How do we get from 1000s projects to one product
summary
Explains how RHEL turns thousands of Fedora upstream packages into one release by doing periodic mask import of selected Fedora sources into internal repos, running weekly snapshots to build ISOs, and iterating alpha/beta cycles over multiple years; calls out a concrete surprise: PowerPC and mainframe compiling lag required months of internal work to get those architectures booting; stresses rigid documentation for team turnover, short six‑month planning windows, regular retrospectives, and the challenge of a 10‑year ABI lifecycle.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 1:15Upstream to distributions
- 2:23RHEL 6 release journey
- 6:04RHEL 6 lessons
- 7:29Documentation role
- 9:17Project cycles
- 11:44Cone of uncertainty
- 12:49Teamwork in development
- 14:15Work-life balance
- 15:59Large project challenges
- 17:30Documentation structure
- 19:23Upstream and downstream dev
- 20:46Fedora and PESC collaboration
- 21:40RHEL lifecycle and ABI
- 22:41Closing and Q&A
related talks
18mProject, Process, Product(ion)
37mFinding the Path from Embedded to Edge using Product Lines
41mInnovation takes courage
44mReflections around SW obsolescence
14mThe state of Morello software and projects
41mWhy is there no new Release? Nobody pays for the basics 🙁
57mLessons on Open Source from a 25-Year-Old Company
17mA Brief History of Release Engineering