EVi, a hard-fork of Vim
21 points by krig
21 points by krig
a hard-fork of Vim v9.1.2073 before AI was used
I am afraid this fork is too late, AI slop has been integrated way before that, see e.g.: https://github.com/vim/vim/issues/18800#issuecomment-3568099543
I remember testing out the allegedly improved completion menu from the above issue and being surprised myself how bad the implementation is, with flickering and random crashes everywhere.
This bizarre LLM era is so much against anything that Bram stood for. From my times interacting with him, he always was so careful in review and specifically chose to move slowly to ensure everything is properly done. The famous case of reverting cmdheight=0 comes to mind, as it was just too buggy at the time.
I don't think Bram would have approved of these strange new times and he's probably rolling in his grave about dissolving his charity with a heartless AI generated bullet list; at the same time it feels toxic to blame the maintainers, when they are already at the verge of burnout.
it feels toxic to blame the maintainers, when they are already at the verge of burnout
There are no other people that could be blamed. I have to wonder why they are maintaining the project in the first place, if it isn't to keep the spirit of the project alive? Maybe rather than burning out or outsourcing maintenance to a machine, they could simply slow down and do less.
Tangential: I sympathize with the vim maintainers, but if it has reached to the point where it cannot be maintained naturally, isn’t it concerning that it comes pre-installed in some Linux distros? Unless it’s already version-locked, in which case, it’s fine I guess.
Aside: Have they considered feature-locking vim to give themselves some temporary reprieve?
I am happy to see this, but what I really want to know is what the fork wants to do with the codebase, as well as what they're against.
Right now, it seems like most of the work is around killing old platforms.