Hi, could anyone help me with understanding why the lambda papers deserve a whole book? What I understand from the history is that they are a set of papers that define and explain the Scheme programming language.
If so, shouldn't the book be broader by including more of literature on LISP or even AI programming in general?