Maybe we need more innovation in reading code. Language servers and autoformatters already are doing a lot. Visualizers could make it way easier too.
There's a general issue though that we haven't figured out the silver bullet for abstraction. Traits/typeclasses are convenient but they're inherently global. Interfaces are similarly global and are quite annoying when you have to take in two objects of the same type. Omitting traits and using something like modules can work but you lose quality of life stuff like easy printing (quick, how do you say something is printable in OCaml?). I'm not familiar with Erlang/Elixir but I suppose message passing could be an option, although I'm not sure how it'd work for memory intensive situations.
You could go the PL nihilism route and be against abstraction as a whole, but I don't think that's a valid answer. Abstraction is necessary for code to function at scale.
There's a general issue though that we haven't figured out the silver bullet for abstraction. Traits/typeclasses are convenient but they're inherently global. Interfaces are similarly global and are quite annoying when you have to take in two objects of the same type. Omitting traits and using something like modules can work but you lose quality of life stuff like easy printing (quick, how do you say something is printable in OCaml?). I'm not familiar with Erlang/Elixir but I suppose message passing could be an option, although I'm not sure how it'd work for memory intensive situations.
You could go the PL nihilism route and be against abstraction as a whole, but I don't think that's a valid answer. Abstraction is necessary for code to function at scale.