On 22/05/25 18:10, Matthias Geiger wrote:
On Thu, 22 May 2025 17:41, Joerg Jaspert <joerg@debian.org> wrote:Thanks for the vote of confidence; I concur that this might be a nice goal in the long run.On 17602 March 1977, Matthias Geiger wrote:While it is not that mature yet and some features are still being worked on, I think it already suitable to host small, non-key packages.I would like a switch, but i think its currently not realistic.
A few weeks ago I brought up a few points during a discussion on #d-mentors about forgejo (and other alternatives to GitLab) and I believe these points are also relevant in this thread:
a) Debian must plan a post-Gitlab future because GitLab (the org) will stop one day providing FLOSS/DFSG-compliant sources.
b) That day is not tomorrow; there is no need to push too much now. c) forgejo is the only viable forge alternative.d) forgejo's CI (and external tools like woodpecker) are way behind what Salsa CI requires and are not integrated UI-wise into the forge.
e) glab is a very good CLI; forgejo has no equivalent CLI.f) It makes no sense to plan how to move away from GitLab before forgejo(+CI) reaches parity.
g) Whatever replaces GitLab must support 100% of what is needed to run tag2upload (including all security-relevant details and assurances).
Regards, -- Gioele Barabucci