For reasons that I’ll explain below, this is not an informed post. Rather, I’m just passing along something that I found on the Web and that was merely one man’s opinion. Still, it seems like useful information so I’m repeating it here but caveat emptor is definitely called for.
Over at the Emacs subreddit, kmlkclkmlkcl says that after two years he’s changed from LSP-mode to Eglot and he’s really happy with the change. He says, “It just works”, feels like it’s part of Emacs and not a third party app, and that unlike LSP-mode it was easy to configure and get working.
As I say, this is uninformed commentary on my part. Probably because I’ve spent most of my programming career programming in C and various Lisp dialects, I’ve never felt the need for LSP mediated programming. I know you can use it with C but why bother? There aren’t objects with a ton of methods to negotiate. At most you have to deal with some libraries but to an experienced C programmer, these become second nature.
Added to that, I really hate having my editor pop up miscellaneous information—it’s why I don’t use the automatic completion so beloved by others—so the last thing I want is my editor telling me what function or method to call. I prefer to be left alone when I program; it’s probably why the idea of pair programming fills me with loathing and dread.
The point of all this is that I’m the last person to ask about the relative merits of LSP applications. But I do think kmlkclkmlkcl has some useful information so I’m passing it on.