3 ms·
The other thing that bugs me about a count like this as a metric, is it doesn't speak at all to the complexity of the keywords. For example, think of how comple
by JackMorgan 13y ago
The other thing that bugs me about a count like this as a metric, is it doesn't speak at all to the complexity of the keywords. For example, think of how complex the extends keyword is in Java/C#. Just that one keyword represents usually one of the hardest things to learn about Java. Or the defmacro special form in Clojure. Macros are a huge area of confusion when learning a language. The count metric makes very little sense in explaining the difficultly to learn a language.
- vorg 13y agoAgreed, counting keywords in Java or macros in Clojure core isn't comparable. For example, keywords such as `:when`, `:while`, and `:let` inside a Clojure `for` macro support the macro name, perhaps in the same way `extends` supports the `class` or `interface` keywords in Java. It would even be possible to have only one publicly visible macro in Clojure with keywords in 2nd position to chose the behavior, e.g. `(macro :defn myfn [a] (print a))`.