The List is our Process: An Analysis of the Kernel's Email-based Development Process
Talkat Open Source Summit Europe + Embedded Linux Conference Europe 2019 · on Monday, 2019-10-28· at 16:20· Forum 1·
summary
Paper-quality measurement of how patches travel from mailing-list post to merged commit by reconstructing histories from mailing-list + kernel.org trace analysis and applying network clustering and distance metrics. The work quantifies pipeline steps, measures time-to-merge and categorizes off-list integrations and ignored patches, finding 24% ignored patches (~1,100) in the sample and numerous off-list fixes that never appear in thread history. Dataset and tooling reproduceably flag where review breaks down and which repositories absorb out-of-band changes.
AI-written summary · about
chapters (10)
- 0:00Kernel email-based process
- 1:37Linux kernel collaboration
- 4:18Mailing lists as core
- 9:04Patch popularity and clusters
- 11:05Kernel.org repositories limits
- 15:06Patch trends and ignored
- 19:25Patch and list statistics
- 23:14Open patch inclusion
- 26:48Mailing list community
- 32:21Patch integration challenges
related talks
22mKeynote: A Fireside Chat with Greg Kroah-Hartman, Linux Foundation Fellow
45mFrom an Idea to a Patch in the Linux Mainline
37mGetting Your Patches in Mainline Linux: What Not To Do (& a Few Things You Could Try
56mMaintainer_s Diary - We Have a Scaling Problem
39mThe Life of a Kernel Bugfix
21mSyzbot To Mainline: How I Merged 21 Kernel Patches as a First-Time Contributor
53mUpstreaming 201
58mUpstreaming 101