envy VS post-rfc

when you already know how to compile and run single-module interactive console programs, it takes about a day to understand basics of Cabal, and about a week to learn about input parsing and output formatting. Do you need CLI args? Use optparse-applicative. Env vars? Use envy. JSON? Use aeson. Don't think about performance and/or API conventions, that's not what you should be concerned of at this point, as you are just learning to compose things together from indivdual parts.

State of the Haskell Ecosystem lists libraries for many of its domains.

Recently saw this repo that shows the state of Haskell in a lot of programming applications. It classifies it as Immature for Numerical Programming; I would like to know what are the challenges in it.

Good resource: https://github.com/Gabriel439/post-rfc/blob/main/sotu.md

I know haskell has https://github.com/Gabriel439/post-rfc/blob/main/sotu.md, clojure has https://clojure.org/news/2021/04/06/state-of-clojure-2021, and scala has https://www.jetbrains.com/lp/devecosystem-2021/scala/. I only find the haskell survey complete enough to use in evaluating haskell for production use on a new project.

Seriously though, Haskell is a very practical programming language that especially shines for some very useful real world application domains.

https://github.com/Gabriel439/post-rfc/blob/main/sotu.md will probably help here

At work, we tend to build so-called "serverless" APIs, running on AWS. Haskell's story for maintainable software, provided you have the libraries, is excellent. The State of the Haskell Ecosystem document is a good starting point. Once you have a toe-hold running systems in production, the cost of building "just that one missing library" is often not too bad.