Agility in Software Development and Project Value

Agile Development Methods, considered as an alternative to the traditional plan-based methods, have received much attention since their inception. These practices have evolved and developed over time, culminating in 2001 with the Agile Manifesto. Since that time, preferred methodologies, implementations, and best practices have continued to evolve with a focus on doing what works best for the individual company or project. However, the concept of agility in software development has remained quite nebulous, lacking in clarity particularly about its underlying dimensions. In this research the authors conceive agility in terms of four distinct dimensions. Drawing from the theoretical perspective of holographic organization, they develop a model explaining how each of these underlying dimensions of agility contributes to project value in software teams. The authors test the model using survey data collected from industry practitioners and discuss findings.

The Role of Functional Diversity, Collective Team Identification, and Task Cohesion in Influencing Innovation Speed

This article opens up the black box of innovation and examines the relationship between functional diversity in software teams and the often neglected dimension of innovation – speed, over the two phases of innovation: creativity and idea implementation. By combining information processing view and social identity theory, the authors hypothesize that when collective team identification is low, functional diversity positively affects the time spent in the creativity phase; however, when collective team identification is high, this relationship is inverted U-shaped. When task cohesion is high, functional diversity negatively affects the time spent in the idea implementation phase; however, when task cohesion is low, this relationship is U-shaped. Results from 96 IT software-teams confirmed the authors' hypotheses. Theoretical and managerial implications are discussed.

Translation of Activities in a Global Virtual Teams Software Development

. This study explores Global Virtual Software Teams’ development practices and try to demystify some of the misconceptions about global software development practices based on findings from the global virtual software teams’ experiment that was carried out at DePaul University from 2011 – 2018. The moments of translation from the perspective of actor-network theory (ANT) was employed in the data analysis, to examine how development approach was selected by the global virtual teams. One of the key findings from our research is that the success of a global software development project does not have a strong dependency on the development approach. While we agree that it is one of the key influencing factors, there are other equally strong factors for global virtual software team’s success.

On the use of hidden Markov model to predict the time to fix bugs.

A significant amount of time is spent by software developers in investigating bug reports. It is useful to indicate when a bug report will be closed, since it would help software teams to prioritize their work. Several studies of this problem have been conducted during the past decade. Most of these studies have used the frequency of occurrence of certain developer activities as input attributes in building their prediction models. However, these approaches tend to ignore the temporal nature of the occurrence of these activities. In this thesis, a novel approach using Hidden Markov Models and temporal sequences of developer activities is introduced. The approach is empirically demonstrated using eight years of bug reports collected from the Firefox project. The model correctly identifies bug reports with expected bug fix times. The approach is also compared against the frequency based classification approaches. The results indicate around 10% higher accuracy .

On the use of hidden Markov model to predict the time to fix bugs.

A significant amount of time is spent by software developers in investigating bug reports. It is useful to indicate when a bug report will be closed, since it would help software teams to prioritize their work. Several studies of this problem have been conducted during the past decade. Most of these studies have used the frequency of occurrence of certain developer activities as input attributes in building their prediction models. However, these approaches tend to ignore the temporal nature of the occurrence of these activities. In this thesis, a novel approach using Hidden Markov Models and temporal sequences of developer activities is introduced. The approach is empirically demonstrated using eight years of bug reports collected from the Firefox project. The model correctly identifies bug reports with expected bug fix times. The approach is also compared against the frequency based classification approaches. The results indicate around 10% higher accuracy .