This thread’s scope largely overlaps with this one 
, but doesn’t necessarily point to actionable output, and doesn’t purpose to many examples.
Examples shared there:
- Julia’s native
AbstractArraytype hierarchy is used pretty much everywhere without much fuss. (Some do experience otherwise though). - Python’s built-in complex type is only for floats, c.f. Julia’s ability to define complex types over any other
Realisbitssubtypes, and for function methods to be efficiently defined for parametric subtypes effectively (I added that last point).
And some caveats on preaching multiple dispatch:
- Different packages working together through APIs is not a novel concept. EDIT: This point is my misinterpretation of @Benny’s explanation in the link, which he kindly clarifies below. The composability is arguably easier to accomplish in Julia c.f. other languages.
- Composability, especially in comparison between languages, is difficult to measure.