Hacker News Reader

If you view member access as a (global) function on a value (possibly an lvalue), then it actually makes a whole lot of sense.

I find code dealing with structures that have their members prefixed with a unique prefix much easier to read, usually.

Given the pervasiveness of APIs that contain struct definitions nowadays, it's also understandable that the unique-constraint was removed.

View on HN

Indeed and Haskell still has this issue today, because member access is a top-level function.

Still, it creates all kinds of problems for code generation (templates) and code by convention (as with generic).

I have not seen a case of it creating problems where there weren't already plenty of other problems with the code and metaprograming had to be abandoned anyway. But it does create problems.

It would make more sense if fields actually were functions, or at least function-like in syntax.

Then again, OCaml still has module-scoped scoped record field names today, even though it also uses the dot syntax.

Reply on news.ycombinator.com