Никола Радојевић:
>The title pretty much sums it up
I was going to reply to your previous (personal) email with the link to
the archive of the announcement. You beat me on time :)
I have really no clear plans. As I said in the announcement, I would
first like to take some time to study the LSP (the protocol itself) and
take a look at some existing implementations (I was thinking of gopls
and rust-analyzer, as those are two languages I am proficent with).
The following step, ideally, would be to write some fist implementation
that supports the simplest use-cases: many useful existing
implementations in other languages do not yet support refactor, for
example, and that could be a feature we could leave out for a while.
I will try to document what I am learning in the wiki. I guess this
could be a collaborative effort in that one may add considerations to it
(to the wiki, I mean) via a PATCH to this mailing list, on the wiki
branch (at least, this is how I undestood the mechanism behind the wikis
on sourcehut).
Another interesing topic to cover is how to test an LSP implementation.
Again, I guess that to copy/take inspiration from others is the best way
forward.
Greetings
--
blallo