Analyze a project and its management
A fully client-side GitHub/BitBucket repo analyzer. It finds out: Is a project well managed? Does it accept help? How does it compare to other options? And then: what should you do to take your repo from in the weeds to super effective?
It performs the hard work entirely in the browser so that each user's own API rate limits are in effect. This avoids problems similar projects have had, and means faster and more varied results.
- A developer looking for a library wants to compare management styles of a few alternatives.
- A developer thinking about contributing to a project wants to know if it's worth it.
- A job candidate wants to know how a prospective employer manages their projects.
- A company wants to see if all its projects are being managed similarly.
One metric is implemented (effectiveness) and I'm brainstorming about next steps:
- More metrics
- Advice for improving the issues found
- New visualizations, and better UX
Quantitative measurements which can determine qualitative assessments.
- Num. of active branches
- merged
- un-merged
- Num. of stale branches
- merged
- un-merged
- Num. of open pull requests
- proportion open:closed pull requests
- aging of pull requests
- Num. of open issues
- proportion open:closed issues
- Num. of stars
- Num. of forks
- Num. of subscribers
- ratio of # of tags to # of tags in use
By using the raw signals listed above, we can calculate useful metrics. These metrics, then, will be the pieces that can form a judgement for the big questions.
E.g.:
- How well do the maintainers keep up with the queue of incoming PR's?
- How quick do the maintainers respond to new issues and PR's?
- How does the activity level affect the other metrics?
- Code Climate score
- Travis CI passing