I think if Zig wanted to, it could introduce lightweight linear types using the concept of proof variables and interleaving from ATS. Since resource management

I think if Zig wanted to, it could introduce lightweight linear types using the concept of proof variables and interleaving from ATS. Since resource management is explicit in Zig anyways, there's not much additional overhead in "consuming" proof values to signal that you've dealt with a resource.