Mining knowledge on technical debt propagation

Technical debt has gained considerable traction both in the industry and the academia due to its unique ability to distinguish asset management characteristics for problematic software project trade-offs. Management of technical debt relies on separate solutions identifying instances of technical debt, tracking the instances, and delivering information regarding the debt to relevant decision making processes. While there are several of these solutions available, due to the multiformity of software development, they are applicable only in predefined contexts that are often independent from one another. As technical debt management must consider all these aspects in unison, our work pursues connecting the software contexts via unlimited capturing and explanation of technical debt propagation intraand inter-software-contexts. We mine software repositories (MSR) for data regarding the amount of work as a function of time. Concurrently, we gather information on events that are clearly external to the programmers’ own work on these repositories. These data are then combined in an effort to statistically measure the impact of these events in the amount of work. With this data, as future work, we can apply taxonomies, code analysis, and other analyses to pinpoint these effects into different technical debt propagation channels. Abstraction of the channel patterns into rules is pursued so that development tools may automatically maintain technical debt information with them (the authors have introduced the DebtFlag tool for this). Hence, successfully implementing this study would allow further understanding and describing technical debt propagation at both the high level (longitudinal technical debt propagation effects for the project) and the low level (artifact level effects describing the mechanism of technical debt value accumulation).