Jorge A. Colazo

Performance implications of stage-wise lead user participation in software development problem solving

The problem-solving view of new product development sees the innovation process as a series of problem-solving loops broken down into three stages: problem detection, analysis and removal. We link this framework with lead user-driven innovation regarding software and show that effort by lead users (LUs) in each stage of the innovation problem solving process is, in varying degrees, associated with the source codes quality, the productivity of the development process and the softwares popularity. We also test whether front loading the problem solving process is associated with development performance and we find that front loading is associated with increased code quality but decreased development productivity. Empirical tests are carried out with data from open source software projects. Findings potentially impact the design and management of online communities to help product development. The paper addresses software development seen as a series of problem solving loops helped by user input.Effort in user-led problem analysis is related to better product quality.Front loading the problem solving process yields better software quality.Engaging users in either problem solving stage always negatively impacts productivity.

Structural Changes Associated with the Temporal Dispersion of Teams: Evidence from Open Source Software Projects

This study relies on Media Synchronicity Theory and Social Network Analysis to analyze how the structure of collaboration networks change when collaborating teams become temporally dispersed. The empirical test of hypotheses using ordinary least squares with archival data from 230 Open Source Software projects shows that the collaboration structure networks of more temporally dispersed teams are sparser and more centralized, and these associations are stronger in those teams exhibiting higher relative performance.

Exploring the Association between Temporal Dispersion and Virtual Team Performance

A relatively new research stream inquires into phenomena related to temporal separation, or temporal dispersion (TD) in distributed virtual teams. There is little empirical work to date on the association of TD with different indicators of team performance. This paper explores the relation between TD, product quality and product development speed, using open source software development teams as research framework. TD is defined and operationalized into two dimensions: coverage and overlap, and an exploratory regression model is tested on data collected from multiple archival sources comprising 100 open source software development distributed project teams. Results show that TD has a negative association with software interrelease time and that TD interacts with product complexity in its association with product quality.

Team Network Evolution While Undergoing a Lean Transformation: A Missing Success Indicator?

The Lean paradigm for production process management and improvement can be considered a benchmark for managing operations in many areas, from manufacturing to software development. Several studies found factors for the success or failure of Lean implementations, but these factors are aggregated or company-level and do not capture the nuances of teamwork at the shop floor level. In this paper it is contended that as companies embark along a lean transformation their work teams undergo a change in their collaboration patterns, and the degree of such transformation indicates and is correlated to how successful the implementation of Lean has been for that particular team. After developing propositions based on literature and practical experience, three hypotheses are tested with a small sample of work teams at an industrial company implementing Lean. Results support that as companies become Lean, there is an increase in communication among workers of a team, among different teams and between teams and management. These changes are more pronounced in more successful work teams. The conclusions may be helpful to understand a lean transformation as a communication network transformation, and hence open the game to improving the odds of a successful lean implementation by facilitating communication rather than by insisting in technical or tool-based skills.

A COGNITIVE LOAD VIEW AND EMPIRICAL TEST OF COLLABORATION NETWORK STRUCTURE VERSUS LEARNING RATES IN NEW SOFTWARE DEVELOPMENT

This study explores whether characteristics of the collaboration structure in software development teams affect development learning rates, with a secondary goal of testing product complexity as a moderator. We develop suitable hypotheses under the theoretical lens of cognitive load theory. The empirical study uses archival data on an ordinary least squares model to find significant associations between collaboration structure, product complexity and the learning rates exhibited by 230 development teams producing open source software. Results show two distinct subgroups of projects: The first subgroup exhibits an average 78% learning rate, and the other subgroup “unlearned”, i.e., productivity deteriorated over time instead of improved. In the learning subgroup, collaboration network density negatively impacted learning, while product complexity interacted with collaboration network centralisation and boundary spanning activity. In the unlearning subgroup, only network density impacted learning rates and no moderating effects were found. Practical implications and future opportunities for research are discussed.

The interplay among software volatility, complexity and development outcomes: evidence from open source software

The study posits a mediating role of software complexity in the association between software volatility and different software development outcomes. Empirical tests using data from 326 open source software projects support such a partial mediating role of software complexity in the association between software volatility and development outcomes. Archival data is tested using an ordinary least squares mediated model. The paper uses productivity, defect count and development speed as dependent variables.

STRUCTURAL CHANGES ASSOCIATED WITH TEMPORAL DISPERSION IN SOFTWARE DEVELOPMENT TEAMS: EVIDENCE FROM OPEN SOURCE SOFTWARE PROJECT TEAMS

Collaboration structure and temporal dispersion (TD) in teams have been studied independently so far. This study uses Media Synchronicity Theory (MST) to derive hypotheses positing that the structure of collaboration networks in distributed teams changes when those teams are more temporally dispersed. The empirical test of hypotheses using ordinary least squares with archival data from 230 open source software (OSS) projects shows that the collaboration structure networks of those OSS teams that are more temporally dispersed are sparser and more centralised, and these associations are stronger in those teams exhibiting higher relative performance. Theoretical and practical consequences are discussed.