Recent searches
Search options
CentOS Stream wants it both ways. It wants to brand itself as a community distro like Fedora just with more QA and stricter stability guarantees. When your distribution is entirely developed by people working at Red Hat and the people with the most weight in decision making are RHEL customers and not the actual users/community of CentOS Stream, it's not a community distro.
@maxgot CentOS Stream is RHEL built in the open in collaboration with the community. Decisions are still made by Red Hat, so it's not community led, but it's more community oriented than classic CentOS ever was.
@carlwgeorge What percentage of CentOS Stream contributions (excluding Fedora commits) are from the community (i.e. non-Hatters)?
@carlwgeorge@fosstodon.org@maxgot This would be incredibly difficult to measure. The bulk of the dataset would come from GitLab merge requests. You would also have to go find bugzillas with patches attached, which was the workflow for CS8 before it was onboarded to the CS9 GitLab workflow.
@maxgot You'll have to identify the authors, not just as Hatters or not, but RHEL maintainers or not. I would argue that a MR from @Conan_Kudo working in a different department counts as community because working on RHEL is not his job. Or my submissions working in CPE to upstream debranding stuff for classic CentOS 8, or to resolve issues in support of EPEL. It gets even harder because I don't believe the maintainers are required to use their at-redhat-dot-com email addresses.
@maxgot @Conan_Kudo Even the GitLab metrics will be misleading, because there are several types of repos. A single dist-git commit probably encompasses more work than single src-git commit. Commits to releng things like comps would also be difficult to compare directly.
@carlwgeorge @maxgot I'm not sure anything of mine has landed since I joined Red Hat, though? Regardless, I'm not doing anything in CentOS as part of my job, it's all community work as I've been doing for years now.