Reconsidering Implicit Return Types

[snnsnn]

Hi, seeing the issue #22, it’s clear that you value the experiences of seasoned language designers, particularly when reflecting on what they would do differently if they could start over. In this context, I’d like to mention a notable perspective from Nico Matsakis, a key figure in the Rust community.

I remember Nico expressing his thoughts about changing how implicit returns work—or maybe it was about making semicolons not significant. While I don’t recall the exact details of his critique, his reflection on the issue stayed with me. I searched for this video and recall it being a one-on-one interview—possibly one from the async interview series. It might be worthwhile to explore these interviews further or reach out to Nico directly to ask about the specific challenges or drawbacks he encountered.

I wanted to raise a point about implicit return types in Moonbit, in the spirit of benefiting from the collective experiences of designing programming languages. The design decisions of other languages can often provide valuable insights and help avoid potential pitfalls.

With the thoughtful approach evident in Moonbit’s design decisions and by drawing inspiration from Rust’s journey, I am confident Moonbit is poised to become a promising and influential language.

Best regards