Question: How are short timeframes of intense activity, followed by a corresponding return to a typical pattern of activity, observed in a project?
There are a number of reasons that may prompt a sudden increase or decrease in the amount of activity within a repository. These increases and decreases appear both as a sudden change in activity against the average amount of activity. Burstiness is a way of understanding the cycle of activity in existing metrics, like issues, merge requests, mailing lists, commits, or comments. Examples of root causes for bursts in activity include:
- Release cycles
- Global pandemics
- Hackathon activities
- Mentorship programs
- Conferences, meetups, and other events where tools are presented
- Conventional and social media announcements and mentions
- Critical bugs as raising awareness and getting people’s attention
- Community design meetings or brainstorming meetings to address a particular issue
- Community members show up from another community that is relying on your project (e.g., dependencies)
- To identify impacts of root causes of a burst in activity
- To provide awareness when project activity unknowingly goes up
- To help capture the meaningfulness of increases or decreases in project activity
- To help the community and maintainers prepare for future bursts that follow a pattern
- To help measure the impact of influential external activities
- To differentiate skewed activity versus normal activity
The usage and dissemination of health metrics may lead to privacy violations. Organizations may be exposed to risks. These risks may flow from compliance with the GDPR in the EU, with state law in the US, or with other law. There may also be contractual risks flowing from terms of service for data providers such as GitHub and GitLab. The usage of metrics must be examined for risk and potential data ethics problems. Please see CHAOSS Data Ethics document for additional guidance.
- Issues or bug reports
- Release Tags
- Change Requests
- Mail List Traffic
- Documentation additions or revisions
- New Repositories
- Feature Requests
- Messaging Conversations
- Conventional and Social Media Activity
- Conference Attendance and Submissions
Tools Providing the Metric
- Grimoire Lab
Data Collection Strategies
- Time box activities identifying deviations away from some norm
- Outliers for certain thresholds, using statistics like Bollinger Bands to measure stability or volatility: https://en.wikipedia.org/wiki/Bollinger_Bands
- Qualitative Interview Questions
- Why do you contribute more during a period of time?
- What do you believe to be the root cause for particular bursts?
- What impact do different events (e.g., hackathons, mentorship program, or conferences) have on project activity?
This metric was inspired by the work of Goh and Barabasi (2008): https://arxiv.org/pdf/physics/0610233.pdf