First screen, captured 9 hours ago. Open the video.

A technical vision for the Haskell ecosystem

What HN thinks

Haskell fans and skeptics reenact the annual 'is it useful yet?' debate, now with LLMs as the new excuse.

Strict typing should suit AI agents
Some argue Haskell's terseness and strong types make it ideal for LLM coding, while others note sparse training data, slow feedback loops and poor dev experience make agents struggle.
Haskell lacks polish and broad reach
Critics say Haskell's libraries for GUI, mobile, games and web feel uncoordinated, so it rarely gets picked for everyday apps; defenders reply it can already do all of that.
Tooling and install story fragmented
Commenters debate the Unix 'do one thing' philosophy versus Cargo-style unified tooling, and joke about choosing among ghcup, stack, cabal and Nix.

Consensus: Most agree the language itself is strong and the real bottleneck is developer experience: tooling, compile times, error messages and onboarding.

Recommended in the comments

OCaml (tool): Suggested as a more comprehensible strict language with fast feedback for AI-driven coding.
F# (tool): Used for teaching because it has Haskell's aesthetics with practical mutability escape hatches.
Lean (tool): An AI researcher works in it as a strict language well suited to agents despite verbosity issues.
Nixpkgs ↗ (tool): Offers easy overrides and Haskell helpers like callCabal2nix for managing dependencies.
Yesod (library): Cited as a production-used Haskell web framework.
Servant (library): Cited as a production-used Haskell web framework.

Summarised by Claude from the comment thread 6 hours ago.

Points: 175
Comments: 79
Best rank: #2
Posted: 2026-11-13 by cosmic_quanta
Category: Programming

haskell haskell-foundation ghc developer-experience open-source ai-assisted-coding

All captures:

Full page

Captured 9 hours ago. Click for the whole screenshot.

Looks similar

Screenshot of github.com: Llama.cpp: 6.5x faster prompt processing on Arm with a NEON kernel

Llama.cpp: 6.5x faster prompt processing on Arm with a NEON kernel (github.com)

6 points by hardwiredagent 14 hours ago | 1 comments

GitHub repo with an Arm NEON SDOT kernel patch for llama.cpp's tiled K-quant matmul, plus benchmarks showing up to 6.5x faster prompt processing on Neoverse cores.

llama-cpp arm-neon quantization simd

Screenshot of github.com: Mocktastic: A Meshtastic Node on Arduino

Mocktastic: A Meshtastic Node on Arduino (github.com)

25 points by kimi 9 days ago | 1 comments

GitHub repository for Mocktastic, an educational C++ Meshtastic-compatible LoRa client for ESP32 boards (LilyGO T3) built with Arduino IDE, with decoder and DHT22 sensor modes.

meshtastic lora esp32 arduino

Screenshot of github.com: ZapFast: A fast, native WhatsApp client in Rust and egui

ZapFast: A fast, native WhatsApp client in Rust and egui (github.com)

87 points by handfuloflight 5 days ago | 59 comments

GitHub repository for ZapFast, an MIT-licensed native desktop WhatsApp client written in Rust with the egui GUI library for Linux, macOS and Windows.

rust egui whatsapp desktop-app

Screenshot of github.com: A ZIP-style Prolog VM for trilog

A ZIP-style Prolog VM for trilog (github.com)

22 points by triska 4 days ago | 6 comments

GitHub repository for zipc, a work-in-progress compiler from Prolog to Clocksin's ZIP abstract machine plus a C VM, intended as a backend for trilog.

prolog virtual-machine compilers c

Screenshot of github.com: MoofTCP – A modern replacement for MacTCP on 68k Macs

MoofTCP – A modern replacement for MacTCP on 68k Macs (github.com)

56 points by marcusbasant 8 days ago | 28 comments

GitHub repository for MoofTCP, an MIT-licensed, memory-lean drop-in replacement for MacTCP on 68000 Macs running System 6, written mostly in C.

mactcp classic-mac 68k system-6

Screenshot of github.com: Show HN: Maple – Sub-2-BPW Moe Pruning to Run Gemma 4 26B at 45 TPS on a 6GB GPU

Show HN: Maple – Sub-2-BPW Moe Pruning to Run Gemma 4 26B at 45 TPS on a 6GB GPU (github.com)

7 points by AmSach 5 days ago | 3 comments

GitHub repository for MAPLE, a Mixture-of-Experts pruning framework that compresses Gemma 4 26B-A4B to a 4.75 GB GGUF to run fully in VRAM on 6 GB GPUs.

moe-pruning gemma llm-quantization gguf