Greg's suggestions for reviewing

1. read the lecture and section notes -- these are the
  key sources of information.  If there's something that's
  even remotely fuzzy, ask Greg about it.

2. create exercises for yourself or better yet, with a study
  partner.

  * 42 problems -- particularly involving state, streams,
    and higher-order functions.  One example we went over
    yesterday is this one:

    (if (= (greg) 42) 0 (greg))

   Define greg so that this evaluates to 42.  Another one we
   did was:

    ((greg) (greg))

  * box-and-pointer diagrams -- if I give you a diagram of
    some (mutable) lists as laid out in memory, can you write
    some Scheme code that will realize those lists?  If I give
    you Scheme code, can you show the box-and-pointer
    diagram of the data structures it will generate?

 The goal of all of this is to make sure you understand how
 Scheme works -- that you have the correct model of evaluation.

3. You can ignore parsing.  What you should take away from this is
  (a) using higher-order functions to factor out code, and (b) using
  higher-order functions to generate code.

4. You should understand the concepts of OO (e.g., classes, objects,
  inheritance, subtyping) but need not worry about how to realize
  these things in PLT-Scheme.  Typical questions:  What is the difference
  between a class and an object?  Why is inheritance useful?  etc.

5. review problem sets and projects.