An improvement for LISPs

So I'm starting to learn LISPs quite a lot lately, and I'm fascinated by how everything is a list! It's brilliant! Absolutely brilliant. But I had second thoughts when I found out that LISP has _primitives like integers and strings. This disturbed me greatly, so I was forced to come up with a superior LISP model that only uses lists:

1. An integer i is a list li of empty lists such that (length li) == i.
2. A character is an integer i such that i < 256.
3. A string is a list of characters.
4. A float is a list of integers of length two.

What do you guys think of my idea?

0/10

I do not support this idea. It would greatly reduce the speed of my fact and fib programs, which I use very often.

'((()()()()()()())(()()()()()()()()()()))

(apply '((x) (+ x 1)) '(2))

>>1
How about handling a string as a list of arbitrary objects which might or might not be characters. Then you get ropes for free, as well as being able to handle arbitrary formatting, embedded images and other objects, etc. with standard data structures.

>>5 You mean (apply ''((x) (+ x 1)) '(2))

Alonzo Church, is that you?

>>5 , >>7
A procedure ought not become an unprinted object in >>1 's LISP.

characters in their fucking   language If it   was there and   wiped the file   you just saved   r n4 r?