Hacker News Reader

Comment by zepolud | Hacker News Reader

Defining a sane syntax while the language is still in its infancy is anything but bikeshedding. You have only one chance to do it right before an entrenched codebase develops or interest in the language fizzles.

MS Research did this right with the experimental hardwhite support while F# was still in alpha. People liked it so much that it became the default style for the compiler and the standard library. It would be impossible to try to make the switch now that the language is widely used.

It might be childish for me to choose one language for a project than another over trivial things like semicolons, but for the language itself it's an irreversible decision once the spec gets set in stone.

View on HN

To me, one of the most interesting design decisions of Rust is that the creators strongly try to avoid inventing new language features. Instead, they attempt to find the best combination of tried-and-true existing features from a variety of other languages.

I wonder whether language syntax could be considered such a "feature" -- that is, the Rust creators would consciously prefer to combine existing syntaxes (at this point, mainly C++, Haskell, OCaml and Ruby) than invent their own.

An argument could be made that they could just "use Python's syntax", but perhaps a counter-argument would be that Python's syntax would need to be extended so much (for types, pointers, templates, etc.) that the result would effectively be a new invention.

That logic sounds fishy to me. After all, they took C++'s semicolon syntax over Haskell, OCaml and Ruby's semicolon-free syntax, so there must have been a reason for picking one over the others. So the issue is not about inventing syntax, it's about choosing one (existing) syntax over another.

pcwalton says it was to attract C++ devs, which is interesting considering that Go was also designed to attract C++ devs and made the opposite choice. As a sometime C++ dev, I recognize the historical uselessness of semicolons and would love a replacement language that did away with them.

Since when OCaml is semicolon-free? It not only has semicolons, single semicolon (;) means something different than double semicolon (;;). The former is used as an expression separator, like comma in Erlang, and the latter denotes the end of a top-level declaration, like period in Erlang. That's a pain while refactoring, but I can live with it. Anyway, that's just a style, nothing more, you can learn to love it or you can write a "transpiler", it's really a non-issue.

True, Ocaml has semicolons, although it's a little different given that you can construct deeply nested functional expressions without ever using a single one. If I remember correctly the semicolon is a separator, not a terminator, which means that declarations don't need to end with a semicolon, for example, and you pretty much only need it to separate statements.

AFAICT, if you were to write in Lisp-like purely-functional style, you wouldn't need any semicolons in your Rust code either.

I'm not sure I prefer 'in' over semicolon in declarations. I mean:

    let x = ref 0 in
    (* something *)

in Ocaml is not much worse/better than

    let mut x = 0;
    // something

in Rust. As I said above, that's really a minor, stylistic issue, one probably would write "let x = mut 0 in" for a week or so (coming from OCaml) and that's it, after a week it becomes as natural to write semicolon as it is to write 'in'.

I agree, I didn't really put OCaml forward as the ideal solution. Ruby and Go solve the semicolon problem in much better ways, in my opinion.

Reply on news.ycombinator.com