Remove dead URL #44984
Remove dead URL #44984
Conversation
The `rubysl/rubysl-yaml` repo and its issue tracker are gone, and there doesn't seem to be an alternative URL.
|
This is 8 years old: #16267. It is unclear to me if Rubinius is dead or not, but I think we can leave it for now. Like JRuby, we don't support the platform officially, but we accept patches related to alternate rubies. |
Yeah, I think we probably should. While JRuby is actively working on compatibility (and is just not [yet?] at the point where it's automatically our fault/problem if something breaks), I think Rubinius has been on a different path for a good while now, the lack of recent activity being only the latest incarnation of that. We can revisit/reintroduce things if or when that changes, but for now, IMO it's at best pointless to keep such special cases, and more likely slightly harmful, by implying there's a hope of something working when (at least to my knowledge) that's just not realistic. |
|
🤔 Maybe @brixen can share some details on state of Rubinius. |
|
Thanks for reaching out @simi. I think removing Rubinius-specific code is fine. Rubinius has over 400 contributors and many people have commit bits. The core idea behind Rubinius to build advanced technology supporting Ruby remains. If people find it worthwhile to work on it, then I imagine they will. My personal time is very limited, but there are many areas of very recent research that could be tremendously beneficial that I may begin incorporating in Rubinius. If you're curious, check out Koka and FBIP (functional but in-place update of data structures), and semantic subtyping. |
Rubinius has not been maintained since May 2020 and based on the discussion at rails#44984 , I think we can remove Rubinius specific code from Rails.
The
rubysl/rubysl-yamlrepo and its issue tracker are gone, and there doesn't seem to be an alternative URL.From a cursory search, there are a couple dozen places where we have Rubinius-specific code. Rubinius hasn't had a commit since May 2020 (rubinius/rubinius@b7a755c), and we do not officially support it. Is there interest in removing this code?