Haskell Ecosystem Survey 2025: What Developers Want Changed

Source: Haskell Blog | Date: April 2, 2026

        Key Finding: 553 developers shared their pain points. The top issues: Text instead of String, onboarding, documentation, tooling, and language complexity.
    

Major Topics

1. Text Instead of String

A laser-focused suggestion: replace String with Text as the canonical type for textual data.

"The text library must live from base, because none of the APIs of base can use Text, and when people look at base for examples of well-written Haskell code, they end up using String, which is not the right choice."

2. Onboarding of Beginners

Need for up-to-date content
Centralized entry point in the language
Haskell Wiki decreasing in contributions

3. Documentation

Library authors encouraged to provide more working examples and documentation beyond simple API references.

4. Complexity of the Haskell Language

Language perceived as too complex
Historical cruft and lack of consistency
Requests to stop adding language extensions or redesign them holistically
Desire for new Haskell Language Report

5. The Record System

Requests for OverloadedRecordDot and NoFieldSelectors to be enabled by default.

Tooling Issues

Compilation Times

"Make GHC 10x faster" is one of the prevalent moods amongst participants.

Ecosystem Consolidation

Newcomers confused by fragmentation. Referenced uv as example of fully-integrated toolchain.

Compilation error messages seen as confusing (for end-users vs implementers)
Haskell Language Server memory usage and reliability issues

Pervasive Laziness

"Anxiety about laziness is a recurring theme, indicating lasting difficulties developers have in reasoning about lazy evaluation."

Requests for data structures strict by default, using StrictData extension.

Minor Topics

Dependent Haskell

Viewed as unified solution to replace loosely related language extensions.

Effect Systems

Interest in base providing algebraic effect systems out of the box. Requests for more MonadIO usage.

Debugging

Haskell Debugger mentioned as addressing pain points. "HasCallStack everywhere" mentioned as symptom.

Academia vs Industry

"There needs to be a recognition that there is documentation produced for/by the type theory / heavy mathematics enthusiasts which doesn't translate well as developer documentation."

Why This Matters

This survey provides rare insight into a major functional language ecosystem's pain points. The themes - tooling, complexity, documentation - are universal across programming languages. The tension between academic and industry needs is particularly relevant in 2026's AI-dominated landscape.