i think they're considered two sides of the same coin - exhaustive vs extensible, each with its benefits and tradeoffs. this is known as "The Expression Problem

i think they're considered two sides of the same coin - exhaustive vs extensible, each with its benefits and tradeoffs. this is known as "The Expression Problem"; what follows is a quick and messy intro. sum types: you can easily add new functions that accept your sum type, but adding a new "variant" to the type will require modifying every function you've written to support the new case polymorphism (interfaces): you can easily add new types that implement your interface, but if you want to add a new method/function to it, you're going to have to implement it for every type that implements your interface essentially, we've got two "extensibility dimensions": one for adding new operations, and one for adding new types. sum types are good on the operation axis, interfaces are good on the type axis. (it's an open research question if there's a solution that's good on both.) i hope this makes sense! as a side note, when discussing this, it's useful to distinguish ad-hoc polymorphism (subclassing, interfaces/typeclasses) and parametric polymorphism (generics) – here, we're talking about the "ad-hoc" variety. I mention this because i saw some confusion about this in a sibling thread but can't reply there.