Scheme interpreter in Erlang

I'm going to make a Scheme interpreter this weekend. I'm wondering how I should represent lists: Same as Erlang lists, or one 2-tuple per cons?

>>10
No.

Nr. >>4 probably never got his head wrapped around the functional paradigm. FP is used in many real world applications, and it's smart to use it for problem domains that fit it well. How can someone who doesn't understand something at all see how it would be useful to him?