It's close, but I think NeoVim's RPC model could be huge. Writing a new frontend for vim no longer seems that crazy of a goal which makes it easier for people to experiment with new editor ideas while relying on the very solid core of vim (well at least I find that exciting, idk if the general vimmer cares).
Or rather, a day where its easier, more flexible, and more performant for IDEs to always(only) use an embedded neovim. Getting plugin support "for free".
The keybindings would then be the only thing that the IDE changes.
I dream of the day when the IDE is in vim. There has been some great work around c# for this lately, where the compiler is running as a service and vim makes an async call for file completion, etc.
The neovim project is still a massive undertaking about refactoring such old and non trivial piece of code in what appears to be a smooth way. I hope they can keep working on it, in symbiosis with vim.
