises must prepare their internal

forces to first become acquainted with community standards, in order to be able to

further strengthen this relationship and provide efficient support. Relying exclusively

on external vendors is not enough, because organization would not reap the real

advantages of FLOSS. Companies may encourage developers to participate in

FLOSS projects as this (Hars and Ou, 2002) can expand their skill base, promote

peer recognition as external factor to individual motivation and the same may be

applied for users as well. This task is far from easy.

Like large social groups (Elliott and Scacchi, 2003b) with beliefs and values

manifested in norms that form behavioural patterns, enterprises have their own

unique cultures that form guidelines for employees how to do things internally, deal

with uncertainties and establish some degree of order in the organization’s social life.

These internal expectations can be very different from the ones shared within FLOSS

communities. The FLOSS community is differentiated by a distinct ideology that

consists of mutually related norms, beliefs, and values (Stewart and Gosain, 2001;

Stewart and Gosain, 2003; Elliott and Scacchi, 2003b; Scacchi, 2004a). Dahlander

(2004) points out that enterprises involved in FLOSS projects have to balance the

possibility of appropriating returns, while maintaining good relationship with the

community and compliance with its norms and values. Otherwise, if interests collide

then conflicts may arise. As Domino et al (2003) define, conflict begins when

personal goals, attitudes, values or beliefs are not accordant with those of another

individual. The same applies to social groups.

Over time, subcultures within an organization, such as occupational

subcultures (Elliott and Scacchi, 2003a) may manifest their own beliefs and practices

that make them distinguished from other groups (Guzman et al, 2004). But as teams

develop through time (Trimmer et al, 2000) so do their norms. Gallivan (2000) found

out that formation of attitudes, values, and norms is strongly influenced by opinions

within social and professional circles. The development of the internal teams in a

36

LSE and their mutual collaboration in different projects can actually reshape the

internal subcultures in the enterprise. This is a spontaneous process, but also

represents a deep reflection of management decisions, quality of human resources,

cultural determinants, etc. Therefore it is important how FLOSS migration impacts

on whole teams.

The main question is how enterprises can intentionally and systematically

streamline this development in the direction to make their own subcultures

ideologically closer to the FLOSS community. Thus provide the foundation for

efficient internal support. Training about the founding principles of the FLOSS

community would seem to be one of the first logical steps forward. However,

Gallivan (2000) identifies the problem that trainers and researchers mostly

conceptualize IT training at the individual level, rather than the sharing of attitudes,

values, and norms relevant to groups. Furthermore, he claims that assimilation of

such values, norms, and beliefs occurs more effectively directly on the job.

Considering the above, special emphasis has to be put on group preparation for

FLOSS migration.

It was discussed that internal groups in LSEs distinguish from the FLOSS

community and it is important for enterprises to contribute to the FLOSS community

in order to maintain proper support for FLOSS migration. Training is thus important,

especially for whole teams, for adoption of FLOSS concepts.

4.1.3. Coping with differences between IT personnel

IT personnel may react differently on FLOSS migration efforts. Particular

incentives may help in resolving problems, but may not be suitable for every job

position. The analysis below distinguishes the attitudes of developers and

administrators towards FLOSS migrations. It also raises the issue of preparing the IT

personnel for concepts they are not familiar with.

37

The social structure inside the IT department can have distinct and dynamic

complexity, unique for a particular organization. Common interests may create

different kinds of bonds and relationships, but people at work tend to group

themselves if they share similar problems. In general, two groups inside the IT

department have crucial role for the success of a FLOSS migration project: in-house

developers and administrators. The motives that drive these two groups can be

conflicting at times. While the developers are focused on the creative process of

providing functionality in software applications, the administrators would like to

have stable systems that keep running at best performance. Even though developers

and administrators express same affinity for working in teams, according to Griesser

(1993) developers seem to accept changes more readily and prefer jobs that offer

chance for higher personal growth, which is less the case with administrators. This

finding suggests that new technology should be evaluated by developers first and

after that by maintenance staff. For the technical staff to be motivated it is crucial

(Wiederhold, 2003) the maintenance tasks to be highly valued in the enterprise.

The practice of sharing (Nuvolari, 2003) source code of programs is not novel

feature of the software industry. Enterprise development (Augustin et al, 2002) often

reflects many positive aspects of the open source community, like software and

knowledge sharing and can be thought of some sort of open source microcosm, but

there are differences. Building a positive attitude of the IT personnel towards FLOSS

is very important for the success of the migration process. The main problem is how

to prepare the IT personnel for the concept which most of them are not familiar with.

Although with a long history, FLOSS is still unknown to many people (Daffara et al,

2000). Industry cases (Miller, 2004) show that developer training is the hardest part

of FLOSS migration. Developers typically resist to changes in the already developed

software (European Communities, 2003; Zuliani and Succi, 2004; Lederer and

Mendelow, 1989; Greenemeier, 2004), many personally attach to their preferred

tools and there is also a feeling of loss of power experienced by both developers and

administrators. Some developers underestimate FLOSS (Lussier, 2004; Johnson,

2002).

38

The qualitative analysis of several case studies conducted by Lederer and

Mendelow (1989) suggests that developers prefer to solve problems using old

technology and methodologies. This is not much consistent with the findings of

Griesser (1993) that information system personnel morale can be maintained through

implementation of advanced technology. But in typical settings, IT personnel show

tendency to conformism. Anecdotal evidence shows that change can be so stressful

to some people (Miller, 2004), that they are ready to leave their jobs and managers

have to find incentives for the majority, like buying new hardware, etc. Hodgson and

Aiken (1998) point out that an individual dislikes changes in general, but may have

positive attitude towards a specific change if it provides significant personal benefits.

Therefore if new technology brought some kind of benefits to IT personnel, they

might be willing to accept it.

The acceptance of FLOSS by IT professionals seems to be related to the

personal beliefs and professional motivation. It may be difficult to persuade an IT

professional to accept FLOSS instead of COTS development tools, if the individual’s

professional motivation is primarily based on commercial attractiveness of the

profession and high wages. The acceptance of the FLOSS concept seems to be

somehow related to the strongest personal motive to work in the IT field: whether it

is because of financial attractiveness, or because of experiences of joy and

creativeness in the profession. According to the equity theory (Joshi et al, 1986)

when individuals are placed in inequitable social interchanges, they become

distressed. Major reason for left jobs after a decision to migrate to FLOSS is the CV

dilution effect (European Communities, 2003), when some people feel that not using

“industry standard” software impairs their ability to develop their career. As

Blanchard and Blanchard (2005) point out, employees are more likely to leave the

enterprise prematurely if they are not developing their career toward their dreams.

It was described how among IT personnel there may be different reactions on

FLOSS migration efforts and that developers in general accept changes more readily

than administrators. Personal career development has strong impact on how FLOSS

migration is accepted.

39

4.1.4. Influence of personal motivation

An insight into how personal motivation influences the acceptance of FLOSS

migration is presented in this section. Comparison between intrinsic and extrinsic

motivation of FLOSS and in-house developers is also made. Emphasis is put on job

satisfaction and fun and on the need for additional incentives for retaining personnel

who do not favour FLOSS.

The problems presented in the previous section are a huge obstacle for

FLOSS acceptance by IT personnel in LSEs. They also reflect the differences in

motivations that drive the FLOSS community and the internal IT personnel in LSEs.

Seifert and Wieland (2003) explain that motivation of FLOSS developers is quite

different from the one of commercial programmers. It is of crucial importance to

determine the reasons behind this gap between the two cultures in order to

understand how to provoke internal changes and under what conditions. Research

shows (Rossi, 2004; Hars and Ou, 2002; Ye and Kishida, 2003; Weerawarana and

Weeratunge, 2003) that the individual motivations of FLOSS developers can be

divided into intrinsic and extrinsic.

Intrinsic motivation is very important (McHugh and Donnelly, 2004; Ye and

Kishida, 2003), because it is a precondition for involvement in FLOSS projects.

Intrinsic motivations (Rossi, 2004) can range from the pure joy of programming to

more social-oriented motives. Based on the results of a large survey, Ghosh (2004)

claims that social motivation is most important between FLOSS developers. There is

also considerable evidence that altruism and knowledge sharing are also among the

most important motivations of FLOSS developers since economic benefits are not

significant in their decisions (Rossi, 2004; Kogut and Metiu, 2001; Bonaccorsi and

Rossi, 2005; Noronha, 2005).

Motivation of IT personnel in LSEs is based on inherently different structure

of incentives than the one that sustains the FLOSS community. While it was

described that personal motivation is prime for FLOSS developers, the self-

40

motivation of developers in enterprises is hardest to achieve (Seifert and Wieland,

2003). Griesser (1993) points out that developers in organizations have lower social

needs compared to other IT staff. Programming job can be very individualistic.

Actually many individuals prefer autonomy (Griesser, 1993) as opposed to increased

team activity. In a large organization developers are typically given very limited

freedom in choosing projects and usually have to follow strict deadlines, which may

be the explanation to this introvert behaviour of in-house developers. In commercial

organizations some developers pay more tribute to other incentives, like economical

benefits for example. This is a natural result of the different career path that people

have followed in their life. People who participate in FLOSS projects are driven by a

specific framework of incentives, not necessarily applicable to all developers, or at

least not with the same priorities.

Sometimes developers express reluctance to programming tasks because of

high levels of stress, non-challenging or dull assignments, conflicts with managers or

end users, etc. This can be compensated with other motives. Salary is one of the

prime motives for IT personnel to work in commercial organizations. Dickson (1969)

shows how in a large organization low salary is the most frequently given reason to

leave job by IT employees. Parker (2000) points out that mere training leads to

having servile workers who do not do things freely and intelligently, but for the

salary earned. Igbaria and Guimaraes (1992) found out that intentions to leave the

job are result of low job satisfaction.

Since migration to FLOSS can influence on the job satisfaction of IT

personnel because of possible personal fears from affected career improvement, it

can be explained why it happens that people are ready to leave jobs when such

migration projects are announced. These situations can be balanced only with

offering other incentives to IT personnel. Managers have to offer (von Krogh, 2003)

both extrinsic and intrinsic incentives to encourage innovation: from higher salary to

enhanced peer recognition and a sense of belonging. And on the opposite, as the

FLOSS community grows, the demand for FLOSS educated IT personnel and the

visibility of FLOSS in general will rise, which will have positive effect on the

individual perceptions of FLOSS between IT personnel inside LSEs as well. Griesser

(1993) points out that developers and administrators generally have the same

41

motivating potential for their jobs, but it has to be considered that different

subgroups react differently on change. Therefore, managers must carefully choose

the right incentives for different IT job positions in order to influence better response

to FLOSS migration.

A fundamental intrinsic motivation of FLOSS developers is fun, or joy of

programming. Many researchers have pointed out the importance of this motivation

(Ambati and Kishore, 2004; Elliott and Scacchi, 2003b; Lakhani and Wolf, 2003) as

one of the most pervasive drivers. But this is not related only to FLOSS developers.

Feelings like joy are proven to have (Han, 2003) major impact on an individual

response to any systems. It is an important factor for job satisfaction in IT personnel

as well. The problem is that (Seidel and Niedermeier, 2003) in large companies,

hierarchy, bureaucracy and strict projects plans are factors that massively impair

freedom to pursue ideas like in FLOSS style of development, while developers

participating in FLOSS projects do not need sophisticated incentives to be motivated.

While (The Swedish Agency for Public Management Publication Service, 2003)

many developers enjoy creative programming and are fascinated by software

development, Waterman (2003) points out that not every developer would code for

the joy of programming alone. For different IT developers, joy of programming may

have different impact on personal decisions. Not all developers are the same and this

has to be considered.

In the previous paragraphs it was explained that intrinsic and extrinsic

motivation are different in FLOSS developers and in-house developers. Job

satisfaction is very important for IT personnel and fun as well, although fun has

bigger role for FLOSS developers. It was also explained why management needs to

consider additional incentives for IT personnel for better acceptance of FLOSS.

42

4.1.5. Influence of age and gender factors

The influence of age and gender on FLOSS migration is discussed in this

section.

Blanchard and Blanchard (2005) point out that young people want to have

more fun on the job. The studies of Ghosh (2003) and Meehan (2005) show that

FLOSS developers are young (Figure 6) and predominantly male. Experiences of

large enterprises are that young people usually become more proficient than older

colleagues (Kleintop et al, 1994; WatchIT, 2004). However, Griggs and Manring

(1986) point out that age is the most powerful way to distinguish differences in how

individuals view their jobs, careers and organizations. Older employees usually

express more loyalty to the companies they work for. The same applies to IT

personnel. Luthiger (2005) found out that age correlates negatively with the

readiness of developers for future commitment. Therefore, for younger developers in

LSEs, the setting up of proper conditions for experiencing the joy of programming

would ease their acceptance of FLOSS, while in older colleagues that task would be

more difficult to accomplish in the same way.

Ivancevich et al (1983) points out that the stress level in IT personnel grows

with age, i.e. with years of employment. Older IT personnel though are less likely to

leave job because of personal views of the FLOSS ideology or fear from affected

career. Job stability and security play more important role for them. Hadfield (2005)

points out that IT personnel suffers burnout from overwork more than professionals

in most other professions, but are less likely to leave their jobs because of that. Since,

FLOSS community is dominantly represented with young people, it is very important

to educate older employees about what FLOSS is all about, but also what will be the

benefits to the enterprise.

43

Figure 6: Age of FLOSS developers

(Source: Ghosh et al, 2002; p.9)

According to the above described, reasonable explanations about how FLOSS

impacts the organization, are more likely to be accepted by older IT personnel, while

younger workers look for challenges. Sockel et al (2001) suggest that focusing on

loyalty building can be used for fighting against turnover. It is very important to gain

equal level of loyalty of younger IT personnel. Younger developers would not

necessarily accept FLOSS better in different circumstances. The problem as pointed

out by Graham (2005) is that the most productive young people will always be

undervalued by large organizations. This risk has to be mitigated by IT managers in

FLOSS migration projects, especially in those where age plays important role in the

internal social relationships.

The fact that the FLOSS community is predominantly male, can also produce

potential problems with FLOSS migration in LSEs, where gender structure may be

more balanced. Research shows (Ghosh et al, 2002; Lin, 2005) that women do not

play a significant role in the development of FLOSS. As Lin (2005) points out, the

44

gender problem is marginalized in current FLOSS research, most FLOSS policies are

gender blind, but also a far more women-friendly environment for FLOSS

development is needed. One interesting observation of Lyman (2005) is that women

involved in FLOSS tend to be older than male participants and they are mainly

involved in non-technical areas like management, marketing, and leadership.

It is observed (Lin, 2005) that female developers usually obtain their

knowledge through the formal education system. The research of Ivancevich et al

(1983) shows that men report significantly higher commitment to organizations.

Gallivan (2003) proved that women IT professionals appeared to be more sensitive to

the total amount of change that imposes stress and pointed out an example project

where women reported considerably higher levels of stress when they were provided

more autonomy and less supervision. He also found out that women respond more

negatively than men to devoting considerable personal time for improving their skills

by themselves.

Kleintop et al (1994) also reported that gender has effect on skills acquisition.

A reasonable explanation of this is that (Lin, 2005) women need to take care of the

housework and require a more stable working time. Many talented and experienced

professional women (Blanchard and Blanchard, 2005) want to stay home. Dickson

(1969) indicates that home duties are major reason to leave job for female IT

workers. Obviously gender differences may have great impact on FLOSS adoption

among IT personnel and may cause problems. However, the relationship between

gender and intention to leave job is not proven to be consistent (Gallivan, 2003).

Ivancevich et al (1983) suggests that gender differences are also related to cultural

attitudes and values.

The influence of age and gender on FLOSS migration was discussed in this

section. It was discussed what kind of different treatment must be done for younger

and older personnel, as well as for female and male personnel, during the FLOSS

migration project. It was found out that young personnel respond to challenges better

than older colleagues, while male employees are better self-learners than their female

colleagues for whom low level of stress is of more importance.

45

4.1.6. Improvement of skills and knowledge sharing

In the text that follows, the importance of skills and training for FLOSS

migration projects will be discussed. The importance of peer recognition as a team

factors will also be reviewed.

Learning new skills (Lakhani and Wolf, 2003; Ghosh et al, 2002; Maass,

2004; Seifert and Wieland, 2003) is also one of the top motivators for FLOSS project

participation (Figure 7). Companies that exploit FLOSS (Bonaccorsi and Rossi,

2004a) have greater chances of improving their programming skills. But a question

remains if all IT personnel can be equally personally motivated to improve their

skills. Since the IT profession has little skill stability compared to traditional

professional careers, (Sawyer et al, 1998) IT professionals must keep themselves

skilled and not rely on corporate guidance.

Figure 7: Reasons to join and stay in the FLOSS community

(Source: Ghosh et al, 2002; p.45)

46

As described earlier, in LSEs personal prioritization of skill improvement is

strongly related to other factors, like age and gender. Social influences also matter

(Gallivan, 2000) to skill development. Decision for migration to FLOSS (Almond et

al, 2004) can cause to IT personnel the impression of disapproval of the already done

work or devaluation of their current skills. On the other side, numerous works point

out that adequate technical skills are very important for the success of FLOSS

migration (Larsen et al, 2000; Dedrick and West, 2003; Dedrick and West, 2004;

Winslow, 2005; Kenwood, 2001; Vile, 2005; Almond et al, 2004; European

Communities, 2003; Maqua et al, 2004; FFIEC, 2004). Of course, all participants of

a project team need to have certain skills and knowledge to accomplish a given task

as a team (Leonard, 2002). Therefore, successful FLOSS migration also needs

possession of proper skills by the complete migration team.

One of the issues which any LSE has to face is how to update the skills and

knowledge base of the staff in a timely manner (Fitzgerald and Kenny, 2003). One

approach to fill the gap (Niederman and Webster, 1998) created by new technology

between individual skills and those needed to fully utilize the new technology, is

training. It is very important to train the IT personnel for the FLOSS concept. There

are positive examples of FLOSS migration case studies where (WatchIT, 2004)

training of IT personnel has improved morale and has lowered turnover. However

training initiatives for FLOSS migration must be designed very carefully. The

logistics for organizing proper training in LSEs can be very complicated issue.

Support of skills development in practice is usually an abstract category which

companies pretend to engage in without actually taking clear measures (e-Business

Watch, 2002b).

Other methods than standard training sessions can be more effective in skill

preparation for FLOSS migration. Cooperative learning theory (Janz, 1998) suggests

that productivity of a group will be enhanced if its members have interdependent

goals, help or teach each other the needed skills, and periodically evaluate the

group’s performance. Setting up an effective knowledge sharing between IT

personnel in a LSE can help them to prepare for FLOSS migration.

47

It was described how intrinsic individual motives of the IT personnel are

related to the success of FLOSS migration projects. The analysis of the impact of

knowledge sharing shows that social aspects are also of crucial importance for

successful FLOSS migrations. Stewart (2004) suggests that team factors influence on

the success of FLOSS development efforts. Peer recognition is very important social

motivation for software developers in general (Baldi et al, 2003; Robinson, 2005a).

Peer recognition in a group is possible only if open communication exists between

group members. Since the FLOSS community (Eilola, 2002) places primary value on

communication rather than on program functions, it is understandable why peer

recognition is important to individual contributors. In that sense, need for peer

recognition is one attribute that characterizes both enterprise and FLOSS developers.

But it is based on different premises, because the communities in which peer

recognition takes place are different.

In the previous text it was discussed that technical skills are very important

for success of FLOSS migration projects and how training improves morale and

lowers turnover. It was elaborated that peer recognition is an important team factor

that influences on success of FLOSS development efforts and improves knowledge

sharing.

4.1.7. Approaching the FLOSS community through

acceptance of new collaborative practices

In this section a discussion follows about the issues related to encouraging IT

personnel to become involved in the FLOSS community.

With FLOSS migration and approaching the FLOSS community for support

for the first time, IT personnel from a LSE will face unknown territory, defined with

rules different from those they are used to see in the enterprise environment. It was

explained in detail how both communities will express different values, norms and

48

beliefs. The main role of IT management would be to try as a first step to acquaint

the IT personnel with the values, norms and beliefs of the FLOSS community. A

company that begins to cooperate with a FLOSS project (Seifert and Wieland, 2003)

must be aware that the motivation paradigm of the community will influence the

motivation of the employed developers. The open-source movement (Markus et al,

2000) suggests ways how to motivate and direct knowledge workers.

Developers may have more fun and identify themselves as part of a larger

community (Hars and Ou, 2002). Ye and Kishida (2003) point out that sense of

belonging to a community is precondition for successful FLOSS projects. The

community spirit (Elliott and Scacchi, 2003b) in the forum of collective beliefs,

values, and norms fosters collaboration and resolution of conflict. But allowing IT

personnel to communicate more closely with the community is not possible if they

don’t receive a high degree of freedom and autonomy. Seifert and Wieland (2003)

indicate that high degree of autonomy seems to be more efficient incentive for

employed developers to participate in FLOSS projects than typical task assignment

or usual salary incentives.

Griggs and Manring (1986) confirm that financial incentives are weak

motivators. Open collaboration (Michlmayr et al, 2005) with other developers and

receiving information from others also increase the motivation to develop software.

Finally, job characteristics theory suggests (Janz, 1998) that autonomy is positively

correlated to job satisfaction and performance. If IT personnel expresses positive, or

at least neutral attitude towards FLOSS migration, with proper management

guidance, the chance of adopting the FLOSS philosophy may increase. But for

individual cases where opposition to FLOSS is encountered, other incentives have to

be offered with greater priority. Both organizational and FLOSS culture are dynamic

systems and change over time. As Ghosh (2003) explains, over time other aspects

gain in importance for FLOSS contributors, such as commercial and political aspects.

Similarly, the same may happen with IT personnel after the FLOSS migration.

The key (Sharma et al, 2002) to having successful hybrid FLOSS

environment in an enterprise is to identify appropriate projects and personnel. The

human potential in LSE is a result of different selection criteria. FLOSS community

49

is filtering the best players on the basis of self selection between contributors, while

LSEs are typically facing problems with the dynamics of their human capital. Most

of the IT personnel in LSEs will typically be so called B players (DeLong, as cited in

Fonda and Locke, 2003; Lagace, 2003), or the middle of the workforce: very loyal

employees, who balance their work with personal lives, do the job and most

important do not leave the company. Domenick (2004) points out that star

performers in enterprises are also the players most likely to leave the organization for

opportunities elsewhere.

On one side, star performers are expected to deal better than their averaged

colleagues with technicalities of FLOSS migration, but on the other side the risk with

some of them from having their personal disapproval of adoption of FLOSS ideology

as a reason to leave job, is higher. As Domenick (2004) further explains, best players

behave the way they do because they are looking for something they want, not

necessarily for something the company wants, and if they get what they want from

the company, they may stay and begin to behave like average performers (Blanchard,

2004). Fonda and Locke (2003) explain that middle players tend to think of the

company as a family and they often take the time to guide and train inexperienced

employees.

That is why recognition and appreciation of employees is very important

prerequisite for a FLOSS migration project. Otherwise, it is hard to expect from

people to try to devote themselves to the FLOSS ideology which they are not

familiar with. So, even though it is important to have key skilled IT people for the

migration, it is more important to involve everyone. However, because of (Grudin,

1994) the social and political factors at work, achieving groupware acceptance is

much trickier than individual acceptance and some people must adjust more than

others. Involving the complete IT personnel and strategically changing its culture by

paying attention to individual differences is needed not only for successful FLOSS

migration, but also for efficient support of the new platform after the migration.

The practice of peers evaluating each other's work, rather than relying on

external judgement is called peer review (Stalder and Hirsh, 2002). Peer review has

always played important role in FLOSS projects, especially for quality assurance

50

(Ambati and Kishore, 2004; Seidel and Niedermeier, 2003). But there are strong

social motivations behind peer review in the FLOSS community, which Augustin et

al (2002) define as peer glory. Developers compete to produce better software code

and peer recognition distinguishes the best performers in this process. This has

positive influence both on existing and new developers. Gacek et al (2004) confirm

that peer recognition has its role in motivating contributions to the FLOSS

community.

In typical large enterprises there is no peer review in the in-house

development process (Augustin et al, 2002; Parker, 2000). Augustin et al (2002)

point out that enterprises instead often promote competition and suppress mutual

support through individual performance reviews and bonus incentives. There are

cases where some developers keep their code jealously from other colleagues and

avoid potential criticism, which may create false impression for their value. Augustin

et al (2002) found out that enterprises are not cultivating collaborative efforts and

management does not promote knowledge and code sharing between developers.

This means that in LSEs there is profoundly weak culture of sharing, which may be a

very serious problem with FLOSS migration initiatives.

Large enterprises are simply used to different kind of development. In other

situations, when they are given tasks to change piece of code produced by another

colleague, some in-house developers may be reluctant to undertake such tasks,

because they are not always sure about other’s code quality and legibility, especially

if there is no proper documentation. Nobody would like to risk his or her reputation

after all. Strand (1980) points out that for most programmers it is important not to be

considered by colleagues as the creator of the least understandable programs. Peer

review (Hirschheim and Klein, 1989) can be seen as “group think” and its pressure

has positive impacts on the loyalty of developers.

Unlike implementations of COTS systems, FLOSS migration introduces

unique positive collaborative practices in LSEs and peer review is one of them. The

effort of adopting norms and practices of the community in LSEs with the goal of

successfully supporting FLOSS migration can be extended with promoting peer

review inside enterprises. Internal peer review may be both a quality control

51

mechanism and a potential motivating factor if accepted by the personnel. This may

be extended to the FLOSS contributions made by the team. From passive users, the

internal development team (Schmitz, 2001) will feel promoted as a member of a

larger community of peers and will contribute with code, ideas, tips, etc. Even

though these activities may consume time, the important gain may be higher

motivation and better in-house competence.

To summarize the key points in this section, developing sense of belonging to

a community is precondition for successful FLOSS projects. In order to accomplish

that, enterprises first need to encourage autonomy and internal collaboration

practices like peer review, but also pay strong attention to recognition and

appreciation of the work done by the personnel.

4.1.8. Issues with personal and shared ideology and beliefs

In this section, issues with ideology and beliefs are examined in the process

of narrowing the gap between the in-house personnel and the community. Both

individual and group perspective will be discussed in respect of the ideological

transformation that is needed to position closer to the community.

Ideological perspectives of software development are very important factor of

FLOSS adoption among IT personnel in LSEs. The belief in free software is a core

motivator of free software developers (Elliott and Scacchi, 2003c; Payne, 2002;

Weiss, 2001). But not all IT professionals share this belief. Actually as Ferris (2003)

points out, the history of corporate IT is riddled with holy wars. There is a distinct

polarization between IT professionals on the values of FLOSS. This normally affects

individual attitudes of many employed IT experts. Weiss (2001) observes that

ideology is involved in the choices of software developers and systems

administrators.

52

Ferris (2003) has discussed this phenomenon from the perspective of two

extreme camps which he calls zealots and the heretics. The first blindly believe in

FLOSS, while the others are fanatical about their disbeliefs. He points out that even

though FLOSS is firmly penetrating in corporate IT and is even implemented in basic

stages even without the knowledge of senior managers (Figure 8), the conflict of the

believers vs. non-believers can be damaging for enterprises. Seifert and Wieland

(2003) found out that FLOSS also affects the attitudes on mid-level management

who can be convinced much more difficult for what are the benefits and the necessity

for organizational changes. Problems may occur if the IT managers express fanatical

behaviour about certain technology, either pro or against FLOSS.

Figure 8: Unofficial experimentation with Linux in enterprises

(Source: QNB Intelligence, 2004; p.3)

When an IT manager has been involved in building the previous

infrastructure, much of the manager's reputation is tied to it and the manager may be

rejecting ideas before they are even presented (Robinson, 2005a; Robinson, 2005b).

Robert and Schutz (2001) claim that religious wars are not commonly found in

decision makers, due to strong hierarchical and commercial pressures. Zealot

developers on the other side can be a terrific resource (Robinson, 2005b) for problem

solving and knowledge sharing with others in the team. If the IT personnel are

divided in the acceptance of FLOSS, then this risk has to be mitigated.

53

Implementation of FLOSS in cases where developers and administrators

don’t favour it actually redistributes their internal power position and new

technology that they don’t understand may directly affect their reputation in the

organization, as well on their career. Ideologies are more emotionally intense and

resistant to change than rational behaviour because they help individuals deal with

uncertainty (Elliott and Scacchi, 2003b). IT managers have to address these issues

carefully in the migration policy. Stewart and Gosain (2003) explain that subgroups

which are formed around projects may vary in the level of conformance to the

ideology of the community. Therefore, each subgroup in the IT department formed

around specific projects that involve specific technology, has to be addressed

separately. In that direction, further discussion about this issue should be centred on

shared ideologies.

Stewart (2004) generalizes that the existence of a shared ideology in a group

is believed to facilitate group efforts. Guzman et al (2004) point out that cultures are

usually dynamic and while some cultures may accept changes easily, others react

with different levels of resistance. In that sense, the FLOSS migration project must

provide means for adequate ideological transformation of the IT personnel. Stewart

and Gosain (2003) suggest that enterprises may encourage the development of

compatible ideological elements that will help the IT personnel build trust in the

community. Much of that can be performed through different tasks in which the

personnel have to address directly to the community. When a specific task has to be

accomplished, (Guzman et al, 2004) groups need to adjust the way how they do

things and work together with a shared goal and with minimum conflict. LSEs have

to determine specific elements of migration policies how to bridge this gap. FLOSS

implementations require cultural changes inside enterprises (Seidel and Niedermeier;

2003; Seifert and Wieland, 2003; Treasury Board of Canada Secretariat, 2005).

To summarize the above, issues with ideology and beliefs in the process of

narrowing the gap between the in-house personnel and the community were

examined. The polarization on the values of FLOSS means that ideology is involved

in the choices of IT professionals and this risk has to be mitigated. It was described

how FLOSS migration projects must provide means for adequate ideological

transformation of the IT personnel and special attention has to be given to groups. It

54

was concluded that successful FLOSS implementations require cultural changes

inside enterprises.

4.1.9. Issues with online communication and usage of online

resources

The FLOSS community performs activities through Internet. In this section

the issues with online communication and the involvement of personnel in online

collaboration together with accompanying risk factors are discussed.

FLOSS development is an international activity and the communication

within the community is performed through the infrastructure of Internet (Scherler,

2004; Weerawarana and Weeratunge, 2003; Augustin et al, 2002; Peeling and

Satchell, 2001; Osterlie, 2004; Sharma et al, 2002; President’s Information

Technology Advisory Committee, 2000). Actually the whole Internet infrastructure

is laid on technology rooted in the FLOSS community. The unique social aspects of

the FLOSS community are reasonable explanation for the findings of Lerner and

Tirole (2002) that rapid responses are not as common in commercial software

projects, as in FLOSS projects. Internet also provides resources to the FLOSS

community to express its incredible ability to mobilize members (Morgan, 2002).

Schweik and Semenov (2002) point out that collaboration over the Internet

greatly increases the speed in which innovations can be published. Kogut and Metiu

(2001) discuss that Internet can be utilized for communication and collaborative

work. The power of the global Internet network in connecting communities of people

with common interests is exceptional (Dempsey et al, 2002). Internet has very strong

impact on the corporate communication with the outside world. Access to Internet

has not been a problem for LSEs (Robert and Schutz, 2001) because they have

resources to take proper care about information security. When enterprise IT

personnel need to communicate with the FLOSS community online, they would

normally use various communication resources such as different newsgroups,

55

mailing lists, archives, databases, wikis, blogs and chat rooms (Wheatley, 2004;

Sarma and Hoek, 2004; Wang and Whitehead, 2001; Williams et al, 2005; Seifert

and Wieland, 2003; Relevantive, 2003; Thronicke and Berger, 2004; Kenwood,

2001; McHugh and Donnelly, 2004).

In one case study, Fitzgerald and Kenny (2003) found out that the biggest

learning has been to orientate the internal IT support staff to effectively utilize the

Internet and other resources to deliver support for FLOSS. Many enterprises

(Fitzgerald and Kenny, 2003) may face internal resistance form IT personnel because

some prefer to communicate with vendors instead of search solutions to problems on

bulletin boards. It is crucially important to stress that using the Internet

communication resources described above, is also a skill. In enterprise environment,

not every IT employee has the same level of this skill. Some developers or

administrators use resources of the community more efficiently; others even don’t

understand certain communication technologies.

It is common in LSEs where the employees may have highly diversified age

and gender structure, to notice different quality of the Internet usage patterns

between the IT personnel. On the other side the communication in LSEs for in-house

development is (Ambati and Kishore, 2004) based on person to person

communication. There are certain rules (Bergquist and Ljungberg, 2001) for

communication that distinguishes each different group which is part of the FLOSS

community. If members (Erenkrantz and Taylor, 2003) do not share a common

language, it becomes hard to communicate effectively. Shaikh and Cornford (2004)

suggest three main elements have to be learned: the source code, the community that

surrounds it, and the engaged process. Important implication out of these findings is

that the IT personnel have to be properly trained to improve their online

communication skills, but also to understand the language and ethics of the FLOSS

community.

Things can become complicated if IT personnel are of different ethnical

origin than the community they are going to approach for support. The language may

also be a serious barrier, especially for employees that are not good with foreign

languages. Wang and Whitehead (2001) point out that language and culture issues

56

may affect online usage by in-house staff. On the other side FLOSS provides

(UNDP, 2003) means for creation of local and relevant content. This may mean that

enterprises may establish new bonds and friendships with other local organization,

through the use of Internet. Local communities are characterized with (Maggioni,

2002) relatively uniform systems of attitudes and values. The cohesion of the

national culture has specific impact on the relationship with the community. Payne

(2002) for example points out that some countries may not trust software developed

in another country that is a potential enemy.

Cultural differences can be an obstacle for providing self support for FLOSS

migration and FLOSS related projects and these issues should be considered as risk

factors. Cultural dysfunctions between enterprise IT personnel and the community

can cause problems to implementation and contribution efforts (Bleek and Finck,

2004; Guzman et al, 2004; Garcia, 2001; Ambati and Kishore, 2004; Elliott and

Scacchi, 2003a). Enterprises also have their distinct cultures which have direct

influence on the social behaviour of IT personnel. Through IT personnel, the values

and principles of the FLOSS community indirectly influence on the social dynamics

of enterprises. And vice versa. As long as the number of employed members of

FLOSS community grows, its social dynamics will also change.

One of the prime forms of contact between the IT personnel and the FLOSS

community needed to support FLOSS migration is lurking. As mentioned previously,

the internal IT staff must be able to reach diverse Internet community

communication hubs like mailing lists, newsgroups etc. Since at the beginning they

do not have established social contact with the community, at first they behave as

passive observers and information consumers, i.e. they become peripheral members,

i.e. lurkers. In their extensive analysis of the behaviour of lurkers, Takahashi et al

(2003) define two types of active lurkers: propagators, who propagate knowledge

gained from an online community to others outside it; and practitioners, who use

such knowledge in their own organizational activities. Active lurkers are therefore

important assets in the early stages of FLOSS migration because they disseminate

and apply knowledge essential for providing self support from FLOSS platforms.

These are skills that IT personnel have to develop.

57

Takahashi et al (2003) believe that it is useful for managers to focus on

lurkers with the aim of making appropriate decisions. Passive lurkers, that do not

share or use information gained from the community, may also be important for the

enterprise that plans to migrate to FLOSS. Passive lurkers are important in the early

stage of understanding how the community works and what the rules that guide it

are. Part of the passive lurkers may become active lurker candidates, if the online

community affects the lurker’s thought (Takahashi et al, 2003). As Zhang and Storck

(2001) suggest, even though these peripheral members behave differently from active

members of the community, they should still be treated as members because they

share many similar attributes with the community and also contribute to it. The gap

may become smaller over time if IT managers provide adequate resources and slack

time for IT personnel to access community resources.

It is important to note that (Maass, 2004) plain users and lurkers of FLOSS

express negative levels of trust towards community members, while those who start

to contribute demonstrate very high trust in other members. Involvement in FLOSS

projects can be of various forms (Figure 9). Deeper involvement may improve the

collaboration based on trust between the IT personnel and the community.

Bonaccorsi and Rossi (2004a) found out that enterprises that are more involved in

FLOSS are stronger linked with the community.

Figure 9: The classification of FLOSS users and developers

(Source: Gacek et al, 2004; p.10)

58

It is also important to address the risk factors associated with more open

communication with the virtual environment outside of the enterprise boundaries. IT

personnel have to take special care about integrity of enterprise security and

intellectual property during their communication with the community for supporting

FLOSS. Large enterprises (Lambe, 2002) due to the big number of controllable

internal environments sense risk and danger less acutely, even though they have the

resources to address them. The amount of open source information that can be found

about enterprises (Rues, 2003) also appears to be a cultural issue. Lassen and

Sorensen (2001) discuss a case study of a large financial enterprise that has forbidden

participation in discussion forums on the Internet, which heavily restricted the

support of an internally deployed FLOSS product. This example is in line with the

findings of Sharma et al (2002) that it is difficult for large traditional enterprises to

provide an environment to their personnel similar to the one provided in the FLOSS

community, due to rigid structures, processes and culture.

The exposal of proprietary information can be devastating to the internal

information security. IT personnel have to be given clear guidelines about these

issues during the FLOSS migration. Benkler (2001) points out that exclusive private

right in information exists in tension with individual freedom to read and express

oneself. Melian et al (2002) claim that collaboration may have negative aspects for

enterprises such as: control, visibility and exposing to competitors. Therefore

enterprises should properly address and mitigate these risks in their information

security policies (Rues, 2003; Balduino, 2005).

In this section it was examined how important online communication is for

success of FLOSS related projects. It was argued that in-house personnel should be

trained to improve online communication skills and understand the standards of the

FLOSS community and that cultural differences may also affect. The importance of

lurkers was further argued, because they disseminate knowledge essential for self

support. It was concluded that due to numerous risk factors, organizational size and

culture it is difficult for LSEs to provide an environment similar to the FLOSS

community, but with clear policy and guidelines these issues may be addressed.

59

4.1.10. Process of changing the internal culture

This section discusses the differences between the internal culture in LSEs

and the FLOSS community and the need for transformation with the goal of

successful implementation of FLOSS.

Weerawarana and Weeratunge (2003) point out that community participants

in FLOSS projects are individuals and not organizations. But this may change over

time. The survey results of Ghosh (2003) show that one third of FLOSS developers

are paid for their contributions, while project leadership is primarily done by

professionals. Similar findings which suggest that developers tend to be highly

educated and employed (Figure 10) can be drawn from the two largest surveys of

FLOSS developers conducted by David et al (2003) and Ghosh et al (2002), as well

as from the work of Lakhani and Wolf (2003). These are indications that the role of

enterprises in the community is growing. The boundaries between the professional

culture and the community are constantly changing (Holtgrewe and Werle, 2001).

Figure 10: Employment status of FLOSS developers

(Source: Ghosh et al, 2002; p.14)

Enterprise surveys show (SoftwareMag, 2005) that internal collaborative

culture is important to overwhelming majority of enterprises. But there may be

specific requirements. Moreover, these distinct corporate cultures (Griggs and

60

Manring, 1986) affect the priority issues identified by professionals. Ven and Verelst

(2004) claim that there is no way to force people to comply with certain guidelines

that are set by the project team. The biggest difference between the software

development styles of the FLOSS community and typical LSEs is that enterprise

developers can not choose the projects on which they work (Augustin et al, 2002;

Seidel and Niedermeier, 2003; Lee and Chan, 2004). If enterprises want to benefit

with means for better support of FLOSS migrations by improving their contribution

to the community, they have to adapt their internal organization (Seidel and

Niedermeier, 2003). Enterprises have to change the internal development style and

provide more autonomy to the IT personnel.

There are indications that FLOSS projects are more successful in areas that

are more interesting to the developers (Bezroukov, 1999). In the in-house specific

domains, it is a real challenge for IT managers to find means how to make projects

more interesting and foster creativity between IT personnel. The element of

playfulness is very important for the motivation of IT personnel in that direction.

Like mentioned before, the collaboration efforts of LSEs for embracing the

principles of the FLOSS community cannot be improved without giving higher

autonomy to the IT personnel. As Blanchard (1996) points out, empowerment is a

key tool for any organization that wants to grow. Empowerment however depends on

many external factors. Clement (1990) claims that empowerment cannot be

accomplished without redefining existing social relationships.

The flexible automation paradigm discussed by Bianchi (1996) provides

reasonable alternatives with new organizational principles which oppose to existing

rigid organizational structures. These alternatives are effected through development

of organizational flexibility. Autonomy of IT personnel in LSEs can be fostered only

in emerging organizational systems where higher degree of flexibility exists (Figure

11). Benkler (2001) adds that enterprise regulation may serve or defeat autonomy. So

organizational flexibility has to be deeply covered within internal policies. However

Agarwal and Sambamurthy (2002) point out that even though the federal model of

distributed governance is particularly appropriate for LSEs, it may not adequately

address the strategic, organizational, and technological realities faced by IT

executives.

61

Figure 11: Comparison between traditional and emerging organizational systems

(Source: Klenke, 1996; p.33)

These implications suggest that autonomy of IT personnel needed for

providing highest quality of internal support of FLOSS can not be accomplished to a

higher extent without some social transformation of the whole enterprise and without

introducing changes in the overall organizational strategy. Blanchard (1996) found

out that autonomy reflects boundaries which encourage self-control and

responsibility and which are different from the boundaries found in bureaucratic

systems. The findings from the case studies of Janz (1998) also support the opinion

that autonomy and cooperative learning are positively correlated with work outcomes

of self-organized teams. These positive outcomes are possible only through shifting

the context of IT management.

Seifert and Wieland (2003) advice that developers should be given the

freedom to establish and contribute their own ideas, especially with relaxed time

schedules and management. However autonomy which is fully uncontrolled can also

cause problems, like in the example when a developer who changes FLOSS source

code of one in-house application leaves the company (Stafford, 2005).

62

Figure 12: Example of community structure of a big FLOSS project

(Source: Barahona et al, 2004; p.45)

Figure 13: Example of social network of FLOSS developers

(Source: Scacchi, 2004b; p.13)

One very important attribute (Melian et al, 2002; Lin, 2004; The Swedish

Agency for Public Management Publication Service, 2003; O’Reilly, 1999) of the

FLOSS community are dynamic social networks (Figure 12 and Figure 13). But the

ability to create social networks or involve in them is not a characteristic of a typical

LSE. Parry (2000) strongly points out that there is contractual view of social

63

networking and creating relationships inside enterprises, and as a result of that there

is less trust and absence of social capital. Research indicates that a paradigm shift

towards network oriented enterprise is possible only if there are competent managers

who will undertake the task of reversing these negative dispositions (Parry, 2000;

Melian et al, 2002; Galup et al, 1998), which can be possible sometimes only with

hiring adequate professionals.

In their two year study, Agarwal and Sambamurthy (2002) found out that one

of the main characteristics of leading innovative enterprises is the ability to nourish

social networks. Leading companies are also attributed with excellent management.

Sarin and McDermott (2003) point out that best team leaders are well networked

both inside and outside the enterprise. Lack of social networking is not only a result

of enterprise authoritativeness and rigidness, but may also be a result of lack of

personal skill of enterprise employees. Sawyer et al (1998) found out that social

skills of IT personnel in enterprises are much harder to improve. Thus enterprises

simply need new form of leadership that is more networking based (Van de Loo,

2000).

LSEs have to understand (Windley, 2005) that people work a lot harder on

things they like, but in majority of cases the average office is a not associated with

fun. There are indications that software development in FLOSS projects is more fun,

while in-house development in LSEs is more associated with time pressure and

boring work (Luthiger, 2005; Ghosh et al, 2002; Rossi, 2004). Furthermore as

Patricia (2005) describes, departments within large enterprises often work like in

closed silos. The factors that lead to these negative perceptions have to be changed to

the root wherever FLOSS migrations of wider scale are to be performed.

In this section the differences between the internal culture in LSEs and the

FLOSS community were presented. It was argued that for better support of FLOSS

migrations, LSEs have to adapt their internal culture, through empowerment and

autonomy of employees. It was pointed out that management that understands and

promotes social networking is a prerequisite for this process.

64

4.1.11. Establishing bonds with the community and self

identity of the IT personnel

This section discusses how bonds can be established with the community and

how IT personnel will perceive themselves in the new environment.

Socializing with community (Zhang and Storck, 2001) is difficult without

regular communication between members. Enterprise IT personnel have to be

encouraged for this communication and one of the ways to do that is to offer them

autonomy in the task performance, as was described earlier. For enterprises that are

planning to rely at some extent on their own resources to support the FLOSS

migration, it is very important that they understand the conditions and process of

transformation through which their IT personnel would become part of the global

FLOSS community. Additionally, this process has specific challenges attributed to

the different organizational and social behaviours and rules that define the two sides.

Normally, in the very beginning they will have to adopt the standards of the

community first. Hawryszkiewycz (1999) defines this initial step as socialization,

when information is exchanged informally and enterprise personnel become familiar

with terminology and perspectives. Bergquist and Ljungberg (2001) characterize

exchanging of information as very important step for socialization and creating

alliances with the community. The social identification theory argues (Gefen and

Ridings, 2003) that people acquire values, attitudes and behaviour based on their

personal experience and social processes in which they participate. Sawyer et al

(1998) point out that skills needed for successful contribution in a centralized group

may be much different from those needed in a distributed group. Based on the

previous, it may be argued that after the socialization phase which establishes the

basis of communication, IT personnel has to be encouraged to develop their sense of

belonging to community groups.

65

Since software development in LSEs is traditionally centralized, the gap is

obvious. The essential layer for this to be possible to overcome is personal

communication skills. The transformation process is not easy for the IT personnel.

Research has shown (Ivancevich et al, 1983) that for IT personnel communication is

the most stressful job characteristic. As discussed before, organizational setting

influences on development of personal communication skills. It has been observed

that trying to force people to use or be involved in the development of FLOSS when

it is not their own choice is very difficult (McHugh and Donnelly, 2004; The MITRE

Corporation, 2003). The concept is simply better applicable if participation is

voluntary.

Experience from successful FLOSS migrations is that enterprise developers

have to change their attitudes and adopt new habits (Wagner, 2005; SoftwareMag,

2005). FLOSS also changes the way how software is developed (Sarma and Hoek,

2004; Greenemeier, 2005). Other factors besides intrinsic motivation also affect

developers in this transformation process, like age and gender, which were discussed

in-depth previously.

Figure 14: The layered behavioural model of software development.

(Source: Curtis et al, 1988 ; p.1269)

66

Figure 15: Single community built around source code

(Source: Gabriel and Goldman, 2002; p.6)

Last is the challenge of identity. If developers are given sufficient autonomy

and are encouraged to develop their individual participation in social networks, then

question may be raised how this affects their identity. Klenke (1996) addresses the

importance for IT employees not to lose their sense of identity of who they work for

and where they belong (Figure 14 and Figure 15). According to the social

identification theory (Gefen and Ridings, 2003), social identity is the self-perceived

participation in real or imagined social groups, named “ingroups”, which are distinct

from other groups, named “outgroups”. According to Lin (2004) enterprise

developers have a hybrid identity in the FLOSS social groups. If the duality is

supported by the enterprise and does not affect the social attributes of both groups,

then individual self-identity conflict can be avoided.

An interesting finding is that off-line communication can be additionally very

stimulating to enterprise developers and strengthen social bonds between them

(Takahashi et al, 2003; Sarma and Hoek, 2004; Lin, 2004). Erenkrantz and Taylor

(2003) found out that enterprise personnel shows clear bias towards communicating

with people who are close in proximity, rather than with remote peers. Scacchi

(2002b) points out that social interactions in FLOSS projects are rarely performed

off-line, through face-to-face communication. Even though at first this may seem

contradictory with the dual identity problem, the existing tension between the two

concepts may be released with improvement of communication techniques.

67

Weerawarana and Weeratunge (2003) put emphasis on the intrinsic sense of

belonging to a global community, which in some individuals propels imaginative

feeling of power. The sense of belonging may be differently structured even on level

of smaller groups. If the gap between groups becomes larger then the level of trust

becomes lower. In extreme cases even discrimination between groups is possible

(Eckert et al, 2005). Organizational issues can also cause much trouble for the

dualism of identity of enterprise IT personnel. The community groups that maintain

FLOSS projects are highly organized teams (Lawrie and Gacek, 2002; Gallivan,

2001; Heimer, 2003). But they do not have well defined boundaries and leadership is

pluralistic (Klenke, 1996). This can cause confusion to the involved enterprise

personnel if the organization of their working environment has opposite principles.

IT managers will have to find a way how to balance between the social

impact of FLOSS migration and the internal cultural dependencies. Communication

and control of activities (Garcia, 2001) are deeply linked to the organizational

structure of an organisation. Enterprises can not afford a spontaneous adjustment to

FLOSS practices of their IT personnel. This defines the desired position of IT

managers in migration processes. They have to keep the integrity of enterprise

resources, but as well have to be promoters of organizational change. Both tasks can

be accomplished successfully by following proper technical guidelines and by fairly

treating non-technical issues in the migration policy at the same time. Melian et al

(2002) claim that participation in discussions with the community will certainly bring

benefit for large enterprises in in-house software development.

In this section it was discussed how bonds with the community can be

established through socialization and developing a sense of belonging, while being

aware of the potential for inter-group conflicts. It was discussed that organizations

must take care for development of personal communication skills and gradual

involvement of IT personnel in the FLOSS community. It was explained that the

issue of dual identity has to be addressed with clear guidelines and policies.

68

4.2. End users and FLOSS migration

4.2.1. Role and profile of end users

This section will focus on the differences between end users and their role in

the FLOSS migration project.

Users have very important role in FLOSS migration projects. Ultimately, the

conclusion whether one migration was successful or not would depend on their

acceptance of the new system. Based on analysis of several case studies, Larsen et al

(2000) found out that user attitude is a critical factor in adoption of FLOSS. Similar

findings have been expressed by other studies (Yee, 2003; Drozdik et al, 2005;

Briggs and Peck, 2003; Vile, 2005; Zuliani and Succi, 2004). But internal end users

in LSEs do not inherently share same beliefs, values, norms and motives with the

FLOSS community.

The FLOSS community, at least at the present stage of its development, has

end users which are predominantly with IT background (Scacchi, 2004a; Scacchi,

2003). This is mostly a result of the types of existing FLOSS projects, the present

state of FLOSS development, but also reflects the social setting described previously,

with end users of FLOSS sharing same beliefs, values and norms with the creators of

these projects. As Ye and Kishida (2003) explain, they form a community of

practice, because they are informally bounded by their common interest and practice.

On the other side, development of software programs is neither of interest to

enterprise end users nor is their practice, therefore it is not realistic to expect high

levels of their involvement in the FLOSS community like that of the IT personnel.

End users have to get accustomed to new applications with FLOSS migration and in

that sense it is important to focus the analysis on how changes introduced with the

migration will impact on them.

69

Not all internal end users in enterprises are the same. Kryt (1980) divides end

users into three distinct subgroups: indirect end users, the most experienced;

intermediate end users, who are more experienced and often experts in their own

field; and direct end users, who do daily routine tasks of inputting data. The last are a

majority and have (Brunelle and Bruce, 2002) limited skills for using small number

of applications, usually ones that are standard. Furthermore, research shows (Davis et

al, 2000) that most end users have little interest in FLOSS, which supports the earlier

statement for what can be expected from them. And while (Kryt, 1980)

knowledgeable end users will not necessarily be friendly, cooperative and willing to

apply technology, these sophisticated users, also called champion users, are still a

precious resource in system development (Pliskin and Shova, 1987; Vile, 2005;

Briggs and Peck, 2003). Rayksund and Heltn (1988) claim that distributed expertise

can be accomplished by having key users at all levels. Finally, Ambati and Kishore

(2004) point out that the growth of the FLOSS community makes non-developers

more knowledgeable.

For planning the migration outcomes, it is important to anticipate the

structural and individual differences between end users. Different user groups inside

LSEs may have different impact on FLOSS migration projects. It is important to

understand the characteristics and needs of each user group before the project starts.

Otherwise users involved in pilot projects may not adequately represent end users

(Bostrom and Kaiser, 1981). De and Ferrat (1998) found out that those users who

have actively participated in the system development, share better opinion of the

project. As De and Ferrat (1998) claim, different groups perceive system problems

and recommend changes on the basis of their own perspectives. These findings

suggest that in large enterprises there is a risk of having distorted view of end users

and their specifics. Even though it is obviously a challenging task, any activity in the

direction of improving end user satisfaction should consider differences between user

groups (Shaw et al, 2002). Therefore with migration to FLOSS, IT personnel have to

deal with the group-specific problems of different end user groups. Joshi et al (1986)

suggest that IT personnel must avoid favouring any particular user group in order to

prevent conflicts.

70

In this section it was explained that it is important to anticipate differences

between end users, because they may have different impact. User attitude was

defined as a critical factor in adoption of FLOSS. It is explained that it is not realistic

to expect high involvement of end users in the community due to their little interest

in FLOSS. The importance of champion users, as well as the impact of differences

between user groups has also been addressed.

4.2.2. Importance of user training

The following text will discuss why user training is an important factor in

FLOSS migrations and how organizations should cope with the challenges associates

with it.

In order to increase the user’s acceptance (Han, 2003), organisations have to

support and encourage their employees to the information system in their work.

Niederman and Webster (1998) define the goal of training to produce a motivated

user who would continue to learn on the job. In that sense, training is very important

contributing factor to the success of a FLOSS migration project. Many research

studies have pointed out the positive influence of training (Kleintop et al, 1994;

Mahapatra and Lai, 1999; Fardal and Tollefsen, 2004; Shaw et al, 2002; Scott, 1995;

Olfman and Shayo, 1997). Training is essential for the switching to FLOSS (Yee,

2003; European Communities, 2003; WatchIT, 2004). Fardal and Tollefsen (2004)

explain that training can help to reduce computer anxiety but also stimulate

individual motivation in regards to IT.

Training has also many challenges of its own. It is also (Vile, 2005) one of

the top barriers of migration to FLOSS. Training does develop motivation to use the

system by building a mental model of it (Olfman et al, 1986). But this process is

based on many factors, like: organizational culture, personality of the trainees, social

ambient in the enterprise, etc. Gallivan (2000) points out that the role of user training

in facilitating organizational change induced by technology has often been

71

oversimplified. For example, if training displaces roles and responsibilities, it is

noticed that some end users do not transfer learned skills (Shayo and Olfman, 1993;

Shayo and Olfman, 1994).

It is also observed that in practice (e-Business Watch, 2002b) formal training

can only teach basic skills and enterprises consider learning on the job as very

important, especially for IT personnel (e-Business Watch, 2002a). Han (2003) points

out that informal training in form of knowledge sharing also increases willingness to

use a new system. Training on the job is often spontaneously supported by end users

themselves. Nilsen and Sein (2004) point out that end users first turn to their

colleagues for help because of social concerns and needs. And colleagues are the

ones who they would trust most, because they share same problems in the everyday

working life. End user training through knowledge sharing may be a powerful tool

for solving user issues during FLOSS migrations, because the concept is compatible

with the collaborative principles and tools offered by FLOSS.

There are also cultural differences in how end users get skilled and use

computers (e-Business Watch, 2002a). Those also have to be addressed carefully by

migration managers who have to carefully determine the user groups that have to be

influenced. Han (2003) suggests that training can be tailored for each user group

separately. Nelson et al (1995) recommend end user training to be performed through

cross functional cooperation, like with the human resource department. To become

closer to the online tools behind the FLOSS community, e-learning methods can be

also utilized to improve human resource management functions (e-Business Watch,

2002a). This supports the previous argument about training through knowledge

sharing.

During migration projects, communication between project members as well

as proper updating of the internal staff is of crucial importance. The best way to

disseminate information and enable communication for FLOSS migration projects is

to use the same technology which is know to the FLOSS community. In that sense it

is worth considering putting up information about the migration project and

supporting knowledge bases on a company intranet (Vile, 2005; European

Communities, 2003). If same communication technology is used like in the

72

community, later if the enterprise decides to contribute more, it would be easier for

the internal staff to accommodate to the unique collaboration techniques of FLOSS.

To summarize the main key points, training is important for FLOSS

migrations, but also brings many challenges. Besides formal training, enterprises

should promote training through knowledge sharing. Best conditions can be provided

by deploying the principles and techniques that are well known to the FLOSS

community.

4.2.3. Challenges of end users participation

This section explores the importance of end user participation in FLOSS

migration projects and the challenges of this process, primarily related to internal

power issues.

Usability is one of the top barriers to migrating to FLOSS (Vile, 2005).

FLOSS migration projects should be streamlined in direction to enhance the key

applications and provide better tools for the business processes, thus gain more

support from end users. And this is exactly the area through which end users can be

involved in the migration project, especially if changes of user interfaces are

expected. This is because resistance is higher with systems that are not user friendly

(Markus, 1983).

User participation is one of the most important ways of overcoming

resistance. Many research findings suggest that users should be involved in any

system development (Koop and Grant, 1993; Klenke and Kievit, 1992; Hodgson and

Aiken, 1998; Patricia, 2005; Kryt, 1980), and same applies to FLOSS migration

(Almond et al, 2004; European Communities, 2003). However (Klenke and Kievit,

1992; Butler and Fitzgerald, 1997) high end user involvement does not guarantee

project success. User involvement can be problematic in large enterprises and the

knowledge of end users does not guarantee that they will be willing to participate.

73

How users depend on the information system in their work (Jarvinen, 1983), affects

their desire to participate. Gefen and Ridings (2003) suggest that focusing on the

responsiveness of IT personnel instead, thus reducing group boundaries, can increase

user acceptance. According to the social identity theory, (Gefen and Ridings, 2003)

group boundaries can also be reduced through joint activities. Creating cross

departmental teams may be an elegant way to narrow the gap between users and IT

personnel.

Based on top of the above analysis, it can be suggested that user involvement

in the project is an important aspect of FLOSS migration, but not decisive. However,

by user involvement in the design process, not only better design could be provided

and resistance can be fought (Markus, 1983), but also political balance could be

achieved (Koop and Grant, 1993). The importance of IT responsiveness on end user

satisfaction means that inter-organizational balance of power between IT personnel

and end users is very important for the migration outcome. Koop and Grant (1993)

explain that unbalance between power and system design structures can lead to user

resistance or system failure. They also stress out that a structural view of power leads

to a better understanding of the collective resistance of groups involved in

implementation. Similar findings are expressed by Barker and Wright (1997) and

Markus (1983).

The work of Clement (1990) provides some logical explanations of these

phenomena. He claims that enterprise environments are not always characterized by

rational behaviour and are primarily social systems, driven among other things by

conflicting personal aspirations. End users behave as they do because that is their

survival tactics. This means that besides differences in individual attitudes, the inter-

group relationship between end users and IT department is crucial for outcomes of

projects. Furthermore (Koop and Grant, 1993), power in the structural sense reflects

properties of the organization as a social system and is vital for the success of the

project. This also raises questions whether migration of server infrastructures can be

more successful than applications with user interface, since issues of power conflict

are obviously less visible.

74

In this section it was argued that end user participation is very important for

fighting resistance, though it is not a decisive factor. It was discussed that creating

mixed teams may narrow the gap between users and IT personnel and that power

issues have to be properly addressed.

4.2.4. Impact of conflicts and internal power distribution on

the migration

The impact of internal conflicts and changes of internal power in respect of

FLOSS migration projects are discussed in this section.

The research of Warne (2003) indicates that not every type of conflict has

same impact on the successful progress of a project. Most significant threat

according to this research is posed by the conflict between users and developers and

between different user groups. Finally, past research shows (Koop and Grant, 1993)

that if a new system design changes the relationships and restructures power within

an organization, then it often results in user resistance. This finding correlates to one

of the main points drawn from the earlier analysis that IT personnel must avoid

favouring of any particular user group. Managers must try to streamline the FLOSS

migration in the direction which will preserve existing power structure in the

enterprise, or even change it in positive direction, especially if applications have to

be redesigned or new built from scratch, because a new system can produce loss of

power (Markus, 1983) for some and more power for others.

However, power is not (Clement, 1990) a personal matter but depends on

organizational relationships. For a successful FLOSS migration project, it is very

important what is the quality of the relationship between end users and the IT

personnel (Ramakrishna and Lin, 1999; Nagler, 2005). Leonard (2002) made even

more detailed analysis and claims that the quality of this relationship is directly

correlated with the maturity level of the end users. This finding positively supports

other papers (Zuliani and Succi, 2004; European Communities, 2003) which suggest

75

that the most skilled and enthusiastic users should be involved in greater extent in the

FLOSS migration. The ideal time for their initial involvement is pilot projects.

Nandhakumar and Jones (1997) claim that organizational power relations

may restrict interactions between users and developers. They present an important

finding that these interactions have to rely on the developers social skills, instead on

mere cooperation based attitudes. In that sense it is wrong to overemphasize the role

of end users in the FLOSS migration project. This statement is supported by the

research findings of Leonard (2002), but Alter (2000) stresses out that it is dangerous

to generalize it, since there are different types of users, organizations and projects.

These research findings suggest that end user involvement has to be analyzed with a

holistic approach. This also implies that if FLOSS migration succeeds to input new

values and norms in the organization especially for IT personnel, but as well for end

users, not only resistance may be suppressed, but support may be increased as well.

This analysis indicates that power issues have to be incorporated in the

strategy for management of human and social issues with migration to FLOSS.

Management has very important role in solving such problems. Mahring (1998)

points out that stable dominant coalition of line managers and IT managers can

sustain the organizational commitment to projects, since they have decisive

influence. It is important that line managers have well understood not only the main

purpose of the FLOSS migration project, but also the changes that it can impose to

their department and to the organization as a whole. As Sauer et al (1997) point out,

they need to perceive IT projects not as merely technical, but as organizational

initiatives. Managerial attitude towards FLOSS migration, especially in cases where

the migration initiative is started by IT departments, is crucial for the success of the

migration.

The individual characteristics of the IT manager strongly influences on the

acceptance of FLOSS (Koch, 2003). This job is not easy because many senior

executives view IT merely as a cost of doing business, (Hirschheim and Lacity,

2000) mostly because of its questionable value with lots of failed projects. Therefore

it is claimed that end user acceptance of new technology is greater when top

managers fully understand and support changes (Kleintop et al, 1994). The

76

management contributes significantly towards success of FLOSS migration projects

(KBSt, 2003). Sirkin et al (2005) and Briggs and Peck (2003) point out that if senior

executives do not communicate the need for change, the success of their projects

may be endangered. From the previous we understand that managers have key role in

the migration project and their impact is crucial. They may save the migration

project, but as well destroy it. Not all managers possess the same qualities.

Managers differ (Grosfeld and Roland, 1996) in their learning skills and some

are not able to take initiative in an unknown environment. Blanchard (2004) also

claims that most managers are not able to maximize the performance of their

employees. Sherer and Alter (2004) point out that this inadequacy in managers

should be treated as one of the risk factors in FLOSS migration projects. Due to fear

of disruption, line managers may also resist migration (Novell, 2005). In his

philosophical approach to innovation, Lessig (2001) stresses out that there are also

people for whom the idea of cultural production is only associated with the state or

the corporation. These people are almost impossible to change since their ignorance

is rooted very deep and are the most difficult to cope with during FLOSS migration

initiatives. The worst that can happen is these people to be managers. Based on the

whole discussion, it is understandable why IT managers (Hodgson and Aiken, 1998)

must also deal with the human issues of managing change, but as well must be aware

of their own weaknesses and ready to assess them.

As a summary, it was discussed how internal conflicts and redistribution of

power may have strong impact on FLOSS migration projects. It was pointed out that

significant threat is posed by conflicts between users and IT personnel, as well as

between different user groups, which may be reduced through user participation and

mixed teams. It was argued that power issues have to be addressed in the migration

policy and that management has a key role in resolving such issues.

77

4.2.5. Resistance to change

This section discussed how to fight resistance to change expressed by end

users during FLOSS migration projects.

FLOSS migration induces platform changes and changes in user interfaces. It

is very important that these changes are controlled and minimal, because they can

influence on user opinion (Figure 16). Dedrick and West (2004) point out that

research findings find availability of specific key applications as most important

platform adoption criteria for end users. Larsen et al (2000) point out that software

products are evaluated on how they fit into existing architecture and facilitate

business processes. In addition, they also argue that attitudes, weakly founded

assumptions and conservatism are a strong obstacle for user against a potential

switch away from COTS.

Figure 16: Knowledge domains involved in system building

(Source: Curtis et al, 1988; p.1273)

Change (Figure 17) is especially problematic in large, complex organizations

(Hodgson and Aiken, 1998). The problem presented by Neilson et al (2005) is that a

large number of LSEs suffer from “passive-aggressiveness”, while resistance to

changes and rigidness are deeply rooted in their histories. Opposite of them are

78

resilient enterprises. The second are obviously better candidates for FLOSS

migrations. In LSEs resistance is very likely to be encountered with such initiatives,

because large number of end users would be automatically be affected, especially if

desktop platforms are subject of migration.

Figure 17: Impact of changes

(Source: Almond et al, 2004; p.34)

Resistance to change has very powerful negative influence on FLOSS

migration projects. Undertaking such projects can easily increase the levels of stress

and frustration among end users (Drozdik et al, 2005; Novell, 2005; Northwest

Educational Technology Consortium, 2005). FLOSS migration case studies suggest

that there are little incentives for end users to switch from COTS to FLOSS (Dalle

and Jullien, 2001; Fitzgerald and Kenny, 2003; Miller, 2001). But not all users will

express their resistance or conformance to changes in the same way and with the

same intensity (Blanchard, 1992).

Individual differences between end users exist even in groups where users do

the same job and use the same computer resources (Nielsen et al, 1986). Patterns

may be observed of the behaviour of different user groups. The organizational setting

directly influences on the change acceptance level in end users (Markus, 1983; Han,

2003; Sauer et al, 1997). It has been observed that end users who are more

experienced with IT technology will have more positive attitude to changes in IT

systems than others (Han, 2003). Kleintop et al (1994) found out that if the end users

are familiar with the system to be implemented, their uncertainty felt about how their

work is going to change with it may be reduced. De and Ferrat (1998) found out that

end users of a system recommend more changes than developers. These findings

79

indicate that participation of end users, especially champions, in pilot projects, can

generate multiple factors with positive influence on the migration project. De and

Ferrat (1998) point out that this may be difficult in practice, thus careful plan of end

user involvement has to be prepared.

One of the goals of FLOSS migration should be to provide high level of user

satisfaction with the new system. Unfortunately, the effects of change during the

migration phase can not be stabilized overnight by management. Change processes

take time (Van de Loo, 2000). In contrast to that, resistance may come but may also

go away with change. The goal is (Markus, 1983) to avoid resistance if possible, and

to confront it constructively if not. Clear path to confront potential conflicting

situation is through open communication. Open communication may change personal

beliefs and attitudes. End users have to understand why a change has been initiated

and what the resulting effect is (Bruce, 2005). It is important to be able to persuade

end users that FLOSS migration is to be performed due to professional reasons and

strategic benefits for the enterprise and not due to ideology.

To summarize the key points in this section, as with any change FLOSS

migration may induce resistance on end users. This is more evident with changes in

user interfaces, thus it is important to keep these changes controlled and minimal. It

was argued that individual and group differences have to be considered and that open

communication is the key to resolving potential issues.

4.3. FLOSS migration steps

4.3.1. The benefits of staged migration

This section argues why staged migration is a better approach than full roll-

out to FLOSS platforms, from the perspective of the associated human and social

issues.

80

Assuming the management of a large enterprise has decided to migrate to

FLOSS, it is very important whether this migration will be partial or full. Partial

adoption is one of the possible outcomes for FLOSS platforms (Dedrick and West,

2003). Partial migration could either mean that the enterprise has decided to have

both proprietary and open source software solutions in its repository, or simply

makes one step forward towards full migration to FLOSS through a staged approach.

From a practical perspective, partial migration may also mean that some people in

the IT department may be engaged to work with FLOSS more than others.

Dedrick and West (2004) conclude that even though migration of the server

platform have important implications for the IT department, they are easier to be

decided, have little direct impact on the daily work of end users and is transparent to

them. Their research also shows that power and politics play a less important role in

the server platform adoption decisions. Almond et al (2004) point out that abstraction

from the client is the main advantage of server based computing. In case of Linux,

systems equipped with this OS (McDaniel, 2000) are excellent as file and print

servers for heterogeneous environments. User resistance may not be significant

(Nelson and Joshi, 1993) if a new or changed information system is introduced with

incremental changes. If the change is radical, a significant implementation effort may

be needed. It is also easier (Albrigo, 2005) to sustain corporate commitment by

incrementally developing complex systems.

In the FLOSS migration guideline of European Communities (2003) it is also

suggested that migration should be started first with servers, because it does not have

any adverse negative effects on the users. User resistance is only one of the risk

factors which in this case is non obtrusive. Wiederhold (2003) points out that small,

incremental and rapid upgrades reduce the risk. It is better to have slow and careful

roll-outs with FLOSS. Migrating to FLOSS platforms (Novell, 2004) by completing

well-coordinated stages minimizes operational risks and unnecessary disruptions

(Figure 18). In general, Winslow (2005) suggests starting with FLOSS adoption

among few particular areas in the enterprise IT environment where migration would

return most value with least effort and disruption.

81

Figure 18: Example of a Linux adoption roadmap

(Source: Deb and Srivastava, 2004; p.3)

Larsen et al (2000) observes that large organizations are willing to use

FLOSS partially and in few areas where it remains invisible to the corporate board.

Obviously FLOSS penetrates slowly in large enterprises. Most FLOSS technology

comes to companies (Lawton, 2002) after IT people have played with it and

gradually migrated it into the enterprise. Enterprise surveys show that there is even

unofficial experimentation with FLOSS inside enterprises (QNB Intelligence, 2004;

Bedell, 2005). Dedrick and West (2004) confirm that FLOSS is more likely to be

adopted if tried and assimilated gradually over time. In his guidelines for FLOSS

migration, Vile (2005) suggests that the migration should be done in stepwise

manner, targeting certain departments. Big bang migrations (European Communities,

2003) should be avoided because of the big number of variables that have to be

controlled.

Staged migration through the minimized risks has also economic advantages.

Deb and Srivastava (2004) point out that staged migration will progressively reduce

total cost of ownership and incrementally maximize the return on investments.

Staged migrations also provide enough time for the IT personnel to incrementally

study the new FLOSS platform. Server migrations (Winslow, 2005) should be done

82

first because they would be transparent to end users, while the IT staff would have a

chance to learn the FLOSS products. Deb and Srivastava (2004) confirm that server

platforms, but also typical batch and computation oriented applications can be

migrated first. Unlike with majority of COTS systems, migration to FLOSS

platforms can be performed incrementally on a service by service basis (Robert and

Schutz, 2001).

The conclusion of the above analysis is that staged FLOSS migration has

multiple advantages over full roll-outs, such as: lower risk, easier adoption among

end users, more available time for IT personnel to learn new technologies, lower

potential for internal conflicts and economic benefits in terms of progressive total

cost of ownership.

4.3.2. The impact of staged migration on end users

In the following paragraphs the focus is on how staged migration affects end

users and the social groups they belong to internally in LSEs.

Hirschheim and Klein (1989) point out that systems emerge through social

interaction and that evolutionary learning from participating in partial

implementations is the way technology becomes embedded into the social

perception. As social perception evolves over time, the adoption of next FLOSS

migration phases will gradually stabilize and resistance outbreaks will become less

severe and controllable. From a technical point of view, staged migration to FLOSS

brings less risk than a full roll-out. The first stage would be the change of the server

infrastructure, the layer which is invisible to end users. In the second stage the

developers would switch to multiplatform development tools while retaining the OS

platform. In the third stage the OS platform would be replaced. This gradual change

would help both IT personnel and end users become acquainted with FLOSS and

have chance to see what the concept is all about and how the company can benefit

from it.

83

Pilot projects are very important and they can increase confidence in the new

platform (European Communities, 2003; Deb and Srivastava, 2004; KBSt, 2003;

Almond et al, 2004). Pilot FLOSS migration projects can be used as proof of concept

for next stages (Briggs and Peck, 2003; Sliwa, 2005). Experimentation with FLOSS

may be encouraged, but enough slack time needs to be provided (Dedrick and West,

2004). Slack time is also important because it can lower the stress levels and improve

motivation. Slack time can be provided with pilot projects and that is why the pilot

stage is a funding step for business-critical projects (Schadler, 2004). Almond et al

(2004) suggest that end user surveys should be used for planning a pilot migration

project for the purpose of analysis of client application usage patterns.

A practice phase (Kleintop et al, 1994) is needed not only for building system

specific skills, but for creating competent users as well (Paul, 2005). Large scaled

migrations should be avoided, instead FLOSS migration should be performed in

step-wise manner. Large projects are inherently more complex (Nelson and Joshi,

1993) and always bring big risk to the enterprise. During the FLOSS migration,

enterprises may find out that some applications can not be migrated easily. It is an

opportunity for enterprises to change the technology behind. Almond et al (2004)

suggest that the best way to do this is to use intranet technology and convert these

applications to portal-based. Developed in that way, they will not be a major problem

in any future migrations. And a familiar user interface would be offered to end users.

The most critical phase is when the enterprise has already taken advantage of

the server platforms and plans to upgrade the desktop. Since significant changes will

be visible to end users, the end user resistance as a risk factor becomes more

important to be mitigated. When moving to much visible high-profile desktop

systems (Fitzgerald and Kenny, 2003), top management support is critical for FLOSS

deployment. Weerawarana and Weeratunge (2003) explain that there are three

market segments of FLOSS: server operating systems and applications,

desktop/client operating systems and applications, and enterprise solutions. The first

segment relates to an area which does not impose huge impact on the existing user

interfaces, while the last two do that. Following this categorization, after performing

successful pilots enterprises can perform FLOSS migration in the technical areas

84

which are less visible to end users, but as well to upper management decisions, like

server infrastructures.

In the text above it was argued that staged migration positively affects end

users and reduces resistance. It was elaborated how pilot projects increase confidence

as proof of concept and how they help creating competent users. It was also pointed

out that the most critical phase of FLOSS migration is the upgrade of desktop.

4.3.3. End users and desktop migration

The following analysis is about how desktop migration to FLOSS platform

affects end users.

The advanced stage of FLOSS migration should include more visible

applications, like desktop operating systems and enterprise applications. As

discussed earlier, users in LSEs usually tend to strongly react to these changes, either

positively or negatively. As it was discussed previously, FLOSS migration can be

used to change and enhance enterprise applications, especially in the key areas.

Portal based applications will ensure smoother future upgrades. User involvement is

very important in the design of the solutions, with emphasis on retaining the existing

power distribution structure inside the organization.

Regarding desktop operating systems the situation is different, because they

are accepted merely as commodities. This claim is supported by the findings of

Spiller and Wichmann (2002) and Dedrick and West (2003) that enterprises consider

the availability of FLOSS source code as important criteria for FLOSS adoption in

all usage areas except on desktops. This suggests that FLOSS desktop operating

systems are often used as pure replacement to existing commercial operating systems

in enterprises.

85

The core reasons why end users react to changes of desktop operating

systems have to be well understood, because they should be the foundation behind

the principles in the desktop migration policy. Apparently on desktop (Schmitz,

2001), the adoption of FLOSS solutions cannot be invisible to end users. Almond et

al (2004) claim that a radical change in the desktop interface will cause various kinds

of reactions: from enthusiastic acceptance to outright resistance. There is evidence

(Zuliani and Succi, 2004; Fitzgerald and Kenny, 2003) that FLOSS can be

successfully deployed for desktop. Fitzgerald and Kenny (2003) conclude that even

though the functionality provided by FLOSS desktop products is very similar, end

users prefer an identical interface. Still, majority of end users (Schmitz, 2001) appear

to be unwilling to learn a new FLOSS system with no clear benefit for them and with

interoperability problems with various document types which may negatively affect

their job effectiveness.

The crucial player in the FLOSS market for desktop operating systems is by

no doubt Linux (Relevantive, 2003). It is also one of the strongest symbols of the

FLOSS community and probably the biggest FLOSS project so far, both in terms of

people involved and in terms of complexity. Relevantive (2003) have performed a

unique empirical study on a random sample, with interesting findings about how

Linux is accepted by new users, compared side by side with the acceptance of

proprietary Windows operating system. This study shows that the usability of Linux

as a desktop system was judged to be nearly equal to Windows and had no

significant effect on user performance. The charts show similar distributions of

answers for both systems (Figure 19 and Figure 20). The authors conclude that a

migration to Linux would be positively accepted by employees. This is confirmed by

the study of Olfman and Shayo (1997) who found out that if same task is performed

with two related but different software applications, users will most likely capture the

general rules common to both applications.

86

Figure 19: Survey on Linux usability: Level of enjoyment

(Source: Relevantive, 2003; p.75)

Figure 20: Survey on Linux usability: Time needed for achieving competence

(Source: Relevantive, 2003; p.76)

A possible indication from these findings is that the resistance of end users

during FLOSS migration of desktop environment may be partially attributed to

intrinsic factors like individual level of playfulness and conformance, but in

accordance with the previous analysis of the role of end users in the migration

process, this more likely depends on different external factors such as personal

87

motivation with present job, unequal power distribution between employees, low

communication level inside the organization, inter-organizational conflicts, etc. At

the end, users are forced to use the new system and are not offered a choice, which

may trigger reaction. These are the reasons why (Relevantive, 2003) the personnel

department, managers as well as the employees themselves have to be part of the

migration process. It is worth to point out the claim of Rossi et al (2005) that after

FLOSS migrations all users have a clear opinion about FLOSS, while those that were

neutral have split in either positive or negative oriented camps. Enterprises should

take special care to minimize the second one, following the principles discussed

throughout this study.

To summarize the above discussion, majority of users do not easily accept

changes of desktop. Even though studies show that FLOSS desktop platforms may

offer similar user experience like mainstream commercial OS platforms, the

resistance may be partially attributed to internal organizational problems and specific

social settings.

4.4. Summary

In this chapter a critical evaluation of non-technical FLOSS migration was

performed through the frame defined with the three main viewpoints on the problem

domain: issues related to IT personnel, issues related to end users and issues with

staged migration.

In the first part it was discussed about issues related to IT personnel. It was

found out how low vendor support for FLOSS induces self support in LSEs. The

impact of this is that enterprises begin to link with the FLOSS community. It was

discussed how training is important for adoption of FLOSS concepts, even though IT

personnel might react differently on FLOSS migration efforts, especially because of

the strong impact of personal career development. The influence of intrinsic and

extrinsic motivation was also elaborated. The role of job satisfaction, fun and

88

additional incentives for better acceptance of FLOSS was analyzed further, along

with the role of factors like age and gender. Afterwards it was discussed how

technical skills and peer recognition are important factors and how peer recognition

improves knowledge sharing. After that an analysis followed on how development of

sense of belonging to a community is a precondition for successful FLOSS projects

and can be encouraged through autonomy and internal collaboration practices. It was

explained that ideological frameworks can be an obstacle in this process and that

there is need for wider internal cultural changes in LSEs in order to enable this

ideological transformation. In a separate section the importance of online

communication skills and their encouragement was elaborated, in respect of the risks

that come along. Afterwards it was argued that for successful FLOSS migrations,

LSEs have to change their internal culture through empowerment and autonomy,

supported by proper open minded management. At the end it was pointed out that the

link with the community brings new problems, such as dualism of identity and that it

is best to perform changes gradually.

In the second part the issues related to end users were elaborated. It was

explained why it is important to anticipate the internal differences and that user

attitude is a critical factor in adoption of FLOSS. It is explained that high

involvement of end users is not likely, but that champion users are valuable assets

during migration projects. Then the importance of training was defined, along with

the challenges that it brings. It was suggested that training of end users be done with

the techniques well known to the FLOSS community. Later it was discussed how

internal conflicts and redistribution of power may have bad impact on FLOSS

migration projects. It was argued that these have to be addressed in a migration

policy and closely monitored by management. At the end of this part it was

concluded that controlled and minimal changes may reduce resistance from end users

and that open communication is the key to resolving such issues.

In the third and last part it was discussed that there are many benefits of

staged FLOSS migration, like: lower risk, easier adoption, more slack time, less

conflicts and economic benefits. In addition to that it was pointed out that staged

migration positively affects end users and reduces resistance. It was elaborated how

pilot projects may increase confidence and develop competent users. It was also

89

pointed out that the most critical phase of the migration is the upgrade of desktop,

because majority of users do not easily accept these kinds of changes. It was

concluded that even though research shows little dissimilarity between user

experience with FLOSS and COTS platforms, resistance may be partly attributed to

internal organizational and social problems.

90

Chapter 5. Conclusion

5.1. Discussion of the findings

The main goal of this research project was to investigate the non technical

issues that are experienced by large commercial enterprises (LSEs) when they carry

out FLOSS migrations. In order to accomplish the research objectives, the research

analysis was based on three viewpoints that construct the outline of the conceptual

framework in this study: human and social issues related to IT personnel, issues

related to end users and issues with staged migration. The findings of the study have

been expressed in the previous chapters throughout the analysis of these viewpoints

by extended review of literature from different disciplines and secondary data found

in it. These findings will be further discussed in the text that follows, in respect of the

above defined categorization and the set up research questions.

The human and social issues that are found in FLOSS migration projects are

similar to the ones found in migrations of commercial platforms, but there are also

differences. It was found that FLOSS migration may introduce collaborative

practices that did not exist in the enterprise before or positively influence on already

existing practices. It was pointed out that FLOSS migration projects have the

capability to steer innovation in enterprises. Positive changes introduced with FLOSS

migrations in enterprises may affect both individual and group performance and may

also change the attitude towards innovativeness. As it was explained, case studies

from practice prove that this happens in both directions, i.e. not only early adopters

are more likely to migrate to FLOSS, but FLOSS can also inspire innovation and

positive organizational change.

It was explored how the lack of vendor support motivates and forces LSEs to

learn how to self support themselves. The different support options were discussed

and it was suggested that enterprises have to learn how to help themselves with

91

FLOSS migrations and approach to the community in order to provide appropriate

support for the new systems. This is the initiation phase when the internal IT

personnel establish the starting links with the FLOSS community. It was explored

how lurking is especially important in this phase, because it disseminates knowledge

in-house. It was also elaborated that FLOSS introduces collaborative practices like

peer review, activity which promotes knowledge sharing but might not be at a

compatible level with the practice of code sharing in-house in enterprises, or the

level of autonomy given to the personnel. Implementation of collaboration practices

through empowerment and autonomy narrows the gap between IT personnel and the

community (Figure 21). It was pointed out that adoption of FLOSS has impact on the

internal culture, but also that adopting collaboration practices of FLOSS might ease

migration efforts. The difference between different IT jobs was explored and it was

shown that developers and administrators would not perceive in the same way the

FLOSS migration. Developers are more ready to accept new technologies, while both

groups are equally concerned about their career path. These differences have to be

addressed in the migration policy.

Figure 21: Influence of extrinsic and intrinsic factors on migration outcomes

92

It was discussed in detail how intrinsic or extrinsic individual attitudes can

affect the responsiveness to the FLOSS migration initiative, manifested through

changed level of enjoyment of work, concerns about personal career, etc. The role of

different factors like age and gender on FLOSS adoption was also presented. It was

found out that older employees are less likely to adopt FLOSS concepts, while

younger employees would favour the adoption if it brought challenge for them.

Regarding gender issues, enterprises must work on creating a gender neutral

environment while linking with the FLOSS community. But there are external

difficulties since the FLOSS community is predominantly male, at least for now.

Personal ideological differences may be impediments to the process and they

have to be mitigated by IT management with other incentives. These incentives have

to be tailored to the specific situation and persons, but in general the financial

incentive is proven to be one of the strongest motives for IT personnel to accept

changes and has to be considered as such. Building positive attitude for the FLOSS

concept can be performed through training. It was shown that best results will be

gained with group approach. In combination with peer promotion and knowledge

sharing, best conditions would be set for the transformation of IT personnel. Another

way of introducing these new practices is through adoption of the FLOSS

development model inside the enterprise before the migration, although this approach

might not be appropriate for all enterprises.

The main focus was put on how to identify and resolve the differences

between the internally shared norms, beliefs and attitudes of the IT personnel and

those of the FLOSS community. The successful migration and post-migration period

may be an adjustment phase for the IT personnel, in which they would start accepting

the norms, values and beliefs of the community. The best approach to do this is

through staged migration, because it was proven that gradual changes induce less

resistance to changes. It was also pointed out that managers have to plan how to

improve communication skills in IT personnel, especially online skills.

Communication skills are essential for providing high quality of support to end users

and for effective involvement in the FLOSS community. There are associated risk

factors with this advanced communication and it was pointed out that they have to be

effectively mitigated in a security policy. Each employee has to be given clear

93

guidelines about what is the boundary between the enterprise activities and the

community. This approach is also needed to prevent identity conflict of IT personnel,

during their deeper involvement in the FLOSS community.

Regarding end users, it was explained that they need to be early involved in

the migration project and that special attention should be given to the most skilled

users, especially in areas where changes are expected to be highly visible, like

desktop operating systems. It was described that these users have to be motivated to

disseminate the knowledge to their colleagues and that the process has to be

monitored effectively by their managers. Open communication supported by top

management and active participation of users in mixed teams are the main

preconditions for successful FLOSS migration project and fighting resistance to

change. The importance of the responsiveness of the IT personnel was also

explained, since they are the ones who may bring more closely the FLOSS practices

to end users. It is important that the IT personnel believe in these practices, thus it is

best to undertake staged migration with pilot projects, because that would give

enough slack time to the IT personnel to learn more about FLOSS.

It was described how these issues may vary across different organizations and

that local cultural contexts are very important for FLOSS migration projects. It was

shown that the structure of IT personnel in LSEs may also resemble different

patterns. It was pointed out that those organizations, for which IT is close to the

strategic interests, appear to be more innovative and are more open for emerging

organizational frameworks which are more suitable for successful FLOSS

migrations. Therefore, even though the outlined conceptual framework elaborated in

this discussion offers some general propositions, a holistic view of the specifics of

the particular local context must be taken into account during the creation of a

migration policy. It is suggested the migration policy to include assessment of the

current position of IT personnel and enterprise as a whole, in relation to the norms,

values and beliefs of the FLOSS community. Once the gap between the standards of

the in-house team and the community is identified, the enterprise should prepare a

strategy for gradual cultural transformation with the full commitment of

management.

94

5.2. Recommendations for future research

The result of this study is an outline of a conceptual framework that presents

what are the non-technical issues with FLOSS migrations and what are their

dependencies. This framework is elementary and may be further explored and

extended in some future research. Therefore the main recommendation for future

research would be to conduct first a test of this framework, possibly in a separate

study and later to expand it with new insights.

Based on the detailed analysis of the boundaries of existing research literature

in the problem domain and based on the findings of this study, some of the research

topics that have not been subject to analytical observation before and are

recommended to be explored in some future research through expansion of the

conceptual framework, are as follows:

� Determining the difference in the levels of FLOSS adoption between

administrators and developers in LSEs.

� Exploration of structural patterns of personality typology of IT personnel in

LSEs that have performed successful FLOSS migration, and their correlation

with company maturity levels.

� Exploring the dynamics of collaborative practices in LSEs before and after

FLOSS migrations.

� Research on the impact of cultural differences on the success of FLOSS

migrations.

� Conducting comparative analysis of FLOSS migrations in LSEs in EU and

Eastern European countries.

� Exploring the relationship between staged migrations and the dynamics of

playfulness found in IT personnel.

If the research area of FLOSS is expanded with the above given

recommendations for future research, a solid theoretical foundation will be

95

established for better understanding of some unique attributes of the non-technical

issues that are present with FLOSS migrations in LSEs.

96

References

Agarwal, R. & Sambamurthy, V. (2002). “Principles And Models For Organizing

The IT Function”. MIS Quarterly Executive, 1(1), 1-16.

Agostino, D. (2005). “Open Source Turns Strategic”. CIO Insight [Online], 23

November 2005. http://www.cioinsight.com/article2/0,1540,1893081,00.asp

[Accessed 7 December 2005].

Albrigo, E. (2005). “Taming The Big Technology Project”. Mortgage Banking,

October 2005, 151-157.

Almond, C., Cannon, A., van Hoof, J., Mark, O., Patsch, C., Schwaller, T. &

Vaddadi, S. (2004). Linux Client Migration Cookbook: A Practical Planning and

Implementation Guide for Migrating to Desktop Linux. pp.27-36. International: IBM

Redbooks.

Alter, S. (2000). “Same Words, Different Meanings: Are Basic IS/IT Concepts Our

Self-Imposed Tower of Babel?”. Communications of AIS, 3, article 10, 1-89.

Ambati, V. & Kishore, S. P. (2004). “How can Academic Software Research and

Open Source Software Development help each other?”. In: Feller, J., Fitzgerald, B.,

Hissam, S. & Lakhani, K. (eds) Collaboration, Conflict and Control. Proceedings of

the 4th Workshop on Open Source Software Engineering. 21 May 2004, Edinburgh,

Scotland, UK. pp. 5-8. ACM Press, New York.

Augustin, L., Bressler, D. & Smith, G. (2002). "Accelerating software development

through collaboration". In: Proceedings of 24th International Conference on

Software Engineering. 19-25 May 2002, Orlando, Florida, USA. pp. 559-563. ACM

Press, New York.

Australian Government Information Management Office. (2005). A Guide To Open

Source Software for Australian Government Agencies. Melbourne: Australian

97

Government Information Management Office. (Report as complementary document

to A Guide to ICT Sourcing).

Bac, C., Berger, O., Deborde, V. & Hamet, B. (2005). “Why and how to contribute

to libre software when you integrate them into an in-house application?”. In: Scotto,

M. & Succi, G. (eds) Proceedings of the 1st International Conference on Open

Source Systems. 11-15 July 2005, Genova, Italy. pp. 113-118. University of Genova,

Genova.

Baldi, S., Heier, H. & Bicher, A. M. (2003). “Open Courseware and Open Source

Software Learning from experience?”. Communications of the ACM, September

2003, 46(9), 105-107.

Balduino, R. (2005). “Succeeding with Open Source”. IBM [Online], 15 May 2005.

http://www-28.ibm.com/developerworks/rational/library/may05/reader/balduino.html

[Accessed 7 August 2005].

Barahona, J. M., Lopez, L. & Robles, G. (2004). "Community structure of modules

in the Apache project". In: Feller, J., Fitzgerald, B., Hissam, S. & Lakhani, K. (eds)

Collaboration, Conflict and Control. Proceedings of the 4th Workshop on Open

Source Software Engineering. 21 May 2004, Edinburgh, Scotland, UK. pp. 43-47.

ACM Press, New York.

Barker, R. M. & Wright, A. L. (1997). “End User Computing Levels, Job Motivation

And User Perceptions Of Computing Outcomes: A Field Investigation”. In:

Niederman, F. (ed.) Proceedings of the 1997 ACM SIGCPR conference on Computer

personnel research. 03-05 April 1997, San Francisco, California, USA. pp. 224-233.

ACM Press, New York.

Barton, P. & Nissanka, V. (2001). "Innovation v. Suppression: An English View of

Open Source Software Licensing". Journal of Internet Law, 4(12), 16-20.

Batler Group. (2005). “Application Development Strategies”. Batler Group [Online].

http://www.butlergroup.com/reports/ads/mansum.asp [Accessed 20 July 2005].

98

Bedell, K. (2005). “Five things execs want to see from open source in 2005”. IT

Manager's Journal [Online], 19 January 2005.

http://analysis.itmanagersjournal.com/article.pl?sid=05/01/18/210233 [Accessed 20

July 2005].

Benkler, Y. (2001). “Through the Looking Glass: Alice and the Constitutional

Foundations of the Public Domain”. In: Conference on the Public Domain. 9-11

November 2001, Durham, NC, USA. pp. 192-238. Duke University School of Law,

Durham.

Bergquist, M. & Ljungberg, J. (2001). “The power of gifts: organizing social

relationships in open source communities”. Information Systems Journal, 11(4),

305–320.

Bezroukov, N. (1999). ” Open Source Software Development as a Special Type of

Academic Research”. FirstMonday [Online], 4(10).

http://firstmonday.org/issues/issue4_10/bezroukov/ [Accessed 7 August 2005].

Bianchi, G. (1996). “Double Shift: Towards a convergence between spatial systems

of small and large enterprises”. In: 36th European Congress ETH Zurich. 26-30

August 1996, Switzerland. pp. 1-17. European Regional Science Association, EU.

Blanchard, K. & Blanchard, M. (2005). “Retain Key Talent”. Leadership Excellence,

Jul 2005, 22(7), 7-7.

Blanchard, K. (1992). “The Seven Dynamics Of Change”. Executive Excellence,

7(6), 5-6.

Blanchard, K. (1996). “Empowerment is the Key”. Quality Digest Magazine

[Online], April 1996. http://www.qualitydigest.com/apr/blanchrd.html [Accessed 22

November 2005].

99

Blanchard, K. (2004). “Leadership and the Bottom Line”. Executive Excellence,

September 2004, 21(9), 18-18.

Bleek, V. G. & Finck, M. (2004). “Migrating a Development Project to Open Source

Software Development”. In: Feller, J., Fitzgerald, B., Hissam, S. & Lakhani, K. (eds)

Collaboration, Conflict and Control. Proceedings of the 4th Workshop on Open

Source Software Engineering. 21 May 2004, Edinburgh, Scotland, UK. pp. 9-13.

ACM Press, New York.

Bonaccorsi, A. & Rossi, C. (2003). “Altruistic individuals, selfish firms? The

structure of motivation in Open Source software”. FirstMonday [Online], 9(1).

http://firstmonday.org/issues/issue4_10/bezroukov/ [Accessed 19 August 2004].

Bonaccorsi, A. & Rossi, C. (2004). “Contributing to OS Projects. A Comparison

between Individual and Firms”. In: Feller, J., Fitzgerald, B., Hissam, S. & Lakhani,

K. (eds) Collaboration, Conflict and Control. Proceedings of the 4th Workshop on

Open Source Software Engineering. 21 May 2004, Edinburgh, Scotland, UK. pp. 18-

22. ACM Press, New York.

Bonaccorsi, A. & Rossi, C. (2004). Contributing to the common pool resources in

Open Source software. A comparison between individuals and firms [Online].

Cambridge, MA, USA: MIT Sloan.

http://opensource.mit.edu/papers/bnaccorsirossidevelopers.pdf [Accessed 19 August

2004].

Bonaccorsi, A. & Rossi, C. (2005). “Intrinsic Motivations and Profit-oriented Firms

in Open Source Software. Do firms practise what they preach?”. In: Scotto, M. &

Succi, G. (eds) Proceedings of the 1st International Conference on Open Source

Systems. 11-15 July 2005, Genova, Italy. pp. 241-245. University of Genova,

Genova.

Bostrom, R. P. & Kaiser, K. M. (1981). “Personality Differences Within Systems

Project Teams: Implications For Designing Solving Centers”. In: Awad, E. M. (ed.)

100

Proceedings of the eighteenth annual computer personnel research conference. 04-

05 June 1981, Washington, USA. pp. 248-285. ACM Press, New York.

Briggs, J. & Peck, M. (2003). QinetiQ Analysis of Open Source Solution

Implementation Methodologies – QOSSIModo. UK: QuinetiQ.

Bruce, K. (2005). “Embracing Change for the Better". Foodservice equipment and

supply, September 2005, 86-86.

Brunelle, M. D. & Bruce, B. C. (2002). “Why free software matters for literacy

educators”. Journal of Adolescent & Adult Literacy, March 2002, 45(6), 514-518.

Butler, T. & Fitzgerald, B. (1997). "A case study of user participation in the

information system development process". In: Proceedings of the eighteenth

international conference on Information systems.14-17 December 1997, Atlanta,

Georgia, USA. pp. 411-426. Association for Information Systems, Atlanta.

Ciborra, C. U. & Andreu, R. (2001).” Sharing knowledge across boundaries”.

Journal of Information Technology, 2001, 16, 73–81.

Clement, A. (1990). “Cooperative Support for Computer Work: A Social Perspective

on the Empowering of End Users”. In: Proceedings of the 1990 ACM conference on

Computer-supported cooperative work. 07-10 October 1990, Los Angeles, CA,

USA. pp. 223-236. ACM Press, New York.

Conlon, P. & Carew, P. (2005). “A Risk Driven Framework for Open Source

Information Systems Development”. In: Scotto, M. & Succi, G. (eds) Proceedings of

the 1st International Conference on Open Source Systems. 11-15 July 2005, Genova,

Italy. pp. 200-203. University of Genova, Genova.

Cosovanu, C. (2003). “Piracy, Price Discrimination, And Development: The

Software Sector In Eastern Europe And Other Emerging Markets”. AIPLA Quarterly

Journal, 165, 167-221.

101

Curtis, B., Krasner, H. & Iscoe, N. (1988) "A Field Study Of The Software Design

Process For Large Systems". Communications of the ACM, 31(11), 1268-1287.

Daffara, C., Barahona, J. M., Carlos, R. J., Humenberger, E., Koch, W., Lang, B. &

Laurie, B. (2000). Free Software / Open Source: Information Society Opportunities

for Europe?. European Commission: Working Group on Libre Software. (Technical

Report).

Dahlander, L. (2004). Appropriating the Commons: Firms in Open Source Software

[Online]. Cambridge, MA, USA: MIT Sloan.

http://opensource.mit.edu/papers/dahlander2.pdf [Accessed 24 August 2004].

Dalle, J. M. & Jullien, N. (2001). “Open Source vs. Proprietary Software”. European

Summer School on Industrial Dynamics. 2 - 9 September 2001, Cargese, France.

David, P. A., Waterman, A. & Arora, S. (2003). Floss-US: The Free/Libre/Open

Source Software Survey For 2003. Stanford: Stanford Institute for Economic Policy

Research.

Davis, M., O’Donovan W., Fritz, J. & Childress, C. (2000). “Linux and Open Source

in the Academic Enterprise”. In: Building the future. Proceedings of the 28th

SIGUCCS Conference on User Services. 29 October - 01 November 2000,

Richmond, USA. pp. 65-69. ACM Press, New York.

De, P. & Ferrat, T. W. (1998). “An Information System Involving Competing

Organizations”. Communications of the ACM, 41(12), 90-98.

Deb, S. & Srivastava, M. K. (2004). "Managing Linux Migration". Infosys, Cutting

Edge, March 2004 Issue.

Dedrick, J. & West, J. (2003). “Why firms adopt open source platforms: A grounded

theory of innovation and standards adoption”. MISQ Special Issue Workshop, 236-

257.

102

Dedrick, J. & West, J. (2004). “An Exploratory Study into Open Source Platform

Adoption”. In: Proceedings of the 37th Hawaii International Conference on System

Sciences. 5-8 January 2004, Big Island, HI, USA. pp. 1-10. IEEE Computer Society,

Washington.

Dempsey, B. J., Weiss, D., Jones, P. & Greenberg, J. (2002). “Who Is An Open

Source Software Developer?”. Communications of the ACM, 45(2), 67-72.

Dickson, C. M. (1969). “In-House Recruiting – One Answer To Programmer Force

Losses”. In: Fein, M. (ed.) Proceedings of the seventh annual conference on

SIGCPR. 19-20 June 1969, Illinois, USA. pp. 120-127. ACM Press, New York.

Dinkelacker, J. & Garg, P. K. (2001). Corporate Source: Applying Open Source

Concepts to a Corporate Environment. HP: Hewlett-Packard Labaratories. (Tech

Report No. HPL-2001-135)

Domenick, J. (2004). “A, B, and C Players: Appreciating the Differences”. Baudville

[Online], 15 December 2004. http://www.baudville.com/articles.asp?a=35 [Accessed

20 August 2005].

Domino, M. A., Collins, R. W., Hevner, A. R. & Cohen, C. F. (2003). “Conflict in

Collaborative Software Development”. In: Proceedings of the 2003 SIGMIS

conference on Computer personnel research. 10-12 April 2003, Philadelphia,

Pennsylvania, USA. pp. 44-51. ACM Press, New York.

Dravis, P. (2003). Open Source Software: Perspectives for Development. The World

Bank: Global Information and Communication Technologies Department, infoDev

Program. (Report No. 27680).

Drozdik, S., Kovacs, G. L. & Kochis, P. Z. (2005). “Risk Assessment of an Open

Source Migration Project”. In: Scotto, M. & Succi, G. (eds) Proceedings of the 1st

International Conference on Open Source Systems. 11-15 July 2005, Genova, Italy.

pp. 246-249. University of Genova, Genova.

103

Dubie, D. (2005). "Getting there: Migrating to open source". Network World

[Online], 04 July 2005.

http://www.networkworld.com/supp/2005/opensource/070405-open-source-

migration.html [Accessed 19 July 2005].

e-Business Watch. (2002). ICT & e-Business in the Electronics Industry.

Bon/Brussels: The European e-Business Market Watch. (No.11/October 2002).

e-Business Watch. (2002). ICT & e-Business in the Metal Products Industry.

Bon/Brussels: The European e-Business Market Watch. (No.9/October 2002).

Eckert, D., Koch, S. & Mitlohner, J. (2005). “Using the Iterated Prisoner's Dilemma

for Explaining the Evolution of Cooperation in Open Source Communities”. In:

Scotto, M. & Succi, G. (eds) Proceedings of the 1st International Conference on

Open Source Systems. 11-15 July 2005, Genova, Italy. pp. 186-191. University of

Genova, Genova.

Egri, M. (2001). “Corporate Competitiveness – Conclusions of Some Investigations”.

Periodica Polytechnica, 2002, 10(2), 285–301.

Eilola, J. J. (2002). “Open Source Basics: Definitions, Models, and Questions”. In:

Proceedings of the 20th Annual International Conference on Computer

Documentation. 20-23 October 2002, Toronto, Ontario, Canada, pp. 79-83. ACM

Press, Toronto.

Elliott, M. & Scacchi, W. (2003). "Free Software Developers as an Occupational

Community: Resolving Conflicts and Fostering Collaboration". In: Schmidt, K.,

Pendergast, M., Tremaine, M. & Simone, C. (eds) Proceedings of the 2003

International ACM SIGGROUP Conference on Supporting Group Work.GROUP

2003. 9-12 November 2003, Sanibel Island, Florida, USA. pp. 21-30. ACM Press,

New York.

104

Elliott, M. & Scacchi, W. (2003). Free Software: A Case Study of Software

Development in a Virtual Organizational Culture. University of California: Institute

for Software Research. (ISR Technical Report No. UCI-ISR-03-6)

Elliott, M. S. & Scacchi, W. (2003). “Free Software Development: Cooperation and

Conflict in A Virtual Organizational Culture”. In S. Koch (Ed.), Free/Open Source

Software Development: IDEA Publishing. 2004. pp. 1-39.

Erenkrantz, J. R. & Taylor, R. N. (2003). “Supporting Distributed and Decentralized

Projects: Drawing Lessons from the Open Source Community”. In: Sihling, M. (ed.)

Proceedings of OOPSLA 2003: The 1st Workshop on Open Source in an Industrial

Context (OSIC’03). 26-30 October 2003, Anaheim, California, USA. pp. 21-30.

Technical University of Minchen, Minchen.

European Communities. (2003). "The IDA Open Source Migration Guidelines".

European Communities [Online], 26 September 2003.

http://www.netproject.com/docs/migoss/v1.0/index.html [Accessed 27 May 2004].

European Observatory for SMEs. (2000). Report of the European Observatory for

SMEs. EU: European Comission. (6th Report).

Fardal, H. & Tollefsen, H. (2004). Motivational Factors in Computer Training: The

Influence of Promoting Factors on Preventing Factors. MSc, University College of

Buskerud.

Ferris, P. (2003). “The age of corporate open source enlightment”. QUEUE,

July/August 2003, 34-44.

FFIEC. (2004). Risk Management of Free and Open Source Software. Arlington:

Federal Financial Institutions Examination Council.

Fitzgerald, B. & Kenny, T. (2003). Open Source Software can Improve the Health of

the Bank Balance - The Beaumont Hospital Experience [Online]. Rugby, UK:

105

Netproject. http://opensource.mit.edu/papers/fitzgeraldkenny.pdf [Accessed 12 June

2004].

Fonda, D. & Locke, L. (2003). “Title: It's The B Team's Time To Shine”. Time, 15

September 2003, 162 (11), 62-63.

Gabriel, R. P. & Goldman, R. (2002). "Open Source: Beyond the Fairytales".

Perspectives on Business Innovation Journal, 1(8).

Gacek, C., Lawrie, T. & Arief, B. (2004). “The Many Meanings of Open Source”.

IEEE Software, January/February 2004, 21(1), 34-40.

Gallivan, M. (2000). "Examining workgroup influence on technology usage: a

community of practice perspective". In: Prasad, J. & Nance, W. (eds) Proceedings of

the ACM SIGCPR 2000 conference on Computer personnel research. 6-8 April 2000,

Chicago, Illinois, USA. pp. 54-66. ACM Press, New York.

Gallivan, M. (2001). “Striking a balance between trust and control in a virtual

organization: a content analysis of open source software case studies”. Information

Systems Journal, 11(4), 277–304.

Gallivan, M. (2003). “Examining Gender Differences in IT Professionals Perceptions

of Job Stress in Response to Technological Change”. In: Proceedings of the 2003

SIGMIS conference on Computer personnel research. 10-12 April 2003,

Philadelphia, Pennsylvania, USA. pp. 10-23. ACM Press, New York.

Galup, S., Saunders, C. & Cerveny, R. (1998). “The use of Temporary Managers and

Employees in a Local Government Information Systems Services Department”. In:

Agarwal, R. (ed.) Proceedings of the 1998 ACM SIGCPR conference on Computer

personnel research. 26-28 March 1998, Boston, Massachusetts, USA. pp. 110-115.

ACM Press, New York.

106

Garcia, J. M. (2001). Innovating Without Money: Linux and the Open Source

Paradigm as an Alternative to Commercial Software Development. MSc, University

of Sussex.

Gartner. (2003). "Gartner Technology Demand Index Shows IT Spending Remained

Soft in October, Although Expected Increases Point to Optimism for 2004". Gartner

[Online], 2003 Press releases, 14 November 2003.

http://www4.gartner.com/5_about/press_releases/pr14nov2003a.jsp [Accessed 22

Sep 2004].

Gasser, L. & Scacchi, W. (2003). Continuous Design of Free/Open Source Software,

Preliminary Workshop Report and Research Agenda. Irvine: University of

California; Urbana-Champaign: University of Illinois. (Report for IIS-0350754).

Gefen, D. & Ridings, C. M. (2003). “IT Acceptance: Managing User – IT Group

Boundaries”. ACM SIGMIS Database, 34(3), 25-40.

Ghosh, R. A. (2003). "Open Source Software in Europe: Facts on the ground". In:

Proceedings of Annual AICA workshop. 27 June 2003, Milan, Italy.

Ghosh, R. A. (2004). "Open source: understanding developers". In: EuroIndia

conference. 25 March 2004, New Delhi, India.

Ghosh, R. A., Glott, R., Krieger, B. & Robles, G. (2002). Free/Libre and Open

Source Software: Survey and Study, Final Report. Maastrich: International Institute

of Infonomics. (D18:Final Report, parts 1-5).

GITOC. (2003). Using open source software in the South African Government. South

Africa: The Government Information Technology Officers' Council. (Version 3.3).

Graham, P. (2005). “Hiring is obsolete”. Paul Graham [Online], May 2005.

http://www.paulgraham.com/hiring.html [Accessed 27 August 2005].

107

Greenemeier, L. (2004). “In Search Of Open-Source Experts”. InformationWeek

[Online], 8 November 2004.

http://www.informationweek.com/story/showArticle.jhtml?articleID=52200175

[Accessed 21 August 2005].

Greenemeier, L. (2005). “Open-Source Development Effect”. InformationWeek

[Online], 25 July 2005.

http://www.informationweek.com/story/showArticle.jhtml?articleID=166401753

[Accessed 3 August 2005].

Griesser, J. (1993). “Motivation and Information System Professionals”. Journal of

Managerial Psychology, 8(3), 21-30.

Griggs, W. & Manring, S. L. (1986). “Inreasing the Effectiveness of Technical

Professionals”. Management Review, 75(5), 62-64.

Grosfeld, I. & Roland, G. (1996). “Defensive and Strategic Restructuring In Central

European Enterprises”. Journal of Transforming Economies and Societies, 1996, 3

(4), 1-44.

Grudin, J. (1994). “Groupware And Social Dynamics: Eight Challenges For

Developers”. Communications of the ACM, 37(1), 92-105.

Gustafson, P. & Koff, W. (2004). Open Source: Open for Business. Computer

Science Corporation: Leading Edge Forum.

Guzman, I. R., Stanton, J. M., Stam, K. R., Vijayasri, V., Yamodo, I., Zakaria, N. &

Caldera, C. (2004). “A Qualitative Study of the Occupational Subculture of

Information Systems Employees in Organizations”. In: Careers, culture, and ethics

in a networked environment. Proceedings of the 2004 SIGMIS conference on

Computer personnel research. 22-24 April 2004, Tucson, AZ, USA. pp. 74-80. ACM

Press, New York.

108

Hadfield, W. (2005). “IT staff at higher risk of burnout than other workers, says

survey”. Computer Weekly, No. 00104787, 9 August 2005.

Han, S. (2003). Individual Adoption of Information Systems in Organisations: A

Literature Review of the Intention-based Theories. University of Turku; Abo

Akademi University; Turku School of Economics: Turku Centre for Computer

Science. (TUCS Technical Report No. 539).

Hars, A. & Ou, S. (2002). "Working for Free? Motivations for Participating in Open-

Source Projects". International Journal of Electronic Commerce, 6(3), 25-39.

Hawkins, R. E. (2004). “The Economics Of Open Source Software For a

Competitive Firm”. NETNOMICS: Economic Research and Electronic Networking,

6(2).

Hawryszkiewycz, I. T. (1999). “Knowledge Sharing Through Workspace Networks”.

In: Prasad, J. (ed.) Proceedings on the SIGCPR 1999 conference on Computer

personnel research. 8-10 April 1999, New Orleans, USA. pp. 79-85. ACM Press,

New York.

Heimer, R. L. (2003). “The Common Optimization Interface for Operations

Research: Promoting open-source software in the operations research community”.

IBM Journal for R&D, January 2003, 47(1), 57-65.

Hirschheim, R. & Klein, H. K. (1989). “Four Paradigms of Information Systems

Development”. Communications of the ACM, 32(10), 1199-1216.

Hirschheim, R. & Lacity, M. (2000). “The myths and realities of information

technology insourcing”. Communications of the ACM, 43(2), 99-107.

Hodgson, L. & Aiken, P. (1998). “Organizational Change Enabled By The Mandated

Implementation Of New Information Systems Technology: A Modified Technology

Acceptance Model”. In: Agarwal, R. (ed.) Proceedings of the 1998 ACM SIGCPR

109

conference on Computer personnel research. 26-28 March 1998, Boston,

Massachusetts, USA. pp. 205-213. ACM Press, New York.

Holck, J., Larsen, M. H. & Pedersen, M. K. (2004). "Identifying Business Barriers

and Enablers for the Adoption of Open Source Software". In: 13th International

Conference on Information Systems Development (ISD 2004). 9-11 September 2004,

Vilnius, Lithuania. Information Systems Scientific Laboratory, Vilnius Gediminas

Technical University, Vilnius.

Holtgrewe, U. & Werle, R. (2001). “De-Commodifying Software? Open Source

Software Between Business Strategy and Social Movement”. Science Studies, 2001,

14(2), 43–65.

Igbaria, M. & Guimaraes, T. (1992). “Antecedents and Consequences of Job

Satisfaction Among Information Center Employees”. In: Lederer, A. L. (ed.)

Proceedings of the 1992 ACM SIGCPR conference on Computer personnel research.

05-07 April 1992, Cincinnati, Ohio, USA. pp. 352-369. ACM Press, New York.

International Institute of Infonomics. (2002). Free/Libre and Open Source Software:

Survey and Study. The Netherlands: University of Maastricht; Berlin, Germany:

Berlecon Research. (Final Report, Deliverable D18).

Ivancevich, J. M., Napier, H. A. & Wetherbe, J. C. (1983). “Stress in the Information

Systems Professional”. In: Awad, E. M. (ed.) The Proceedings of the Twentieth

Annual Computer Personnel on Research Conference. 17-18 November 1983,

Charlottesville, Virginia, USA. pp. 57-68. ACM Press, New York.

Janz, B. D. (1998). “The Best and Worst of Teams: Self-Directed Work Teams as an

Information Systems Development Workforce Strategy”. In: Agarwal, R. (ed.)

Proceedings of the 1998 ACM SIGCPR conference on Computer personnel research.

26-28 March 1998, Boston, Massachusetts, USA. pp. 59-67. ACM Press, New York.

110

Jarvinen, P. (1983). “A Role Of A User In The Development And Maintenance Of

An Information System: Empirical And Theoretical Findings”. ACM SIGCPR

Computer Personnel, 9(2), 3-10.

Johnson, J. P. (2002). “Open Source Software: Private Provision of a Public Good”.

Journal of Economics & Management Strategy, 11(4), 637–662.

Joshi, K., Bostrom, R. P. & Perkins, W. C. (1986). “Some New Factors Influencing

User Information Satisfaction: Implications For Systems Professionals”. In: Licker,

P. S. (ed.) Proceedings of the Twenty-Second annual Computer Personnel Research

Conference. 16-17 October 1986, Calgary, Canada. pp. 27-42. ACM Press, New

York.

KBSt. (2003). Migration Guide: A guide to migrating the basic software components

on server and workstation computers. Federal Ministry of the Interior of Germany:

KBSt Publication Series. (Volume 57).

Kenwood, C. A. (2001). A Business Case Study of Open Source Software. Bedford:

The MITRE Corporation. (Report MP 01B0000048).

Kleintop, W., Blau, G. & Curall, S. C. (1994). "Practice makes use: Using

information technologies before implementation and the effect on acceptance by end

users". In: Proceedings of the 1994 computer personnel research conference on

Reinventing IS. 24-26 March 1994, Alexandria, Virginia, USA. pp. 120-132. ACM

Press, New York.

Klenke, K. & Kievit, K. (1992). “Predictors of Leadership Style, Organizational

Commitment and Turnover of Information Systems Professionals”. In: Lederer, A. L.

(ed.) Proceedings of the 1992 ACM SIGCPR conference on Computer personnel

research. 05-07 April 1992, Cincinnati, Ohio, USA. pp. 171-183. ACM Press, New

York.

111

Klenke, K. (1996). “IS Leadership In Virtual Organizations”. Proceedings of the

1996 ACM SIGCPR/SIGMIS conference on Computer personnel research. 11-13

April 1996, Denver, Colorado, USA. pp. 22-35. ACM Press, New York.

Kling, R. (2000). “Learning About Information Technologies and Social Change:

The Contribution of Social Informatics”. Information Society, 16(3), 217-232.

Koch, C. (2003). “Open Source - Your Open Source Plan”. CIO Magazine [Online],

15 March 2003. http://www.cio.com/archive/031503/opensource.html [Accessed 25

July 2005].

Kogut, B. & Metiu, A. (2001). “Open Source Software Development and Distributed

Innovation”. Oxford Review of Economic Policy, 17(2), 248-264.

Koop, R. & Grant, R. (1993). “Information Systems And Power: Structural ‘Versus

Personal Views”. In: Tanniru, M. R. (ed.) Proceedings of the 1993 conference on

Computer personnel research. 01-03 April 1993, St Louis, Missouri, USA. pp. 265-

272. ACM Press, New York.

Kryt, J. (1980). “Knowledgeable And Friendly End Users”. In: Awad, E. M. (ed.)

Proceedings of the 17th Annual Computer Personnel Research Conference. 26-27

June 1980, Miami, Florida, USA. pp. 9-20. ACM Press, New York.

Kuchinskas, S. (2005). "Study: Cost Not Only Open Source Driver". InternetNews

[Online], 14 July 2005.

http://www.internetnews.com/ent-news/article.php/3520066 [Accessed 20 July

2005].

Kwan, S. K. & West, J. (2004). “Heterogeneity of IT Importance: Implications for

Enterprise IT Portfolio Management”. The CRITO Hour. 27 January 2004, Irvine,

CA, USA. University of California, Irvine.

112

Lagace, M. (2003). “Are You Supporting Your B Players?”. Harward Business

School [Online], 30 June 2003. http://hbswk.hbs.edu/item/3558.html [Accessed 20

August 2005].

Lakhani, K. R. & Wolf, R. G. (2003). Why Hackers Do What They Do:

Understanding Motivation Effort in Free/Open Source Software Projects. MIT Sloan

School of Management. (Working Paper 4425-03).

Lambe, P. (2002). “Ignorance Management: The Lessons of Small Enterprises for

Knowledge Management”. Journal of Information and Knowledge Management,

April 2002, 1(1).

Larsen, M. H., Holck, J. & Pedersen, M. K. (2000). “The Challenges of Open Source

Software in IT Adoption: Enterprise Architecture versus Total Cost of Ownership”.

In: Proceedings of the 27th conference on Information Systems Research in

Scandinavia (IRIS'27). 14-17 August 2000, Falkenberg, Sweden.

Lassen, K. B. & Sorensen, J. (2001). “Open Source in a Major Swiss Bank”.

UPGRADE: The European Journal for the Informatics Professional, 2(6), 33-35.

Lawrie, T. & Gacek, C. (2002). “Issues of Dependability in Open Source Software

Development”. Software Engineering Notes, 27(3), 34-37.

Lawton, G. (2002). “Open Source Security: Opportunity or Oxymoron?”. Industry

Trends, March 2002, 18-21.

Lederer, A. L. & Mendelow, A. L. (1989). “Information Systems Planning:

Incentives For Effective Action”. ACM SIGMIS Database, 20(3), 13-20.

Lee, J & Chan, T. (2004). “Organizational Structure of “User Collaboration

Community”: Insights from the Case of an Open Source Software Project”. In:

Feller, J., Fitzgerald, B., Hissam, S. & Lakhani, K. (eds) Collaboration, Conflict and

Control. Proceedings of the 4th Workshop on Open Source Software Engineering. 21

May 2004, Edinburgh, Scotland, UK. pp. 58-64. ACM Press, New York.

113

Leonard, A. (2002). “Enabling End Users To Be More Efficient During Systems

Development”. In: Proceedings of the 2002 annual research conference of the South

African institute of computer scientists and information technologists on Enablement

through technology. 16-18 September 2002, Port Elizabeth, South Africa. pp. 156-

162. South African Institute for Computer Scientists and Information Technologists,

Republic of South Africa.

Lerner, J. & Tirole, J. (2002). "Some Simple Economics of Open Source". Journal of

Industrial Economics, 50(2), 197-234.

Lessig, L. (2001). “The Architecture of Innovation”. In: Conference on the Public

Domain. 9-11 November 2001, Durham, NC, USA. pp. 177-191. Duke University

School of Law, Durham.

Lin, Y. (2004). Hacking Practices and Software Development: A Social Worlds

Analysis of ICT Innovation and the Role of Free/Libre Open Source Software

[Online]. Cambridge, MA, USA: MIT Sloan.

http://opensource.mit.edu/papers/lin2.pdf [Accessed 12 May 2005].

Lin, Y. (2005). Inclusion, diversity and gender equality: Gender Dimensions of the

Free/Libre Open Source Software Development [Online]. Cambridge, MA, USA:

MIT Sloan. http://opensource.mit.edu/papers/lin3_gender.pdf [Accessed 14

November 2005].

Lussier, S. (2004). "New Tricks: How Open Source Changed the Way My Team

Works". IEEE Software, 21(1), 68-72.

Luthiger, B. (2005). “Fun and Software Development”. In: Scotto, M. & Succi, G.

(eds) Proceedings of the 1st International Conference on Open Source Systems. 11-

15 July 2005, Genova, Italy. pp. 273-278. University of Genova, Genova.

Lyman, J. (2005). “Getting in touch with the feminine side of open source”.

SearchWebService [Online], 18 July 2005.

114

http://www.newsforge.com/articles/05/08/08/1449259.shtml [Accessed 17 August

2005].

Lynn, D. (1991). The application of case study evaluations. Practical Assessment,

Research & Evaluation [Online], 2(9). http://PAREonline.net/getvn.asp?v=2&n=9

[Accessed 01 May 2006].

Lyytinen, K. (1987). “Different Perspectives on Information Systems: Problems and

Solutions”. ACM Computing Surveys (CSUR), 19(1), 5-46.

Maass, W. (2004). “Inside an Open Source Software Community: Empirical

Analysis on Individual and Group Level”. In: Feller, J., Fitzgerald, B., Hissam, S. &

Lakhani, K. (eds) Collaboration, Conflict and Control. Proceedings of the 4th

Workshop on Open Source Software Engineering. 21 May 2004, Edinburgh,

Scotland, UK. pp. 65-70. ACM Press, New York.

Maggioni, M. A. (2002). Open Source Software Communities and Industrial

Districts: a Useful Comparison? [Online]. Cambridge, MA, USA: MIT Sloan.

http://opensource.mit.edu/papers/maggioni.pdf [Accessed 9 January 2005].

Mahapatra, R. K. & Lai, V. S. (1999). “Evaluation Of Intranet-Based End-User

Training”. In: Proceeding of the 20th international conference on Information

Systems. 12-15 December 1999, Charlotte, North Carolina, USA. pp. 524-527.

Association for Information Systems, Atlanta.

Mahring, M. (1998). “Managerial Control Over It Projects: Control, Forms,

Commitment, And Dominant Coalitions”. In: Proceedings of the international

conference on Information systems. 13-16 December 1998, Helsinki, Finland. pp.

336-340. Association for Information Systems, Atlanta.

Maqua, L., Kunzler, U. & McAvinue, B. (2004). "EUROSTAT And Open Source".

In: EUROSTAT: Joint ECE/Eurostat/OECD Meeting on the Management of

Statistical Information Systems (MSIS). 17-19 May 2004, Geneva, Switzerland.

115

Markus, L., Manville, B. & Agres, C. E. (2000). "What makes a virtual organization

work?". Sloan Management Review, Fall 2000, 13-26.

Markus, M. L. (1983). “Power, Politics, and MIS Implementation”. Communications

of the ACM, 26(6), 430-444.

McDaniel, D. (2000). Linux in the Enterprise. Computer Science Corporation:

Leading Edge Forum.

McHugh, A. & Donnelly, M. (2004). Open Source Software and Standards.

DigiCULT. (Report Ref. IST-2001-34898).

Meehan, M. (2005). “Price not as important in open source adoption”. TechTarget

[Online], 18 July 2005.

http://searchwebservices.techtarget.com/originalContent/0,289142,sid26_gci1108414

,00.html [Accessed 27 July 2005].

Melian, C., Ammirati, C. B. & Garg, P. Sevon, G. (2002). Building networks of

software communities in a large corporation. Stockholm University School of

Business; HP Labs; Stockholm School of Economics. (Tech Report No. HPL-2002-

12 ).

Michlmayr, M., Hunt, F. & Probert, D. (2005). “Quality Practices and Problems in

Free Software Projects”. In: Scotto, M. & Succi, G. (eds) Proceedings of the 1st

International Conference on Open Source Systems. 11-15 July 2005, Genova, Italy.

pp. 24-28. University of Genova, Genova.

Miller, R. (2001). “Secretaries use Linux, taxpayers save millions”. NewsForge

[Online], 13 August 2005.

http://www.newsforge.com/article.pl?sid=01/08/10/1441239 [Accessed 17 August

2005].

Miller, R. (2004). "Where should you use open source? A scenario-based analysis".

NewsForge [Online], 25 August 2004.

116

http://business.newsforge.com/print.pl?sid=04/05/21/1636232 [Accessed 25 August

2004].

Morgan, J. (2002). Open Source Software and Software Patents [Online].

Cambridge, MA, USA: MIT Sloan. http://opensource.mit.edu/papers/morgan.pdf

[Accessed 19 August 2004].

Nagler, M. (2005). “Open Source Software Migration of the Employers’ Liability

Insurance Association”. European Communities [Online], 28 June 2005.

http://europa.eu.int/idabc/en/document/4396/470 [Accessed 20 August 2005].

Nandhakumar, J. & Jones, M. (1997). “Designing In The Dark: The Changing User-

Developer Relationship In Information Systems Development”. In: Proceedings of

the 18th international conference on Information systems. 14-17 December 1997,

Atlanta, Georgia, USA. pp. 75-88. Association for Information Systems, Atlanta.

Neilson, G. L., Pasternack, B. A. & Van Nuys, K. E. (2005). “The Passive-

Aggressive Organization”. Harvard Business Review, 84(2), 159-160.

Nelson, A. C. & Joshi, K. (1993). “Effectively Utilizing Systems Developers On

Projects”. In: Tanniru, M. R. (ed.) Proceedings of the 1993 conference on Computer

personnel research. 01-03 April 1993, St Louis, Missouri, USA. pp. 188-196. ACM

Press, New York.

Nelson, R. R., Whitener, E. M. & Philcox, H. H. (1995). “The Assessment of End-

User Training Needs”. Communications of the ACM, 38(7), 27-39.

Niederman, F. & Webster, J. (1998). “Trends in End-User Training: A Research

Agenda”. In: Agarwal, R. (ed.) Proceedings of the 1998 ACM SIGCPR conference

on Computer personnel research. 26-28 March 1998, Boston, Massachusetts, USA.

pp. 224-232. ACM Press, New York.

Nielsen, J., Mack, R. L., Bergendorff, K. H. & Grischkowsky, N. L. (1986).

“Integrated Software Usage in the Professional Work Environment: Evidence from

117

Questionnaires and Interviews”. In: Mantei, M. & Orbeton, P. (eds) Proceedings of

the SIGCHI conference on Human factors in computing systems. 13-17 April 1986,

Boston, Massachusetts, USA. pp. 162-167. ACM Press, New York.

Nikula, U. & Jantunen, S. (2005). “Quantifying the Interest in Open Source Systems:

Case South-East Finland”. In: Scotto, M. & Succi, G. (eds) Proceedings of the First

International Conference on Open Source Systems. 11-15 July 2005, Genova, Italy.

pp. 192-195. University of Genova, Genova.

Nilsen, H. & Sein, M. (2004). "What is really important in supporting end-users?".

In: Tanniru, M. & Weisband, S. P. (eds) Culture, and Ethics in a Networked

Environment. Proceedings on the SIGCPR 2004 conference on Computer personnel

research. 22-24 April 2004, Tucson, Arizona, USA. pp. 48-54. ACM Press, New

York.

Noronha, F. (2005). ”The social implications of free software”. Free Software

Magazine [Online], 6 September 2005.

http://www.freesoftwaremagazine.com/free_issues/issue_04/social_implications/

[Accessed 25 July 2005].

Northwest Educational Technology Consortium. (2005). “Users & migration”. Open

Options [Online]. Northwest Regional Educational Laboratory, Open Options.

http://www.netc.org/openoptions/pros_cons/users.html [Accessed 7 August 2005].

Novell. (2004). Linux Migration Solutions. Provo, Utah, USA: Novell, Inc.

Novell. (2005). “Resource Management: OS Migration”. Novell, Inc. [Online].

http://www.novell.com/identity/resourcemanagement/os_migration/ [Accessed 7

August 2005].

Nuvolari, A. (2003). Open Source Software Development: Some Historical

Perspectives [Online]. Cambridge, MA, USA: MIT Sloan.

http://opensource.mit.edu/papers/nuvolari.pdf [Accessed 19 August 2004].

118

O’Reilly, T. (1999). “Lessons From Open Source Software Development”.

Communications of the ACM, 42(4), 33-37.

Olfman, L. & Shayo, C. (1997). “The Role Of Training In Preparing End Users To

Learn New But Similar Software Packages”. In: Niederman, F. (ed.) Proceedings of

the 1997 ACM SIGCPR conference on Computer personnel research. 03-05 April

1997, San Francisco, California, USA. pp. 210-223. ACM Press, New York.

Olfman, L., Sein, M. & Bostrom, R. P. (1986). “Training For End-User Computing:

Are Basic Abilities Enough For Learning?”. In: Licker, P. S. (ed.) Proceedings of the

Twenty-Second annual Computer Personnel Research Conference. 16-17 October

1986, Calgary, Canada. pp. 1-11. ACM Press, New York.

Osterlie, T. (2004). “In the network: Distributed control in Gentoo Linux”. In: Feller,

J., Fitzgerald, B., Hissam, S. & Lakhani, K. (eds) Collaboration, Conflict and

Control. Proceedings of the 4th Workshop on Open Source Software Engineering. 21

May 2004, Edinburgh, Scotland, UK. pp. 76-81. ACM Press, New York.

Parker, K. A. (2000). Open Sources And The Open Society An Essay In Politics And

Technology [Online]. Michigan, USA: Grand Valley State University.

http://agora.phi.gvsu.edu/kap/OSFS/OSFSEssay.pdf [Accessed 9 January 2005].

Parry, K. (2000). “Does Leadership Help The Bottom Line?”. New Zealand

Management, April 2000, 47(3), 38-41.

Patricia, M. (2005). “Getting territorial over technology”. Computing Canada, 31(2),

30.

Paul, R. (2005). “Review: Linux Client Migration Cookbook”. NewsForge [Online],

1 February 2005.

http://business.newsforge.com/article.pl?sid=05/01/24/209254&tid=37 [Accessed 7

August 2005].

119

Payne, C. (2002). “On the security of open source software”. Info Systems Journal,

12, 61–78.

Peeling, N. & Satchell, J. (2001). Analysis of the Impact of Open Source Software.

UK: QinetiQ Ltd. (Report No. CR010223).

Perens, B. (1999). "The Open Source Definition". In: DiBona, C., Ockman, S. &

Stone, M. (eds), "Open Sources: Voices from the Open Source Revolution".

Sebastopol, CA, USA: O'Reilly & Associates.

Pliskin, N. & Shova, P. (1987). “End-User Prototyping: Sophisticated Users

Supporting System Development”. ACM SIGMIS Database, 18(4), 7-17.

President’s Information Technology Advisory Committee. (2000). Panel on Open

Source Software for High End Computing. Arlington: National Coordination Office

for Computing, Information, and Communications. (Report to the President).

QNB Intelligence. (2004). Corporate Linux Adoption in Europe. Berkshire: QNB

Intelligence Ltd, Quocirca. (Special Report).

Ramakrishna, H.V. & Lin, X. (1999). “Perception of the Role of Information

Technology Function in Organizations: Toward the Development of a Measure”.

ACM SIGCPR Computer Personnel, October 1999, 20(4), 39-54.

Rayksund, C. & Heltn, M. M. (1988). “Critical Success Factors In Information

Center Strategy”. In: Proceedings of the ACM SIGCPR conference on Management

of information systems personnel. 07-08 April 1988, College park, Maryland, USA.

pp. 33-41. ACM Press, New York.

Relevantive. (2003). Linux Usability Study Report. Berlin: Relevantive AG. (Linux

Usability Report).

Robert, G. & Schutz, F. (2001). “Should Business Adopt Free Software?”. The

European Online Magazine for the IT Professional, December 2001, 2(6), 12-19.

120

Robinson, S. (2005). “Beware the zealot in an information technology management

role”. builder.com [Online], 13 May 2005. http://builder.com.com/5102-6401-

5706440.html [Accessed 4 August 2005].

Robinson, S. (2005). “Embrace the software zealot in a development role”.

TechRepublic [Online], 28 June 2005. CNET Networks, TechRepublic. [Accessed 8

July 2005].

Rossi, B., Scotto, M., Sillitti, A. & Succi, G. (2005). “Criteria for the non invasive

transition to OpenOffice”. In: Scotto, M. & Succi, G. (eds) Proceedings of the 1st

International Conference on Open Source Systems. 11-15 July 2005, Genova, Italy.

pp. 250-253. University of Genova, Genova.

Rossi, M. A. (2004). Decoding the “Free/Open Source (F/OSS) Software Puzzle: a

survey of theoretical and empirical contributions. University of Siena: Department of

Economics. (University of Siena Report No. 424).

Routio, P. (2006). Models in the Research Process [Online]. Helsinki: University of

Art and Design. http://www2.uiah.fi/projects/metodi/177.htm [Accessed 22 June

2006]

Rues, R. (2003). “Corporate Open Source Information Leakage”. Infocon Magazine,

Issue One, October 2003, 1-11.

Sarin, S. & McDermott, C. (2003). “The Effect of Team Leader Characteristics on

Learning, Knowledge Application, and Performance of Cross-Functional New

Product Development Teams”. Decision Sciences, 34(4), 707-739.

Sarma, A. & Hoek, A. V. (2004). “A Conflict Detected Earlier is a Conflict Resolved

Easier”. In: Feller, J., Fitzgerald, B., Hissam, S. & Lakhani, K. (eds) Collaboration,

Conflict and Control. Proceedings of the 4th Workshop on Open Source Software

Engineering. 21 May 2004, Edinburgh, Scotland, UK. pp. 82-86. ACM Press, New

York.

121

Sauer, C., Southon, G. & Dampney, C. N. G. (1997). “Fit, Failure, And The House

Of Horrors: Toward A Configurational Theory of IS Project Failure”. In:

Proceedings of the 18th international conference on Information systems. 14-17

December 1997, Atlanta, Georgia, USA. pp. 349-366. Association for Information

Systems, Atlanta.

Saunders, M., Lewis, P. & Thornhill, A. (2000). Research Methods for Business

Students. Second edition. Harlow, England: Pearson Education.

Sawyer, S., Eschenfelder K.R., Diekema A. & McClure C. R. (1998). “IT Skills in

the Context of BigCo.”. In: Agarwal, R. (ed.) Proceedings of the 1998 ACM SIGCPR

conference on Computer personnel research. 26-28 March 1998, Boston,

Massachusetts, USA. pp. 9-18. ACM Press, New York.

Scacchi, W. (2002). Open EC/B: A Case Study in Electronic Commerce and Open

Source Software Development. University of California: Institute for Software

Research. (Technical Report).

Scacchi, W. (2002). Understanding the Social, Technological, and Public Policy

Implications of Open Source Software Development. [Online]. Irvine, CA, USA:

University of California, Institute for Software Research.

http://www.ics.uci.edu/~wscacchi/Papers/New/OSS-Policy.pdf [Accessed 14 January

2004].

Scacchi, W. (2003). Socio-Technical Design[Online]. Irvine, CA, USA: UC Irvine,

Donald Bren School of Information and Computer Sciences.

http://www.ics.uci.edu/~wscacchi/Papers/SE-Encyc/Socio-Technical-Design.pdf

[Accessed 14 October 2003].

Scacchi, W. (2004). “Socio-Technical Interaction Networks in Free/Open Source

Software Development Processes”. In: Acuna, S. T. & Juristo, N. (eds): Peopleware

and the Software Process. October 2004. World Scientific Press.

122

Scacchi, W. (2004). Understanding Open Source Software Evolution [Online].

Irvine, CA, USA: University of California, Institute for Software Research.

http://www.ics.uci.edu/~wscacchi/Papers/New/Understanding-OSS-Evolution.pdf

[Accessed 6 September 2004].

Schadler, T. (2004). “Commentary: An open-source plan”. CNET News [Online], 14

April 2004. Special to CNET News By Forrester Research.

http://news.com.com/2030-7344-5191481.html [Accessed 25 July 2005].

Scherler, T. (2004). Open Source Software (OSS) within organization: Critical

factors for consulting [Online]. Cambridge, MA, USA: MIT Sloan.

http://opensource.mit.edu/papers/scherler1.pdf [Accessed 19 August 2004].

Schmitz. P. E. (2001). Study into the use of Open Source Software in the Public

Sector. Brussels: European Commission and DG Enterprise. (Part 1).

Schweik, C. M. & Semenov, A. (2002). “The Institutional Design of Open Source

Programming: Implications for Addressing Complex Public Policy and Management

Problems”. FirstMonday [Online], 8(1).

http://www.firstmonday.dk/issues/issue8_1/schweik/index.html [Accessed 7 August

2005].

Scott, J. E. (1995). “The Measurement of Information Systems Effectiveness:

Evaluating a Measuring Instrument”. ACM SIGMIS Database, 26(1), 43-61.

Seidel, W. & Niedermeier, C. (2003). “Open Source Software: Leveraging Software

Quality in the Industrial Context”. In: OSSIE '03 - 1st Workshop on Open Source

Software in an Industrial Environment. 25. September 2003, Erfurt, Germany.

Seifert, T. & Wieland, T. (2003). “Prerequisites For Enterprises To Get Involved In

Open Source Software Development”. In: OSSIE '03 - 1st Workshop on Open Source

Software in an Industrial Environment. 25. September 2003, Erfurt, Germany.

123

Shaikh, M. & Cornford, T. (2004). “Version Control Tools: A Collaborative Vehicle

for Learning in F/OS”. In: Feller, J., Fitzgerald, B., Hissam, S. & Lakhani, K. (eds)

Collaboration, Conflict and Control. Proceedings of the 4th Workshop on Open

Source Software Engineering. 21 May 2004, Edinburgh, Scotland, UK. pp. 87-91.

ACM Press, New York.

Sharma, S., Sugumaran, V. & Rajagopalan, B. (2002). "A framework for creating

hybrid-open source software communities". Information Systems Journal, 12, 7-25.

Shaw, N. C., DeLone, W. H. & Niederman, F. (2002). “Sources of Dissatisfaction in

End-User Support: An Empirical Study Nancy C. Shaw”. ACM SIGMIS Database,

33(2), 41-56.

Shayo, C. & Olfman, L. (1993). “Is The Effectiveness Of Formal End-User Software

Training A Mirage?”. In: Tanniru, M. R. (ed.) Proceedings of the 1993 conference

on Computer personnel research. 01-03 April 1993, St Louis, Missouri, USA. pp.

88-99. ACM Press, New York.

Shayo, C. & Olfman, L. (1994). “A Three Dimensional View And Research Agenda

For The Study Of Transfer Of Skills Galned From Formal End-User Software

Training”. In: Proceedings of the 1994 computer personnel research conference on

Reinventing IS. 24-26 March 1994, Alexandria, Virginia, USA. pp. 133-141. ACM

Press, New York.

Sherer, S. A. & Alter, S. (2004). “Information System Risks and Risk Factors: Are

they Mostly About Information Systems?”. Communications of the Association for

Information Systems,14, 29-64.

Sirkin, H. L., Keenan, P. & Jackson, A. (2005). “The Hard Side Of Change

Management”. Hardward Business Review, October 2005, 109-118.

Sliwa, C. (2005). “Financial Services Companies Lead the Charge to Linux”.

LinuxInsider [Online], 28 July 2005.

124

http://www.linuxinsider.com/story/NWsbJcz9gPnXzt/Financial-Services-

Companies-Lead-the-Charge-to-Linux.xhtml [Accessed 10 August 2005].

Sockel, H., Chen, K. & Kroeger, J. W. (2001). “Technological Adoption to Combat

Burnout”. In: Colton, D. (ed.) Technology in the 21st century: Where Innovation and

Information Converge. The Proceedings of ISECON 2001. 1-4 November 2001,

Sincinnati, USA.

SoftwareMag. (2005). “Application Development Product Coverage: IT

Development Survey Shows 74% Use Open Source”. SoftwareMag [Online], 6 April

2005. http://www.softwaremag.com/L.cfm?Doc=newsletter/2005-04-06/vasoftware

[Accessed 19 August 2005].

Spanos, Y., Prastacos, G. & Papadakis, V. (2001). “Greek Firms and EMU:

Contrasting SMEs and Large-Sized Enterprises”. European Management Journal,

19(6), pp. 638–648.

Spiller, D. & Wichmann, T. (2002). Free/Libre and Open Source Software: Survey

and Study. Berlin: Berlecon Research. (Final Report, parts 1-3).

Stafford, J. (2005). “LinuxWorld preview: The open source waiting game -- pros &

cons”. TechTarget [Online], 2 August 2005.

http://searchenterpriselinux.techtarget.com/originalContent/0,289142,sid39_gci1112

002,00.html [Accessed 10 August 2005].

Stalder, F. & Hirsh, J. (2002). “Open Source Intelligence”. FirstMonday [Online],

7(6). http://www.firstmonday.org/issues/issue7_6/stalder/ [Accessed 7 August 2005].

Stallman, R. (1999). "The GNU Operating System and the Free Software

Movement". In: DiBona, C., Ockman, S. & Stone, M. (eds), "Open Sources: Voices

from the Open Source Revolution". Sebastopol, CA, USA: O'Reilly & Associates.

Steward, J. E., Mangalam, H. & Zhou, J. (2001). “Open Source Software meets gene

expression”. Briefings in Bioinformatics, 2(4), 319-328.

125

Stewart, K. & Gosain, S. (2001). “An Exploratory Study Of Ideology And Trust In

Open Source Development Groups”. In: Proceedings of The 22nd Annual

International Conference on Information Systems. 16-19 December 2001, New

Orleans, Louisiana, USA. pp. 507-512.

Stewart, K. J. & Gosain, S. (2003). Impacts Of Ideology, Trust, And Communication

On Effectiveness In Open Source Software Development Teams [Online]. Cambridge,

MA, USA: MIT Sloan. http://opensource.mit.edu/papers/stewartgosain.pdf

[Accessed 19 August 2004].

Stewart, K. J. (2004). “OSS Project Success: From Internal Dynamics to External

Impact”. In: Feller, J., Fitzgerald, B., Hissam, S. & Lakhani, K. (eds) Collaboration,

Conflict and Control. Proceedings of the 4th Workshop on Open Source Software

Engineering. 21 May 2004, Edinburgh, Scotland, UK. pp. 92-96. ACM Press, New

York.

Strand, E. M. (1980). “Documentation For In-House Sharing Of Software”. In:

Proceedings of the ACM 1980 annual conference. January 1980, USA. pp. 150-155.

ACM Press, New York.

Sumner, M. & Klepper, R. (1987). “Information Systems Strategy End-User And

Application Development”. Database, 18(4), 19-30.

Takahashi, M., Fujimoto, M. & Yamasaki, N. (2003). “The Active Lurker: Influence

of an In-house Online Community on its Outside Environment”. ACM SIGGROUP

Bulletin, 24(1), 4-4.

Tellis, W. (1997). “Introduction to Case Study” [Online]. The Qualitative Report,

3(2). http://www.nova.edu/ssss/QR/QR3-2/tellis1.html [Accessed 01 May 2006].

The Dravis Group. (2003). Open source software: Case studies examining its use.

San Francisco: The Dravis Group.

126

The MITRE Corporation. (2003). Use of Free and Open-Source Software (FOSS) in

the U.S. Department of Defense. Bedford: The MITRE Corporation. (Report MP 02

W000101).

The Swedish Agency for Public Management Publication Service. (2003) Free and

Open Source Software - a feasibility study. Stockholm, Sweden: The Swedish

Agency for Public Management Publication Service.

Thronicke, W. & Berger, F. (2004). “Interacting with the Open Source Community –

Effective models for a Traditional IT-business”. In: OSSIE '03 - 1st Workshop on

Open Source Software in an Industrial Environment. 25. September 2003, Erfurt,

Germany.

Treasury Board of Canada Secretariat. (2005). “Free and Open Source Software

Overview and Preliminary Guidelines for the Government of Canada”. Treasury

Board of Canada Secretariat [Online], 26 April 2005. http://www.tbs-sct.gc.ca/fap-

paf/oss-ll/foss-llo/foss-llo19_e.asp#x1-2500020 [Accessed 7 August 2005].

Trimmer, K. J., Collins, R. W., Will, R. P. & Blanton, J. E. (2000). “Information

Systems Development: Can There be 'Good' Conflict?”. In: Prasad, J. & Nance, W.

(eds) Proceedings of the ACM SIGCPR 2000 conference on Computer personnel

research. 6-8 April 2000, Chicago, Illinois, USA. pp. 174-179. ACM Press, New

York.

UNDP. (2003). International Open Source Network. UNDP: International Open

Source Network. (Asia-Pacific Development Information Programme Report 2003).

Vaden, T. (2003). Intellectual Property, Open Source and Free Software [Online].

Cambridge, MA, USA: MIT Sloan. http://opensource.mit.edu/papers/vaden.pdf

[Accessed 19 August 2004].

Van de Loo, E. (2000). “The clinical paradigm: Manfred Kets de Vries's reflections

on organizational therapy”. Academy of Management Executive, 14(1), 49-51.

127

VaSoftware. (2004). “Executives Should Open Minds to Open Source Development

Techniques”. VaSoftware [Online], 15 November 2004.

http://www.vasoftware.com/news/press.php/2004/1469.html [Accessed 19 August

2005].

Ven, K. & Verelst, J. (2004). “Control Objectives in Open Source Projects”. In:

Feller, J., Fitzgerald, B., Hissam, S. & Lakhani, K. (eds) Collaboration, Conflict and

Control. Proceedings of the 4th Workshop on Open Source Software Engineering. 21

May 2004, Edinburgh, Scotland, UK. pp. 100-104. ACM Press, New York.

Vile, D. (2005). Migrating to Linux on the Desktop: A Practical IT Management

View. Berkshire: Quocirca Ltd. (Business Report).

von Krogh, G. (2003). “Open-Source Software Development”. MIT Sloan

Management Review, spring 2003, 14-18.

Wagner, M. (2005). “Pitfalls In An Open-Source World”. InformationWeek [Online],

25 July 2005.

http://www.informationweek.com/story/showArticle.jhtml?articleID=166401756

[Accessed 28 August 2005].

Wang, Y. & Whitehead, E. J. (2001). International Accessibility of Open Source

Software [Online]. Santa Cruz, CA, USA: University of California, Department of

Computer Engineering. http://www.soe.ucsc.edu/~ywang/research/papers/oss01.pdf

[Accessed 12 May 2005].

Warne, L. (2003). “Organizational Politics and Information Systems Development -

A Model of Conflict”. HICSS Archive, 6(6), 482.

WatchIT. (2004). Linux at Work: Case Studies and Perspectives. New York, USA:

WatchIT. (Program 200407002000).

Waterman, A. (2003). Motivations to Develop Software as Open Source. Stanford

University: Stanford Institute for Economic Policy Research.

128

Weerawarana, S. & Weeratunge, J. (2003). Open Source in Developing Countries.

Sweden: Sida. (Art.no.: SIDA3460en).

Weiss, A. (2001). “The Politics of Free (Software)”. netWorker, 5(3), 26-31.

Wheatley, M. (2004). “The Myths of Open Source”. CIO Magazine [Online], 1

March 2004. http://www.cio.com/archive/030104/open.html [Accessed 20 August

2005].

Wheeler, D. A. (2002). Why Open Source Software / Free Software (OSS/FS)? Look

at the Numbers! [Online]. http://www.dwheeler.com/oss_fs_why.pdf [Accessed 19

July 2004].

Wiederhold, G. (2003). The Product Flow Model [Online]. Stanford: Stanford

University, Computer Science Department.

http://www-db.stanford.edu/LIC/ProductFlowModel.pdf [Accessed 12 May 2005].

Williams, J., Clegg, P. & Dulaney, E. (2005). “The Advantages of Adopting Open

Source Software”. Que [Online], 6 May 2005.

http://www.quepublishing.com/articles/article.asp?p=376255&seqNum=7&rl=1

[Accessed 25 July 2005].

Windley, P. (2005). “Paul Graham on open source and blogging”. ZDNet [Online], 2

August 2005. http://blogs.zdnet.com/BTL/index.php?p=1669&tag=nl.e539

[Accessed 4 August 2005].

Winslow, M. (2005). “Dipping Your Toes in Open Source”. Practical Open Source

[Online], July 2005. http://mariawinslow.com/articles/DippingToes.html [Accessed

19 July 2005].

Wynekoop, J. L. & Walz, D. B. (1999). "Characteristics of High Performing IT

Personnel: A Comparison of IT Versus End-User Perceptions". In: Prasad, J. (ed.)

129

Proceedings on the SIGCPR 1999 conference on Computer personnel research. 8-10

April 1999, New Orleans, USA. pp. 208-218. ACM Press, New York.

Ye, Y. & Kishida, K. (2003). “Toward an Understanding of the Motivation of Open

Source Software Developers”. In: Proceedings of 25th International Conference on

Software Engineering (ICSE2003), 3-10 May 2003, Portland, OR, USA. pp. 419-

429. IEEE Computer Society, Washington.

Yee, S. H. (2003). “Open Source in Small and Medium Size Enterprise Migration to

Linux”. In: Free & Open Source Software Conference (FOSSCON) 2003 [Online].

26-27 August 2003, Subang Jaya, Malaysia. Kuala Lumpur: MIMOS Open Source

R&D Group.

http://opensource.mimos.my/fosscon2003cd/paper/full_paper/seah_hong_yee.pdf

[Accessed 20 August 2005].

Zhang, W. & Storck, J. (2001). “Peripheral Members in Online Communities”. In:

Proceedings of the 7th international conference on Learning sciences. 27-July 01

2006, Bloomington, Indiana, USA. pp. 404-410. International Society of the

Learning Sciences.

Zuliani, P. & Succi, G. (2004). “Migrating Public Administrations To Open Source

Software”. In: Proceedings of e-Society 2004 IADIS International Conference, 16-19

July 2004, Avila, Spain. IADIS Press.

Zymaris, C. (2003). "A Comparison of the GPL and the Microsoft EULA"

Cybersource [Online], 5 May 2003.

http://www.cybersource.com.au/cyber/about/comparing_the_gpl_to_eula.pdf

[Accessed 24 September 2003].