noooit

It's all probably against the policy but if there is a free port open, you can start ssh. If not, I agree there is no way to use local emacs. I would just try to compile emacs on the remote machine which is difficult restrict if the direct or indirect internet access is allowed.

I have a similar requirement.
If folder1 and folder2 have different project root marker, it's possible to start multiple language servers via project-vc-extra-root-markers.

But the tricky part is when it's started from folder2, it needs to cd to folder1 before executing the language server.

I achieve this by creating a wrapper script for language server and create a function that sets an environment variable before executing the script and does do complex things in the script.

It'd be nice if there was a better solution without depending on .dir-locals.el and etc, like passing a function for the server program so that I don't have to use a script + environment variable.

The default one. It's beautiful with white-space mode on.

GNU Emacs is configurable for a reason. Don't use those complete garbages.
The configuration should be made by you. It's really not that hard, just start from the tutorial.

GNU Emacs is configurable for a reason. Don't use those complete garbages.
The configuration should be made by you. It's really not that hard, just start from the tutorial.

The tramp maintainer wasn't really keen on innovating anymore so somebody needs to volunteer.

I feel like network filesystem + some remote commands/plugins handler would stabilise things more. Even with sshfs mode, remote command doesn't work.

It really doesn't make sense that it supports so many network filesystems and none of them support remote command/process. If I could mount network filesystem locally, I wouldn't use tramp to begin with.

Try expand region package from elpa.