Virtual Teams In Global Software Development

Published Date: 02 Nov 2017

Floris Erich

Jurjen Nienhuis

Arjan Staring

University of Twente

Abstract

The development of software by virtual teams with members around the globe is a hot topic for researchers everywhere. Global software development by virtual teams differs in many aspects from development of software by traditional teams. Those differences come with challenges and need to be overcome. This paper aims to research those challenges, find ways to overcome them and elaborate on the impact of these findings in global software development. The research focuses on two major software development paradigms, each of them differ in the approaches to the development process and oppose their own challenges and ways to address them. Research in this paper shows how these software development paradigms hold for different aspects in the global software development. The paper concludes with a list of challenges and ways to address them for each software development paradigm in global software development.

Introduction

Virtual teams in global software development are a hot topic nowadays. One of the prime examples of global software development taking place is within the Linux community. Linux distributions consist of various software packages, which all have different developers. These developers might be colleagues working together at a company, academics working together at a university, or they might not be working together in the same physical location at all. When these projects are distributed communication often takes place using media such as e-mail, with newsletters, chat, with for example internet relay chat and bulletin boards.

These projects form a virtual team, virtual teams can formally be defined as "A team whose members use the Intranet, Intranets, Extranets and other networks to communicate, coordinate and collaborate with each other on tasks and projects even though they may work in different geographical locations and for different organizations" (O’Brien, 2002).

This global development of software by virtual teams differs in many aspects from the development of software by teams who sit next to each other and can meet face to face. In global software development, challenges do not only exist in working in a virtual team, but also in the development process. The widely used software development paradigms have been thought out as to be used by traditional teams. The use of such software development paradigms by virtual teams raises its own challenges.

This research will focus on the challenges and ways to address them of virtual teams in global software development and the software development paradigms used by those virtual teams.

Research outline

In continuation to the introduction, this paper aims to answer the following research question:

How can the challenges of virtual teams in GSD be addressed in the major software development paradigms?

In order to answer the above research question, the mentioned major software development paradigms need to be specified. Paradigms can be placed on a spectrum which is believed to start with traditional software development and ends with agile software development. Often a combination of both is used. By taking the extremes the largest portion of characteristics is covered.

To give a well-structured answer to the research question, it has been split in four sub research questions:

What are the main concerns for virtual teams in global software development?

What are the challenges related to these concerns?

How to address these challenges?

What effect will these challenges have in the traditional and agile software development paradigms?

To answer each of the sub research questions, a literature study will be done. In section 3 research will be done in order to answer sub research question one by constructing a list concerns for virtual teams in global software development (GSD). Section 4 will explore the concerns found in section 3 for challenges and ways to address them in order to answer sub research question 2 and 3, this will largely be based on a literature study. Sub research question 4 will be answered in section 5 by analyzing the challenges and ways to address them in relation to the traditional and agile software development paradigms. This paper will conclude with a conclusion to recapture on the main and sub research questions. A description of the cases used to construct the list of challenges and its ways to address them as to answer sub research question 2 and 3 can be found in appendix A.

Main concerns

In this section, Global Software Development is introduced using various models found in the literature studied. From these models the main concerns for virtual teams in Global Software Development are identified.

In the literature study three models of Global Software Development have been found, those of Carmel (1999), Karolak (1998) and that of Evaristo, Scudder, Desouza, & Sato (2003). The model of Karolak (1998) does not try to describe the problems virtual teams face but rather describes a GSD approach following the traditional project life cycle, starting with predevelopment, continuing with requirements, design, code, test, and finalizing with customer delivery and maintenance. Therefore only the models of Carmel and Evaristo et al. are used.

Carmel (1999) proposes a model of software development which virtual teams can use when going global. In his model teams stand in the center of centrifugal forces which reduce the team’s effectiveness (figure 1). To oppose these centrifugal forces he proposes centripetal forces which should bring the team closer together no matter the physical distance (figure 2).

Figure : Carmel's centrifugal forces. Based on: Prikladnicki et al. (2003)

Figure : Carmel's centripetal forces. Based on: Prikladnicki et al. (2003)

Evaristo, Scudder, Desouza, & Sato (2003) present a dimensional model (figure 3) which can be used to measure how distributed a team is, and could be used to tackle some of these dimensions.

Figure : Model of Evaristo et al.. Based on: Prikladnicki et al. (2003)

By combining the models of Carmel (1999) and Evaristo, Scudder, Desouza, & Sato (2003) on global virtual teams and problems observed in the case studies found in appendix A, a list of concerns has been established. It is believed that these concerns are areas which need attention to ensure the proper functioning of a virtual team. When the virtual team is functioning well, the efficiency will increase as well, which eventually will reduce costs.

The following concerns have been identified:

Communication

Coordination

Collaboration

Culture differences

Human factors: Relationship building, trust, psychological safety

In each of the models these concerns can either be a single component of the models or cross-cutting. The following table shows what applies to each concern. These concerns also match the categories mentioned in the lecture.

Concern

Carmel

Everisto et al.

Communication

Loss of communication richness

Cross-cutting

Coordination

Coordination breakdown

Synchronicity

Collaboration

Cross-cutting

Cross-cutting

Culture differences

Cultural differences

Culture

Human factors

Loss of "teamness"

Trust

Challenges related to the main concerns

In this section, analyzes will be done on the concerns identified in the previous chapter. The challenges, and how to address them, will be identified per concern.

Communication

In this subsection the concept of communication in virtual teams in Global Software Development is introduced. Some of the challenges virtual teams face when communicating globally will be highlighted and some solutions the current literature offers will be discussed. In the cases various challenges in regard to communication were noticed.

Because of the geographic dispersion people do not share the same space anymore and have to communicate virtually. When someone is not physically in the same space, it’s easier to forget tasks related to that person. Team members often forget to communicate changes made to the system during the day, leading to duplicate work (Ramesh & Dennis, 2002). This is called the principle of out of sight, out of mind.

Access to the customer could become complicated for team members because the communication structures involved the team manager as an intermediary for the communication with the customer. (Ramesh & Dennis, 2002).

Face-to-face communication poses a higher barrier in Global Software Development than in regular software development. According to Nardi & Whittaker's (2002) characterization of the unique aspects of face-to-face communication; face-to-face communication supports touch, shared activities, eating and drinking together, as well as informal interactions and attention management. They argue that these activities are crucial for sustaining the social relationships that make distributed work possible. According to Carmel (1999), the force affecting communication is the loss of communication richness.

Several things can be done to address the challenges mentioned above. The geographical dispersion can be addressed by bringing people physically together for important meetings. In several of the cases there was either a team member frequently flying over to the customer or a team member permanently on site.

The effectiveness of communication structures can be increased by introducing collaborative technology as a new platform for communication to take place on and by teaching new managerial techniques to strengthen communication between a manager and employees (Carmel, 1999). Distributed communication can also be introduced into the development methodology, by for example formalizing the moments at which communication takes place. More effort in establishing a product architecture will aid in making communication more clear.

The communication barrier can be reduced by implementing a telecommunications infrastructure. This will enable distributed teams to communicate face-to-face over the internet, but it will not realize all the aspects of face-to-face communication characterized by Nardi & Whittaker (2002). Another way to reduce the communication barrier is to schedule team building activities.

Ramesh & Dennis (2002) propose a new type of organization for global virtual teams, called the object-oriented team. In a traditional virtual team tight coupling between team members is pursued. An object-oriented team should actually be decoupled through the use of processes and rich media clarifying, extending and constraining meaning. While this approach does not solve any of the challenges, it does reduce the amount of communication needed.

To summarize, the main challenges for virtual teams in global software development regarding communication are geographical dispersion, communication structures not being adapted to GSD and a higher communication barrier. Geographical dispersion leads to the principle of "out of sight, out of mind" occurring and loss of communication richness. The main solutions for these challenges are isolating work, team building and implementing a telecommunications infrastructure.

Coordination

When teams are working in the same project, but are geographically dispersed, the coordination of the team is really important and is one of the main concerns. Team coordination can be defined as "the effective management of the mutual dependencies between members of the team" (Van Bezooijen, 2011). In this definition of team coordination, the geographical dispersed factor is not mentioned. For coordination of virtual teams, team coordination can be defined as: ‘the effective management of the mutual dependencies between geographical dispersed members of the team’.

If a virtual team is not coordinated well, problems may arise on the division of tasks or schedules. These problems are caused by the large distances introduced (Casey & Richardson, 2006). When problems arise on for example the division of tasks, team members may do the same tasks or they might expect another team member to do the task. If the one responsible for the coordination does not notice this fast, the results might be catastrophic as development costs may pass the budget or the software may not be delivered on time.

To solve the coordination issue, just putting more time on the coordination is not a guarantee that the coordination problem is solved. Casey & Richardson (2006) suggest that to be able to coordinate a virtual team "policies and procedures needed to be drawn up for the establishment and operation of the virtual software teams that ensured visibility into their activities and operation at both locations, including roles and responsibilities." When setting these policies and procedures, the cultural diversity should not be forgotten, because they can have impact on the operation of the project.

Ebert & Neve (2001) have experience with two development sites at Alcatel. They developed some best practices for managing these both development sites. Although here are two teams involved, which are geographically dispersed, this could be compared with virtual teams whose members are geographically dispersed. The best practices have been analyzed and mainly focus on planning, responsibility and structure in a virtual team.

Summarizing the concern regarding coordination, the major challenges are the division of tasks and scheduling. These challenges can be solved by introducing policies and procedures to ensure visibility of activities and operation, including roles and responsibilities.

Collaboration

The entire purpose of a team is to collaborate since it exists to achieve a mutual goal. Collaboration can be defined as the process of working together to achieve a shared goal. Teams that work collaboratively can obtain greater resources, recognition and reward when facing competition for finite resources (Jablokow & Myers, 2010). For successful collaboration, coordination, communication and a specific form of behavior are required and can be supported by various collaboration tools and techniques. Most issues in respect to coordination and communication have been addressed earlier in this paper and as they affect the team, they directly affect the collaboration process.

Since members are geographically dispersed and one is not able to meet face to face (most of the time), it becomes very hard for members to get to know each other. Therefore it is important to design activities that cause people to get to know each other (Niederman & Tan, 2011; Setamanit & Raffo, 2008), despite the fact that they cannot actually meet face to face.

In order to avoid misperception and miscommunication, but also to improve the way in which the team is able to collaborate, the team needs to agree on standards and terminology to use as well as to embed collaboration technology into everyday work (Niederman & Tan, 2011).

As a team, it is obvious to establish common goals and objectives, but also common rewards (Casey & Richardson, 2006). Unfortunately a pat on the shoulder is not an option, so a new reward structure is needed to create incentives to focus on tasks, as well as to find new ways to focus attention (Niederman & Tan, 2011).

Numerous other point of attention has shown to be effective in improving the collaboration in a virtual team. Members of a team need to be more explicit in their actions, but also need to be trained to self-facilitate since members cannot help each other immediately (Niederman & Tan, 2011). Lastly it is important to implement training and mentoring to help the team evaluate and stay focus on the job (Casey & Richardson, 2006).

In summary in order for a virtual team to collaborate well, it has to overcome some challenges opposed by the fact that the team is dispersed as it hard to get to know each other and content can be misinterpreted by lack or misperception of context. To resolve the challenges, one needs to find way to make communication, goals and objectives more explicit, formalize standards and technology and alter certain aspects which stimulate work, like the reward system in such a way that it is suitable for a global virtual team.

Cultural differences

It is clear that participating in a virtual team has its challenges, but participating in a global virtual team brings its own challenges. A global virtual team consists of members across the globe, each of them bringing their own cultural differences into the team. These cultural differences can build up an obstacle to the performance of the virtual team. Not only can the difference in language or even dialect, but also particular uses and behavior affect the understanding between team members.

Cultural differences, for example difference in language, may cause misperception and miscommunication. This is likely to result in failure as well as to reduced trust (Chinbat, 2010; Patel, Lawson-Johnson, & Patel, 2009). Cultural expectation influence how members relate to each other (Chinbat, 2010; Patel et al., 2009) and are for example used to work together or used to have a ‘team spirit’.

By giving trainings, including a cultural context, and stimulating information sharing people can get to know each other’s background. This will improve cultural compatibility and similarity (Maznevski, Chudoba, & Robey, 2000; Prikladnicki, Nicolas Audy, & Evaristo, 2003). The cultural differences will can be understood and decision can be made easier (Patel et al., 2009; Setamanit & Raffo, 2008).

Summarizing the cultural difference concern, it can be concluded that cultural differences may cause misperception and miscommunication. By having a training on the cultural differences and a stimulation for sharing information, the number of misperceptions and miscommunications can be reduced. This has a positive influence on the decision making and does not form an obstacle for performance.

Human factors

The human factors concern is a union of the concerns related to the interaction between individuals: relationship building, psychological safety, and trust. Relationship building and psychological safety together form the loss of teamness force in Carmel’s centrifugal model.

Each of these concerns will be discussed shortly.

Relationship building

Relationship building is not the same in virtual teams as it is in ‘normal’ teams. Relationships cannot be separated from communication and the nature of the relationship is defined by the communication between its members. Because of the geographical dispersion, it is hard to build up a good relationship.

A good relationship in virtual teams positively influences the task performance and effectiveness. Therefore it is quite important to have a good relationship. Research shows that relational links are associated with enhanced creativity, motivation, increased morale, better decisions, and fewer process losses. It is easier to complete relationship-building activities in a face-to-face context than in a strictly virtual one (Pauleen, 2003)

Jarvenpaa & Leidner (1999) describe how virtual team members may be able to accomplish tasks without first having developed relationships. If a team member feels confident there will be a "payoff' for cooperating with and trusting virtual team members, then he or she will do so. Such trust appears to be fragile and temporary however. This is where the role of a team leader becomes important (Pauleen, 2003).

If it is possible for the virtual team to meet in person, the team should mainly focus on relationship building, to strengthen the team in the beginning (Maznevski et al., 2000). When this is not possible virtual teams should be stimulated to have social conversations as well to achieve higher trust (Jarvenpaa & Leidner, 1999) and better social and emotional relationships (Maznevski et al., 2000). Leaders can stimulate relationship building by scheduling regular chat sessions for socializing with all team members present and using humor to lighten moods (Kayworth & Leidner, 2000). The possibilities of improving the relationship building are merged by Powell, Piccoli, & Ives (2004).

Trust

An important aspect of relationship building is developing trust. "Trust allows people to take part in risky activities that they cannot control or monitor and yet where they may be disappointed by the actions of others. The problem with virtual teams is that there is a lack of shared social context. Factors which should contribute to trust are shared social norms, repeated interactions and shared experiences (assumed by Jarvenpaa & Leidner, 1999).

Jarvenpaa & Leidner (1999) concludes two important recommendations for building trust in virtual teams. The first one is that for the manager of a virtual team, is that a clear definition of responsibilities must be given at an early stage. The manager must also ensure that team members have a sense of complementary objectives and share in the overall aim of the team.

The second recommendation is when a conflict occurs, this should be handled effectively. Emotions which remain undiscussed might cause conflicts within the team afterwards. Furthermore, participants on virtual teams should be open minded at the beginning of the team formation. The participants should also be aware that it is the quality and predictability which is most critical to the communication.

Psychological safety

Psychological safety can be defined as "a shared belief that the team is safe for interpersonal risk taking"… and "a sense of confidence that the team will not embarrass, reject, or punish someone for speaking up" (Edmondson, 1999). Gaudes' (2010) findings show that psychological safety enables collective learning (Edmondson, 2004), increases individual engagement with work (May, Gilson, & Harter, 2004), reduces negative effects of geographical dispersion and electronic dependence on innovation (Gibson & Gibbs, 2006) and enables incremental and radical innovation (Un, 2010).

As Edmondson (1999) described, collective learning may result in psychological safety for team members and effective team performance. But in virtual teams, where the community of practice is geographically distributed and temporally disconnected, opportunities for visual, tactile, and verbal communication are limited. Virtual arrangements potentially threaten the process of situated learning because members are not located together (Robey, Khoo, & Powers, 2000). This will have a negative effect on the psychological safety.

Based on Gaudes (2010), team leaders should do several things. They should make the importance of team tasks clear, ensure the team is tasked with clear goals and responsibilities for assignments, provide the team with collaboration norms to assist in effectively contributing to the team and take necessary action to ensure team effectiveness.

Summarizing the chapter of human factors, relationship building gets complicated because of the geographical dispersion, trust is needed for relationship building but building trust gets complicated because of a lack of shared context and finally psychological safety is negatively affected by team members not being located together.

Relationship building can be improved by dedicating a portion of the time virtual teams meet in person to this very purpose and by stimulating social and emotional relationships between team members. Trust can be improved by giving a clear definition of responsibilities at an early time and effectively handling conflicts by implementing conflict management. Psychological safety can be improved by the team leader who has to make the importance of tasks, goals and responsibilities clear, has to provide the team with collaboration norms and has to take necessary actions to ensure team effectiveness.

Analysis by software development paradigms

In this chapter the challenges found within each concern will be matched with the corresponding paradigms. These paradigms are the extremes of the traditional-agile software development spectrum. In the introduction several characteristics of each paradigm have been identified used to analyze whether the challenges are applicable to a specific paradigm or whether they apply to virtual teams in global software development in general.

Traditional software development paradigm

Traditional software development has a predictive approach in which large parts of the software development processes are planned in detail in advance and has comprehensive documentation where requirements documents play a key part of the documentation. Traditional software development is process oriented in such way that the process is well defined and specific tools are used for the completion of each task (Awad, 2005). Some well-known methodologies based on the traditional software development paradigm waterfall, spiral model and unified process.

When looking at the communication concern in relation to this paradigm, two aspects can be noticed. The traditional software development paradigm may already formalize the communication, because of its comprehensive documentation. But, communication structures of the virtual team may not have been adapted to GSD, which may result in a conflict between the communication structure and process. The main solutions for these challenges are isolating work, team building and implementing a telecommunications infrastructure.

Because of the nature of the traditional software development paradigm, the development is planned in detail and comprehensive documentation exists. Therefore the problems on the division of tasks or schedules may not play an important role for the coordination of virtual teams in global software development.

Just like the coordination concern, from a collaboration concern perspective standards and technologies to be used may have been agreed on in traditional software development paradigm. Because this agreement has been made as part of the paradigm, the collaboration concern is less important in it.

From a cultural concern perspective, no specific issues related to traditional software development paradigm exist. The problems with the cultural concern are rather related to virtual teams themselves. By facilitating training and mentoring the virtual team members can overcome these challenges.

As is the case for the cultural concern, the human factors concern is not related to just one paradigm, but to virtual teams themselves. When the social context of the virtual team is missing the members cannot function to their best capabilities. The challenges can be resolved by scheduling time for socializing (face-to-face or chat), defining goals and responsibilities and resolving conflicts.

Agile software development

Agile software development is a set of properties for a software development process devised in 2001 by a group of expert practitioners, based on existing methodologies such as XP, Scrum, DSDM, Adaptive Software Development, Crystal, etc. (Highsmith, 2001). New methodologies such as Agile Unified Process were introduced after the existing methodologies on which agile is based.

The agile manifesto declares the intentions, motives and views of these practitioners and reads as follows (Beck et al., 2001):

We are uncovering better ways of developing software by doing it and helping others do it.

Through this work we have come to value:

Individuals and interactions over processes and tools

Working software over comprehensive documentation

Customer collaboration over contract negotiation

Responding to change over following a plan

That is, while there is value in the items on the right, we value the items on the left more.

Agile can be seen as a bottom-up approach, originating from a manifesto signed by practitioners (Beck et al., 2001).

Commenting on the communication perspective in relation to the agile software development paradigm, agile does not put an emphasis on the structure of the team and process. Therefore the communication structures can be adapted easily to new circumstances, unlike in traditional software development. Hence the communication concern is less important when dealing with agile software development.

Unlike effects of the coordination concern in traditional software development, coordination in agile software development is very important. Not so much is specified about the process, as plans may change over time. It is suggested to improve or make people aware of planning, responsibility and structure in a virtual team.

Looking at the collaboration concern, it is not in line with the agile software development paradigm to agree on standards and technologies to be used early on. But to improve collaboration it is advised to come to an agreement on standards and technologies to be used at an early stage in the project. Collaboration in agile virtual teams can furthermore be improved by implement a new reward structure, training and mentoring.

Just like in traditional software development paradigm, in agile software development paradigm virtual teams cultural differences are one of the concerns as well. The problems with the cultural concern are rather related to virtual teams themselves. By making members more aware and educating in cultural differences virtual team members can overcome these challenges.

The human factors concern has many similarities with the cultural differences concern in the sense that it is a concern not related to a specific software development paradigm. The challenges can be resolved by scheduling time for socializing (face-to-face or chat), defining goals and responsibilities and resolving conflicts.

Conclusions

When this research was started the question was posed how the challenges for virtual teams in GSD could be addressed. After that the following questions were defined to guide the research:

What are the main concerns for virtual teams in global software development?

What are the challenges related to these concerns?

How to address these challenges?

What effect will these challenges have in the traditional and agile software development paradigms?

The main research question is: How can the challenges of virtual teams in GSD be addressed in the major software development paradigms?

Sub questions 1 will have its own subsection, while sub question 2 and 3 will have be handled in the same sub section, because they complement to each other. Sub question 4 and the main research question will also be handled in the same sub question because they are complementary as well.

What are the main concerns for virtual teams in global software development?

When software development goes global, teams become virtual and get challenged in unique new ways. Looking at the models of Carmel and Everisto et al. five major concerns for virtual teams in Global Software Development were identified: communication, coordination, collaboration, culture differences and human factors (relationship building, trust and psychological safety). Within GSD there are two major paradigms: Traditional software development and agile software development.

What are the challenges related to these concerns? And how can these challenges be addressed?

The main challenges for virtual teams in global software development regarding communication are geographical dispersion, communication structures not being adapted to GSD and a higher communication barrier. Geographical dispersion leads to the principle of "out of sight, out of mind" occurring and loss of communication richness. The main solutions for these challenges are isolating work, team building and implementing a telecommunications infrastructure.

Summarizing the concern regarding coordination, the major challenges are the division of tasks and scheduling. These challenges can be solved by introducing policies and procedures to ensure visibility of activities and operation, including roles and responsibilities.

Summarizing the collaboration concern, in order for a virtual team to collaborate well, it has to overcome some challenges opposed by the fact that the team is dispersed. To resolve the challenges, one needs to find way to make communication, goals and objectives more explicit, formalize standards and technology and alter certain aspects which stimulate work, like the reward system in such a way that it is suitable for a global virtual team.

Summarizing the cultural differences concern, it can be concluded that cultural differences may cause misperception and miscommunication. By having a training on the cultural differences and a stimulation for sharing information, the number of misperceptions and miscommunications can be reduced. This has a positive influence on the decision making and does not form an obstacle for performance.

When looking at human factors, relationship building gets complicated because of the geographical dispersion, trust is needed for relationship building but building trust gets complicated because of a lack of shared context and finally psychological safety is negatively affected by team members not being located together.

Relationship building can be improved by dedicating a portion of the time virtual teams meet in person to this very purpose and by stimulating social and emotional relationships between team members. Trust can be improved by giving a clear definition of responsibilities at an early time and effectively handling conflicts by implementing conflict management. Psychological safety can be improved by the team leader who has to make the importance of tasks, goals and responsibilities clear, has to provide the team with collaboration norms and has to take necessary actions to ensure team effectiveness.

What effect will these challenges have in the traditional and agile software development paradigms and how can they be addressed?

Communication is often formalized in traditional software development but might not have been adapted to GSD. This is a major concern. As communication is allowed to take place more freely in agile software development communication structures grow organically, making it a minor concern.

For traditional software development, the communication concern does not play a large role when communication is adapted to GSD. When it has not been adapted, the challenge of being able to communicate easily can be addressed by for example introducing collaborative technology as a new platform for communication.

Traditional software development puts a lot of emphasis on planning and documentation. This greatly aids in supporting coordination in software development using that paradigm. Agile software development lacks this emphasis (instead the emphasis is on starting with programming early). The changing nature of an agile project results in the need of extra care to be taken to ensure proper coordination when using agile software development in GSD.

Challenges related to the coordination concern, which are about the division of tasks or schedules do not play a large role as much documentation exists in tradition software development.

In traditional software development decisions about standards and technologies are made at an early state in the project. This reduces the collaboration concern. In agile software development responding to change is a core value; hence these decisions are pushed back in time as far as possible. This hinders collaboration however, as for proper collaboration to take place standards and technology have to be shared.

Challenges related to the collaboration concern, which are about misperception, miscommunication and the way in which the team is able to collaborate. In traditional software development the ways of working together are already defined, but still not all can be defined in front. Therefore, teams should get to know each other to get to know the way they work together.

Human factors are, just like cultural differences, are of equal influence to virtual teams using the both traditional and the agile software development paradigms. The characteristics of both traditional and agile software development do not make any difference in these concerns.

Human factors, just like cultural differences, cause challenges on how to get the team work a good together as possible. The human factors cause the relationship between the members, while cultural differences have influence on how communication occurs. These challenges are important to address in tradition software development. By having a training on the cultural differences and a stimulation for sharing information, the number of misperceptions and miscommunications can be reduced. psychological safety is negatively affected by team members not being located together. To increase human factors, it is advised to dedicate social time for the virtual teams and by defining and making clear the responsibilities.

Future work

Various topics have been covered in this paper, but work has still to be done. This paper has focused on the challenges in traditional and agile global software development in virtual teams, but many more software development paradigms exist and need research in relation to global software development.

Other research need to be done to design a software development paradigm specifically for global software development to tackle the challenges of traditional and agile software development in global virtual teams. Research also needs to be done to virtual teams in terms of dimensions. It could be argued that a team on the same location but separated by time (day and night shifts) be considered a virtual team, the same applies to a team with members distributed throughout the same building, but working simultaneously.

Acknowledgements

While writing this research paper we profited greatly from the safe setting of the university, where we could bring out multiple versions of the paper and receive reviews from teachers and fellow students. We would like to thank Klaas Sikkel for giving us clear directions on where to take our research and a lot of feedback to improve our work. We would also like to thank all the students who gave us directions and feedback on our paper and presentation, especially the students who wrote reviews of our draft paper; Jacco Roest, Jurriën Wagenaar, Niels Witte, Pim Dietz, Laurens Hellemons and Milan Schramm.