This page offers an overview of my cumulative, publication-based doctoral thesis.
Abstract#
Background: Modern software systems are often too large and complex for an individual developer to fully oversee, making it difficult to understand the implications of changes. Therefore, most collaborative software projects rely on code review to foster asynchronous discussions about changes before they are merged. Although prior qualitative studies have shown that practitioners view code review as a communication network, no formal theory of code review as a communication network or empirical validation exists. Without confirmatory research, the theory’s validity remains uncertain, hindering its credibility, practical application, and further development.
Objective: In this thesis, our objective is to (1) formalize the theory of code review as a communication network, focusing on information diffusion—the spread of information—as a core characteristic of communication networks, (2) empirically validate the theory by quantifying information diffusion in code review across varied perspectives, contexts, and conditions, and (3) demonstrate its practical application in the context of tax compliance within collaborative software engineering.
Methods: To empirically validate the theory, we use three complementary research approaches. First, we quantify information diffusion using an observational study at Spotify, measuring how information spreads across social, organizational, and architectural boundaries in code review. Second, we developed and conducted in-silico experiments with closed-source code review systems from Microsoft, Spotify, and Trivago, as well as open-source code review systems from Android, Visual Studio Code, and React, to estimate the capability of code review to facilitate information diffusion. Third, we surveyed practitioners on their anticipations for the future of code review to assess how these changes might impact our understanding of code review as a communication network.
Results Throughout our comprehensive empirical validation, we found no evidence that would falsify the theory of code review as a communication network. We observed extensive information diffusion across social, organizational, and architectural boundaries at Spotify. Additionally, we discovered that code review effectively spreads information quickly and widely among participants, even at a large scale. However, we also found that information diffusion patterns in open-source code review systems differ significantly, suggesting that findings from open-source environments may not directly apply to closed-source contexts. Through applying the theory of code review as a communication network, we were able to uncover the significant and uncovered tax risks associated with collaborative software engineering within multinational enterprises. While practitioners consider code review also in the future a core practice in collaborative software engineering, we foresee that generative AI could undermine its role as a human communication network in the future.
Conclusion: Our work on understanding code review as a communication network contributes not only to theory-driven, empirical software engineering research but also lays the groundwork for practical applications, particularly in the context of tax compliance. Future research is needed to explore the evolving role of code review as a communication network.
Introduction1#
Please note that the introduction will be continuously revised and updated, with the date used for versioning. Some company names are temporarily blacked out due to confidentiality agreements. We are currently awaiting official approval to disclose these names. Full details will be made available as soon as authorization is granted.
Download PDFPublications#
This section details the thesis contributions, with each chapter—aside from the introduction—based on a manuscript submitted to a peer-reviewed venue.
| Chapter | Study | Link |
|---|---|---|
| Chapter 1 | Michael Dorner, Daniel Mendez, Ehsan Zabardast, Nicole Valdez, and Marcin Floryan. Measuring Information Diffusion in Code Review at Spotify. Empirical Software Engineering Journal. Accepted as registered report; submission forthcoming. | → |
| Chapter 2 | Michael Dorner, Darja Šmite, Daniel Mendez, Krzysztof Wnuk, and Jacek Czerwonka. 2022. Only Time Will Tell: Modelling Information Diffusion in Code Review with Time-Varying Hypergraphs. In ACM / IEEE International Symposium on Empirical Software Engineering and Measurement (ESEM), 2022. ACM. 10.1145/3544902.3546254 | → |
| Chapter 3 | Michael Dorner, Daniel Mendez, Krzysztof Wnuk, Ehsan Zabardast, and Jacek Czerwonka. The Upper Bound of Information Diffusion in Code Review. 2025. Empirical Software Engineering Journal 30, 1 (2025). 10.1007/s10664-024-10442-y | → |
| Chapter 4 | Michael Dorner and Daniel Mendez. The Capability of Code Review as a Communication Network. Transactions on Software Engineering and Methods. Under review. | → |
| Chapter 5 | Michael Dorner, Maximilian Capraro, Oliver Treidler, Tom-Eric Kunz, Darja Šmite, Ehsan Zabardast, Daniel Mendez, and Krzysztof Wnuk. Taxing Collaborative Software Engineering. 2024. IEEE Software 41, 4 (2024), 143–150. 10.1109/MS.2023.3346646 | → |
| Chapter 6 | Michael Dorner, Andreas Bauer, Darja Šmite, Ehsan Zabardast, Ricardo Britto, and Daniel Mendez. Quo Vadis, Code Review?. IEEE Software. Submission and preprint forthcoming. |
Defense#
When? September 23, 2025, 14:00
Where? J1630 on Campus Karlskrona
Committee#
Opponent: Felix Dobslaw, Mid Sweden University
Members of the grading committee:
- Brian Fitzgerald, University of Limerick and Lero, Ireland
- Emma Söderberg, Lund University
- Burak Turhan, University of Oulu, Finland
In Sweden and other Nordic countries, the comprehensive introduction, which integrates articles of a compilation thesis, is called a “kappa.” ↩︎
