3 ms·

I was really commenting on the fact that GO is absolutely not unique in most regards. For instance my main beef with it is that in a world where Ruby exists th

I was really commenting on the fact that GO is absolutely not unique in most regards. For instance my main beef with it is that in a world where Ruby exists there's no excuse for that kind of crappy all over the place syntax. The brackets seem to almost be there just because anything without C-like syntax gets ignored. [edit] But +1 from me to you anyways.

mahmud 17y ago
For instance my main beef with it is that in a world where Ruby exists there's no excuse for that kind of crappy all over the place syntax. In a world where Lisp exists, there's no excuse for syntax.
- dkersten 17y ago
  Ah, come on! The rules governing how s-expressions are made up are still SYNTAX! But I do agree with your point ;-)
  - smanek 17y ago
    Arguably, 'syntax' is the set of rules to transform code into an Abstract Syntax Tree. Lisp's code directly represnts the AST so you could say there is no syntax (to be a little more precise I might say Lisp is homoiconic, though).
    - dkersten 17y ago
      I'd still argue the textual format is not the AST itself, but a textual representation (just like any code) which is the transformed into the actual AST at parse-time using the syntax rules. Therefore, s-expressions have syntax, even if its minimal. (x y (z a)) is code, its parsed according to the rules of the syntax and is converted to an in-memory AST representation.
      - blasdel 17y ago
        You're barking up the wrong tree, pure s-exprs are a raw expression of the AST. A much better argument would be to point to all the damn quoting operators in most Lisps, which are most definitely syntax (sugar, even).
        dkersten 17y ago
        Oh, good point, thanks! And built-in reader macros.. But really, even pure s-exprs are just one possible textual representation f an AST. Hell, XML is a valid representation too. Anything thats a representation of something follows rules as to how its represented and if it has rules, then its got syntax. But.. you're right, arguing about the s-exprs part isn't going to get me anywhere. The Lisp weenies* will always believe Lisp has no syntax (blatantly ignoring the quoting operator which you mentioned) and nobody else really cares, so... * I'm not trying to knock Lisp; I actually quite like Scheme and Clojure (and probably would like Common Lisp too, if I'd spent more time with it). I just think this whole "it has no syntax, yet I have to follow these precise syntactical rules" stuff is stupid. ;-)

mahmud 17y ago

For instance my main beef with it is that in a world where Ruby exists there's no excuse for that kind of crappy all over the place syntax. In a world where Lisp exists, there's no excuse for syntax.

dkersten 17y ago
Ah, come on! The rules governing how s-expressions are made up are still SYNTAX! But I do agree with your point ;-)
- smanek 17y ago
  Arguably, 'syntax' is the set of rules to transform code into an Abstract Syntax Tree. Lisp's code directly represnts the AST so you could say there is no syntax (to be a little more precise I might say Lisp is homoiconic, though).
  - dkersten 17y ago
    I'd still argue the textual format is not the AST itself, but a textual representation (just like any code) which is the transformed into the actual AST at parse-time using the syntax rules. Therefore, s-expressions have syntax, even if its minimal. (x y (z a)) is code, its parsed according to the rules of the syntax and is converted to an in-memory AST representation.
    - blasdel 17y ago
      You're barking up the wrong tree, pure s-exprs are a raw expression of the AST. A much better argument would be to point to all the damn quoting operators in most Lisps, which are most definitely syntax (sugar, even).
      - dkersten 17y ago
        Oh, good point, thanks! And built-in reader macros.. But really, even pure s-exprs are just one possible textual representation f an AST. Hell, XML is a valid representation too. Anything thats a representation of something follows rules as to how its represented and if it has rules, then its got syntax. But.. you're right, arguing about the s-exprs part isn't going to get me anywhere. The Lisp weenies* will always believe Lisp has no syntax (blatantly ignoring the quoting operator which you mentioned) and nobody else really cares, so... * I'm not trying to knock Lisp; I actually quite like Scheme and Clojure (and probably would like Common Lisp too, if I'd spent more time with it). I just think this whole "it has no syntax, yet I have to follow these precise syntactical rules" stuff is stupid. ;-)