Estimating the Project Lifetime: A Methodological Overview and a Practical Case
summary
Claims a concrete method: GitHub repo lifetimes measured from first to last commit for top 1,000 projects, giving a median 4.5 years survival and a heuristic that a 3‑year old project likely lasts 3 more years. Presents a simple cost model where a bug occurring in 1% of sessions with 5 minutes per occurrence costs about 117 hours over a 4‑year lifetime. Uses public GitHub data and issue timestamps to justify fixing bugs immediately or marking them never to be fixed.
AI-written summary · about
chapters (14)
- 0:00Introduction to project lifetime
- 1:05Software lifecycle stages
- 3:53Maintenance vs development focus
- 5:19Challenges of decommissioning
- 7:29Project lifetime expectations
- 9:12GitHub repository lifespan data
- 12:04Language impact on project longevity
- 14:34Explicit project timelines
- 16:23Release and support proportions
- 17:50Maintenance and feedback loops
- 19:36Lifetime value of defects
- 23:52Bug resolution strategies
- 27:03Handling unresolved defects
- 28:54Summary and key takeaways
related talks
44mThe Math behind Project Scheduling, Bug Tracking, and Triage
46mLet Them Eat Code
20mHow long is a piece of string? Solving the conundrum of software estimation
51mZero-bug policy success
32mTime Driven Development
41mA Human-Centric Quality Assurance Process for Open Source Softwa...
21mHarnessing tenure variety within a team
30mSoftware Estimation Made Easy