pull down to refresh
Maybe I’m reading too much into it, but it seems to show the commit count trending down since May 2025. It makes me think that the drama and general increase in harassment of Core contributors has been at least distracting Bitcoin Core contributors if not outright making them reconsider their career choices.
I'll take this 'concern' as worth exploring.
methodology: compare pre/post commit activity (count and consistency) to see if there's any correlation with the drama. I'll use May 2025 as one cutoff date to start, and we can repeat this for other key moments or drill down into specific developers/maintainers
I was surprised how you identify me, and the labels of my contributions are confusing:
E.g., I opened the following PR in 2024 and it got merged in 2025, but neither of those two bars lists any Tests.
I’m also not sure how you compiled that list of Maintainers. Some of the dates seem incorrect to me (e.g. Pieter stepped back in 2022), Luke and Cory never were Bitcoin Core maintainers, and some other people are missing. Perhaps check out this: https://bitcoin.stackexchange.com/q/176/5406
Issue: 2024 Tests Not Showing
Root cause: My categorisation logic had a priority flaw. I defined:
Wallet category rule: src/wallet/
Tests/QA category rule: src/test/, test/, or src/bench/
Conflict: Files in src/wallet/test/ matched both patterns. The wallet rule took precedence, so test files were incorrectly categorized as wallet code rather than tests/QA.
Fix: Adjusting logic to catch component-level test directories (e.g., src/[component]/test/) before falling back to component categorisation.
On Maintainer Tracking
Fixed the 'relay race' chart using:
Static file source: Your Stack Exchange maintainer list
Merge commits to identify maintainer activity
Verification: Cross-checking against the trusted-keys file to confirm maintainer identities
Thanks for pointing out the discrepancy. I'd appreciate it if you could flag any other data points that don't look right and need fixing
this part is fixed.
I'll refine the impact profile to make it more explanatory, in case it's confusing
Thanks for jumping in. Curious to know what do you see in the chart.
Also,
1: You’ve been in the trenches, so love to get your thoughts on: what numbers or views would actually help you day-to-day?
2: Any buckets or labels feel off? Happy to re-cut the data so it maps to how contributors really think about the code
If this turns out useful, will love to make it some sort of engineering analytics or developer health dashboard.