Since I started writing ES6, I’ve adopted the functions without function approach and have quite liked it.

I’m not going to get into every reason why, but that post does a good job of echoing my thoughts on it. I’m not going so far as to say “never”, but I don’t think I’ve found the need to type the word function even once since I started writing ES6.

I have always wondered about stack traces when using arrow functions though.

What about stack traces? (Hint: they’re fine)

This is one major question that I’ve seen whenever this discussion has come up, and frankly I didn’t know the answer to it. Did using arrow functions equate to use anonymous functions which can make stack traces (and debugging) impossible to navigate?

The answer that I found is: no.

Here is the code I used and some screenshots of running it in Chrome, Node 5, and Chrome after transpiling it with Babel.

const x = (arr) => arr.map(Boolean)
const y = (arr) => x(arr)
const z = (arr) => y(arr)

console.log(z([1, 1, 0]))
console.log(z(5))

Chrome

Node 5

Chrome with Babel

Those all look pretty good to me! I’m going to keep writing arrow functions for most everything since I haven’t seen any significant downsides, and I really, really don’t like typing the word function :)