Every October, Write a Lisp Interpreter In C

I would have thought that writing a C compiler in Lisp would be more interesting... Every time I read a book on compiler design in C, the book passes an inordinate amount of time setting up the infrastructure for creating an AST, yet surely this would be a very natural thing in Lisp?

Do you think they would want us to spend our 'Octobers' reinventing things they built or for us to take their works in computing to greater heights?