Software Development - Mob Programming, Pair Programming or Fly Solo?
summary
Argues ensemble work raises long‑term throughput not raw keystrokes, using concrete cases: a week‑long mob on Couchbase to solve transactional concurrency that prevented data corruption, a Greenfield sprint where the whole team builds a CI/CD “walking skeleton” and chooses Jenkins vs local scripts, and a legacy UI week that produces a small automation DSL for QA. Shows pairing collapses PR delays into real‑time review so teams shift toward trunk‑based delivery and faster feedback loops, and warns that metrics like Jira tickets, LOC or hours obscure real outcomes.
AI-written summary · about
chapters (13)
- 0:00Collaboration in software dev
- 3:28Programming productivity myths
- 10:13Why productivity metrics fail
- 16:09Outcomes over outputs
- 21:46Social nature of programming
- 28:03Pair & ensemble basics
- 37:02When to use whole-team
- 47:05Shared understanding & ownership
- 57:12Fun side of collaboration
- 1:01:04Choosing collaboration style
- 1:10:13Remote & in-person tips
- 1:15:30Adapting to team flow
- 1:20:29Closing & Q&A
related talks
31mMob Programming: Team-Building Included
01h 04mAdvanced Software Teaming
01h 00mMob Programming, A Whole Team Approach
29mAccelerated learning with Mob Programming
42mAdvanced Software Teaming [Mob Programming]
01h 05mSoftware Teaming (Mob Programming) and the Power of Flow.
50mMob Programming, A Whole Team Approach
01h 35mSoftware Teaming [Mob Programming] mini-workshop