Lessons learned by participants of distributed software development

The software industry is facing a paradigm shift towards distributed software development (DSD). This change creates situations from which organizations may benefit, and challenges to which they must adapt to (e.g. the absence of opportunities for informal interaction). In this paper we present the results of an analysis of the knowledge flow problems which hinder members of a DSD team to get into synchronous collaboration. We used the KoFI methodology to identify knowledge sources, topics, and existing flows, and to identify knowledge flow problems in the DSD environment. These have been used as a basis for the design and development of CWS-IM, an extended instant messenger tool which implements "Selective Availability" to facilitate "l'entree en collaboration" of the members of a DSD team. Therefore, it is expected that the interruptions during DSD will be less disturbing as this tool helps users to decide when it is the most suitable moment to establish synchronous contact for the sender and the receiver.

Software organizations are increasingly combining agile methodologies and distributed software development (DSD) for efficiently and effectively built software products. There are numerous challenges associated with scaling agile methods in a distributed environment. Our study is intended to explore and prioritize the challenges for scaling agile practices in the DSD environment. This study was divided into 3 stages. In the first stage, 22 challenges were identified from literature review and grouped into 4 categories: management, team, technology, and process based on discussion with experts. In the second stage, an online questionnaire study was conducted to validate the identified challenges. Finally, analytic hierarchy process method was used to prioritize challenges and their categories based on their relative importance. The results highlighted that management is a most significant category as compared with the other categories. Similarly, lack of management commitments, lack of effective communication, lack of knowledge sharing, etc are identified as the most significant challenges that need to be focused by the organizations for scaling agile methodologies. On the basis of the research findings, we could conclude that the identified challenges along with their categories provide a robust framework to scale agile methodologies in the DSD environment.

agile information systems, flexible information systems development, distributed information systems development, global software development, software development, special issue

Age stereotypes in distributed software development: The impact of culture on age-related performance expectations

Teamwork in a software engineering project includes different coordination aspects. Usually many different workflows are implied by a software development process model,which prescribes — among other things — the general course of the development process and all demanded deliverables such as design and system documents or source codes. During the whole course of a development project, communication support and efficient management of the flow of work items between different people or between different groups of people is required.

Purpose – The aim of this paper is to make an attempt to find good values of onsite–offshore team strength; number of hours of communication between business users and onsite team and between onsite and offshore team to reduce cost and improve schedule for re-engineering projects in global software development environment. Design/methodology/approach – The system dynamics technique is used for simulation model construction and policy run experimentation. The experts from Indian software outsourcing industry were consulted for model construction, validation and analysis of policy run results in both co-located and distributed software development environment. Findings – The study results show that there is a drop in the overall team productivity in outsourcing environment by considering the offshore options. But the project cost can be reduced by employing the offshore team for coding and testing work only with minimal training for imparting business knowledge. The research results show that there is a potential to save project cost by being flexible in project schedule. Research limitations/implications – The study found that there could be substantial cost saving for re-engineering projects with a loss of project schedule when an appropriate onsite–offshore combination is used. The quality and productivity drop, however, were rather small for such combinations. The cost savings are high when re-engineering work is sent to offshore location entirely after completion of requirement analysis work at onsite location and providing training to offshore team in business knowledge The research findings show that there is potential to make large cost savings by being flexible in project schedule for re-engineering projects. Practical implications – The software project manager can use the model results to divide the software team between onsite and offshore location during various phases of software development in distributed environment. Originality/value – The study is novel as there is little attempt at finding the team distribution between onsite and offshore location in global software development environment.

In large projects, software developers struggle with two sources of complexity - the complexity of the code itself, and the complexity of the process of producing it. Both of these concerns have been subjected to considerable research investigation, and tools and techniques have been developed to help manage them. However, these solutions have generally been developed independently, making it difficult to deal with problems that inherently span both dimensions. We describe Augur, a visualization tool that supports distributed software development processes. Augur creates visual representations of both software artifacts and software development activities, and, crucially, allows developers to explore the relationship between them. Augur is designed not for managers, but for the developers participating in the software development process. We discuss some of the early results of informal evaluation with open source software developers. Our experiences to date suggest that combining views of artifacts and activities is both meaningful and valuable to software developers.

In this paper, we present a management system for distributed software development in cloud computing environments, called DDMan. Cloud computing environments provide more flexibility than conventional computing environments. In particular, the platform as a service (PaaS) in cloud computing environments provides benefits for users in a variety of aspects, such as application design, development testing, software deployment, team collaboration, web service integration, database integration, and scalability. Although the process of distributed software development evolves and changes by a lot in the past decade, there are still some user needs remaining unsatisfied. For managing distributed software development in cloud computing systems, we present the DDMan system, which extends the WebSD model of distributed software development management. In DDMan, the interactions among the roles in the software development are precisely defined. In addition to the conventional web-based interface, the DDMan system also supports the interface for the Amazon cloud environment, which is an approach to a fully automatic service of distributed software development management in cloud computing environments.

Along the last decade, many companies started using Distributed Software Development (DSD). The distribution of the software development teams over the globe has become almost a rule in large companies. However, in this context, new problems arise, which mainly involve the physical and temporal distance among the participants. Some studies show that deploying a version control system to alleviate this problem is a big challenge for distributed teams. This paper presents a systematic mapping study about works about version control that focus on DSD. We found 29 studies related to DSD version control, published between 2002 and 2012. Using the systematically extracted data from these works, we present challenges, tools, and other solutions proposed to version control in DSD. These results can support practitioners and researchers to better understand and overcome the challenges related do DSD version control, and devise more effective solutions to improve version control in a distributed setting.

The geographically distributed software development has attained a significant attraction in software development industry since last few decades due to its wide acceptance across the globe. The main essence of geographically distributed software development is to expand company working by follow-the-sun strategy, get skilled labors at cheaper rates from low economy regions of world and pool-up company resources to raise their utilization graph. The requirements elicitation process is considered as one of the most challenging tasks during software development process in traditional as well as geographically distributed software development projects due to its social and collaborative nature. This article presents a comprehensive insight about major challenges faced by requirements elicitation teams engaged in geographically distributed software development projects. The article also elaborates the possible strategies, which can be adopted to minimize the negative impact of geological distribution of software development activities. The presented study will help the requirements elicitation teams to better plane their requirements elicitation sessions for geological distributed software development projects by encountering the expected hurdles.


 
 
 
 Several years ago software development has become critical to the global market. In the past decade, as a reflection of globalization, software companies began to distribute their development processes in different places, creating distributed software development (DSD). With the growing of Distributed Software Development, organizations have attempt to sketch the best possible development structure, in order to improve productivity and quality. The aim of this paper is to present a Systematic Review of Literature to identify which ways of collaboration are commonly used by software organizations where teams are temporal and geographically dispersed, with also different native languages and culture. Further, the research was based on the basic life cycle of the traditional development, and where phases of the project are performed: onsite, distributed/offshore and multi-site. The systematic review have examined 868 papers published between 2000 and 2010.
 
 
 

Distributed Software Development (DSD) has become a common approach for many organizations nowadays. This model of work enables better completive advantages, by reducing the overall development costs, shortening the time-to-market and enabling the access to professional resources in a global scale, among others. But there are challenges to be faced in DSD, and Team Motivation is an approach for dealing with its issues. This work has the objective of perform an empirical study on eleven motivational factors proposed by Farias Junior et al. for DSD teams. To achieve this goal, an online survey form was proposed for DSD professionals and researchers, to obtain their perception on the relevancy of these motivational factors. The results include the opinions of the participants and our analysis on these findings. DSD professionals and researchers can benefit from this work by using the findings as a reference for adoption of these motivational factors within their development projects or as a reference for future studies.

Software development in a context where development locations are separated across geographical regions, either nearby or internationally, is quickly becoming a popular practice. This software development agreement is often mentioned as Global Software. "Global Software Development (GSD) is also known as Distributed Software Development (DSD) or Global Software Engineering (GSE)". Plans carried out by a scattered squad have been identified by way of a more dangerous and hard enterprise than projects carried out by teams working below a similar roof. As a result, considering the possible obstacles and appropriate mitigation techniques for GSD is critical for operating a successful project. Software development is divided into stages of requirements, analysis, designs, coding, and testing. The creation of software in globalized settings is frequent and significant in today's world of software development. This paper emphasizes the condition of software product transmission, which deals with software testing in an overseas site, as part of numerous globalized scenarios.

As more software development organizations are increasingly distributing their operations spatially, information systems development researchers are taking perspectives such as transactions costs and resource dependency to explain the effects of spatial distribution on coordination. This paper argues that these perspectives are limited because they do not address all the key relationships between software development resources in a unified and systemic manner. The interactions between people, information, and technology, which are the key software development resources, are characterized by four key relationships – interdependencies, uncertainties, conflicts, and technology representations. Based on the premise that spatial distribution of the resources exposes their relationships to direct environmental stimuli, the paper proposes a complementary resource relationships perspective to explain the spatial effects on the dynamics of coordination. An empirical example of software development distributed between sites in USA and Republic of Ireland is used to illustrate these issues. Analyzing the effects of spatial distribution in terms of resource relationships leads to a more systemic, in-depth and phenomenological understanding of the dynamics of coordination. Implications of this perspective for the analysis of coordination in distributed software development are discussed.

Software engineering project courses where student teams are geographically distributed can effectively simulate the problems of globally distributed software development (DSD). However, this pedagogical model has proven difficult to adopt or sustain. It requires significant pedagogical resources and collaboration infrastructure. Institutionalizing such courses also requires compatible and reliable teaching partners. The purpose of this workshop is to continue building on our outreach efforts to foster a community of international faculty and institutions committed to developing, teaching and researching DSD. Foundational materials presented will include pedagogical materials and infrastructure developed and used in teaching DSD courses along with results and lessons learned. The third CTGDSD workshop will also focus on publishing workshop results and collaborating with the larger DSD community. Longrange goals include: lowering adoption barriers by providing common pedagogical materials, collaboration infrastructure, and a pool of potential teaching partners from around the globe.