Searching for Technical Debt - An Empirical, Exploratory, and Descriptive Case Study

Abstract

Commonly, Technical Debt (TD) is used as metaphor to describe “technical compromises that are expedient in the short term, but that create a technical context that increases complexity and cost in the long term” [1]. Since TD is a metaphor, there does not exist a uniform understanding of what concretely such “technical compromises” are. Practitioners, researchers, and tools all subsume and consider widely different concepts as TD. In this paper, we set out to empirically and exploratorily, identify potential “technical compromises” that increase cost and complexity of modifications of two open-source database systems (Apache Cassandra and GCHQ Gaffer). In a manual investigation of 217 commits that are associated to 40 of the most costly and complex issues, we find that refactorings in the sense of Ur-TD [2] are often related to high complexity of modifications and that high cost is due to organization and coordination of work. Other than that, we cannot identify any “technical compromises” that can explain high cost and complexity of the studied contributions.