munching squares

munching squares n.  A {display hack} dating back to the
   PDP-1 (ca. 1962, reportedly discovered by Jackson Wright), which
   employs a trivial computation (repeatedly plotting the graph Y = X
   XOR T for successive values of T -- see {HAKMEM} items
   146--148) to produce an impressive display of moving and growing
   squares that devour the screen.  The initial value of T is treated
   as a parameter, which, when well-chosen, can produce amazing
   effects.  Some of these, later (re)discovered on the LISP machine,
   have been christened `munching triangles' (try AND for XOR and
   toggling points instead of plotting them), `munching w's', and
   `munching mazes'.  More generally, suppose a graphics program
   produces an impressive and ever-changing display of some basic
   form, foo, on a display terminal, and does it using a relatively
   simple program; then the program (or the resulting display) is
   likely to be referred to as `munching foos'.  [This is a good
   example of the use of the word {foo} as a {metasyntactic
   variable}.]