Re: New Package for NonGNU-ELPA: clojure-ts-mode

[João Távora]

On Sun, Aug 27, 2023 at 2:17 PM Philip Kaludercic <philipk@posteo.net> wrote:

> > I'm a bit perplexed why you picked me as the star of "what if
> > he were to disappear?" but I guess I'm as good a candidate as
> > Michael, Lars, Dmitry or so many others.
>
> Mainly because I am under the impression that of all the contributors to
> Eglot, you have the best understanding of LSP as a protocol.  But you
> are right, it was mostly the first example that came to mind when
> comparing it to CEDET.


LSP is not hard at all, it's just a RPC protocol with a growing but
still fairly digestible spec.  You could "master" it if you wanted to,
perhaps in one or two afternoons, just by reading through it closely.
And in fact I invite you (and others) to do that.

I'm not that much of an expert in it anymore, and I mainly just consult it
for reference when someone requests a particular feature.

My "hard to replace" expertise in Eglot is more in niche areas like
that mammoth eglot-completion-at-point function and the techniques
to keep completion responsive and accurate.  But even others like JD
Smith have developed a decent idea of what's going on there (they also
have ideas of how it should work alternatively, which I don't agree
with at all, but that's another matter).

Anyway, a considerable contribution to Eglot would be just to
to do that close reading of the LSP standard and then survey exactly
what parts of it Eglot uses and what parts it doesn't, then add that
information to the manual or some internal document.  Also, as you read
through the LSP standard I'm sure you would collect lots of ideas,
maybe even some low-hanging fruit, or things to contribute to Eglot
as code.

J