This website serves as status page for my cumulative, publication-based doctoral thesis on code review as communication network.
Abstract#
Background: Modern software systems are often too large, too complex, and evolve too fast for an individual developer to oversee all parts of the software and, thus, to understand all implications of a change. To maintain and enhance a collective understanding of the codebase and the changes to it, most modern collaborative software projects rely on code review to foster informal and asynchronous discussions on changes and their impacts before they are merged into the code bases. Those discussions make code review a communication network where its participants exchange information and, when needed and deemed relevant, the information is passed on in subsequent code reviews. Although prior exploratory studies lay an important foundation in understanding code review as communication network, the confirmatory counterpart for this theory is missing.
Objective In this thesis, we set out to test the theory of code review as communication network by quantifying the extent of information diffusion, the spread of information, in code review from different perspectives that may or may not contradict the underlying theory of code review as communication network or its universality.
Methods: We quantify information diffusion using two complementary research approaches: On the one hand, we conducted an observational study to measure the information diffusion in code review across social, organizational, and architectural boundaries. On the other hand, we developed and conducted in-silico experiments to estimate the upper bound of information diffusion in closed-source and open-source code review.
Results Neither the measurements of information diffusion across social, organizational, and architectural boundaries nor the upper bound of information diffusion in open-source and closed-source code review could falsify the theory of code review as communication network. In fact, we found evidence that code review is capable of spreading information among its participants at scale.
Conclusion: Our work advances the understanding of code review as a communication network. By that, we contributed not only to a theory-driven and empirical software engineering research, we also lay the foundation for a practical implications: Code review as a communication network serves as a measurable proxy for cross-border collaboration, which is taxable. We conclude by exploring how advances in machine learning may shape the future of code review as a communication network.
Introduction1#
Schedule#
Status page available
2025-03-25
Create a concise status page to track dissertation progressDissertation available
July 2025
Provide the final version of my dissertation as a downloadable PDF.Defense
August 2025
Defend my dissertation at Blekinge Institute of Technology
In Sweden and other Nordic countries, this comprehensive introduction, which integrates articles of a compilation thesis, is often called a “kappa.” ↩︎
