5 min read
|
Saved February 14, 2026
|
Copied!
Do you care about this?
This article explores the bus factor concept, which measures the risk of knowledge loss in teams when key members leave. It details a project analyzing open source repositories to assess their bus factors using a specific algorithm, revealing surprising trends in code coverage and author contributions.
If you do, here's more
The concept of the bus factor measures the risk associated with knowledge and capabilities not being shared among team members, highlighting how a team's survival hinges on key individuals. In structural engineering, the author struggled with estimating this risk due to poor documentation and high turnover, leading to issues when critical questions arose long after a team member had left. In software engineering, the bus factor can be quantified by analyzing code contributions. A 2016 research paper provided a method for calculating the bus factor using a degree of authorship (DOA) metric, which determines how many developers contribute to the files in a project.
The author conducted an analysis using this method, focusing on open-source projects. After successfully building the algorithm, they ran tests on various repositories, discovering significant variations in bus factors. The analysis indicated that some projects saw an increase in bus factor, contrary to expectations that code cleanup would reduce it. Notably, the bus factor for the Linux source repository dropped dramatically from 57 to 8. About 30% of projects showed no change, while some even increased slightly, highlighting the ongoing challenges in maintaining open-source projects.
The analysis prompted further questions about the role of code review in knowledge sharing and how developer aliases might impact bus factor calculations. The author noted the limitations of relying solely on authorship data without considering review processes. They also pointed out the difficulty in replicating the initial study's results due to missing metadata in the supporting data. This exploration serves as a groundwork for more detailed investigations into open-source project health and knowledge sharing.
Questions about this article
No questions yet.