A lot of times they are the same people. The Haskell community really isn't as active as it was five or even ten years ago, and many of those Haskell evangelists became Rust evangelists. Some people are just attracted to using the most "advanced" language (for whatever definition of "advanced" that's in vogue) and some of them are extremely vocal about that.
The Haskell community is admittedly less active in promoting niche concepts like zygohistic prepomorphisms, but much more active in building out a solid ecosystem. You won't hear much about the latter because it isn't shiny and weird.
How has the ecosystem become more solid in recent years? All of the important Haskell libraries I use were initially released eight years ago or before then.
While I personally don't like people promoting niche concepts as well (I'd rather have them promote building out a solid ecosystem as you say), I still feel the "brain drain" out of the Haskell community.
* We now have ghcup, a solid way to install Haskell (inspired by rustup, I believe). It used to be that you got access to one GHC package from your Linux distribution if you were lucky, otherwise you had to install manually from a bindist.
* We now have HLS, which integrates superbly with VS Code (it's also fine in Emacs and I guess in other editors, but I don't know). It can be flaky on large codebases, but it's a big step up from what we had before (which was almost nothing).
* GHC and bundled libraries have a much higher degree of API stability. In fact in the last 18 months there has been minimal breakage due to new GHC releases[1]. Community libraries are a bit more of a mixed bag, but there's much more awareness in the community that continual API breakage churn is harmful for adoption.
* We now have the Haskell Foundation[2] supporting all sorts of critical ecosystem activity behind the scenes, and helping people work together harmoniously. There used to be extremely fraught and unpleasant interpersonal battles in the community. Those are a thing of the past.
* We now have solid effect systems (Bluefin[3] and effectful) that combine the best of the transformers world (build effects from components, type safety, handle effects and remove them from the type) and the IO-based world (predictable performance, resource safety, escape hatches). I'm willing to boldly claim that effects are no longer a problem in Haskell.
I'm in the same boat as you: I don't use particularly man new libraries. Is that a problem? I'm given to understand that in the Java world that's called "Monday". Sure, I would have liked it if certain community members had stayed more active, but I don't feel I'm missing anything in particular. (That may just be to do with what I work on.) What are you missing?
