Brooks’s Law and Development Output Trends¶
Brooks’s Law states that adding more people to a late software project makes it later. The reason for this is communication and coordination overhead. While we get more total hours available with more people, those available hours increase lineary but the coordination paths increase exponentially. Hence, there’s a point beyond which each additional person’s hours get comsumed by the increased coordination efforts…and then some.
Hence, use CodeScene’s development output to measure the effects of any chaning in staff as shown in Fig. 117.
The development output graph shows the following date:
- Development Output: This is measured by taking all commits during a week and divide them by the number of contributing authors. The resulting normalized output metric gives you an estimate on the organizations’s output.
- Authors (month): This trend shows how many unique authors that have contributed code over a month. This trend is likely to reflect the total number of developers on the project.
If your project is at risk to fall victim to Brooks’s Law, then you will see this as an increased distance between the total number of authors and their normalized output.
Like always when trying to measure things like developer and organizational productivity, the absolute numbers aren’t that interesting; the interesting thing is the trend. Do you get an increase or decrease in response to a chaning in staffing?
Follow-up with in-depth Analyses¶
Should you identify a decrease in development output, then we recommend getting more information via the following analyses:
- Coordination Needs: Check if the teams and/or individual developers need to coordinate their work in the hotspots using the Parallel Development and Code Fragmentation analysis.
- Decrease in Code Health: The decrease in development output could also be due to an increased level of technical debt, so inspect the Code Biomarkers–A Virtual Code Reviewer aware of Code Health analysis.