Type systems define how we reason about code, enforce safety, and manage complexity — but not all type systems approach these goals the same way. In this talk, I pit structural typing, as in TypeScript, against nominal typing, as in Kotlin, in a practical exploration of real-world scenarios. You’ll see side-by-side examples of how each system handles interoperability, composition, and code evolution, revealing where explicit identities enforce discipline and where shape-based typing unlocks flexibility and expressiveness. By the end, you’ll understand the strengths and trade-offs of each approach — and why, in many situations, structural typing gives you a surprising advantage.

Devoxx

A conference talk comparing structural and nominal type systems, using TypeScript, Go, OCaml, Kotlin, Java, and C# as examples. The structural type system is illustrated through a declarative query language built with TypeScript generics, showing dynamic type construction and type-level computation. The nominal type system is demonstrated via a compile-time guaranteed finite state machine in Kotlin, enforcing valid state transitions without runtime logic. The talk concludes that neither system is universally better — the choice depends on project context, with structural systems excelling at flexible API integration and type-level computation, while nominal systems shine for enforcing identity, semantic safety, and controlled state machines.

Type Clash: When Structural and Nominal Systems Collide by Matteo Orlandi and Luca Caputo