I still think that you would find Norvig's book is written in a style that shows you
algorithms stripped to their essence, yet they are fully executable code.
You would soon learn how close Lisp is to the *ideas*.
The way he defines and then builds on tree-search could hardly be any
more elegant. His page 191 shows the arguments are: states, goal-p,
successors, and combiner. The latter three are functions passed as
arguments. Successors is simply a list.
It's nine lines of code plus two lines of comments,
plus the variety of search strategies he presents in the rest of Sec. 6.4,
all built on tree-search by using appropriate variants of the three functions.
- Bob Futrelle
On Jan 1, 2008, at 9:27 AM, Michael B. Enders wrote:
I will keep that in mind, but at this point my strategy for trying to
figure out the pseudocode is to get some of the algorithms that are in
pseudocode in the book in Java, because I some experience with Java and
none with Lisp or Python. For some reason, I find the authors'
pseudocode style to be almost impossible to understand - even when I
find out how the algorithms work from other sources. I figure that if
I compare the pseudocode versions of the algorithm with version in a
language that I am more familiar with, I will finally be able to
understand the pseudocode style. If I look at examples in Lisp then I
am dealing with two languages I don't understand instead of one.