effect-system

name	effect-system
description	Implement effect systems to track and control side effects in programs.
version	1.0.0
tags	["effects","types","purity","popl"]
difficulty	intermediate
languages	["haskell","purecript","koka"]
dependencies	["type-checker-generator","type-class-implementer"]

Effect System

Effect systems track what computational effects a function may perform, enabling reasoning about purity, exceptions, I/O, and other side effects at the type level.

When to Use This Skill

• Building pure functional languages
• Reasoning about side effects
• Implementing effect handlers
• Security and sandboxing
• Optimization (pure code can be optimized more)

What This Skill Does

1. Effect Annotation: Track effects in types
2. Effect Checking: Verify effect constraints
3. Effect Polymorphism: Generic effect handling
4. Effect Row Types: Combine multiple effects
5. Effect Masking: Hide effects from callers

Key Concepts

Concept	Description
Effect	Computational side effect (IO, State, Throw)
Effect Set	Collection of effects
Effect Row	Extensible row of effects
Effect Polymorphism	Generic over effect sets
Effect Handler	Interpret effects
Pure	No effects

Tips

• Start with coarse-grained effects (IO vs Pure)
• Use row polymorphism for flexibility
• Consider effect subtyping
• Effect handlers enable flexible interpretation
• Log effects for debugging

Common Use Cases

• Pure functional languages
• Effect tracking for optimization
• Sandboxing and security
• Effect-based testing
• Resource management

Related Skills

• algebraic-effects - Algebraic effect handlers
• type-class-implementer - Monad type classes
• monad-transformer - Monad transformers
• information-flow-analyzer - Security effects

Canonical References

Reference	Why It Matters
Lucassen & Gifford, "Polymorphic Effect Systems" (POPL 1988)	Original paper on effect systems
Talpin & Jouvelot, "The Type and Effect Discipline" (POPL 1992)	Effect polymorphism formalization
Leijen, "Koka: Programming with Row Polymorphic Effect Handlers" (2017)	Modern effect system design

Tradeoffs and Limitations

Approach Tradeoffs

Approach	Pros	Cons
Monads	Composable	Syntactic overhead
Effect systems	Cleaner syntax	Complex inference
Algebraic effects	Flexible	Runtime overhead

When NOT to Use This Skill

• When side effects are not a concern
• Performance-critical code
• When exceptions suffice

Limitations

• Effect inference can be complex
• Higher-order effects are tricky
• Interaction with existing code

Assessment Criteria

A high-quality implementation should have:

Criterion	What to Look For
Soundness	All effects tracked
Inference	Automatic effect inference
Polymorphism	Effect polymorphism
Handlers	Effect interpretation

Quality Indicators

✅ Good: Sound, infers effects, supports polymorphism ⚠️ Warning: Manual annotations required ❌ Bad: Misses effects, no polymorphism

Research Tools & Artifacts

Effect system implementations:

Tool	Language	What to Learn
Koka	Koka	Row-based effects
Frank	Frank	Handler calculus
Eff	Eff	Direct handlers
Multicore OCaml	OCaml	Effects in OCaml

Research Frontiers

1. Effect Inference

• Goal: Automatic effect tracking
• Papers: "Effect Inference" (various)

Implementation Pitfalls

Pitfall	Real Consequence	Solution
Effect inference	Complexity	Use bidirectional