PF>Bad or NULL pointers jump right out at you under OS/2
  >(especially under 16 bit os/2).  And rather than creeping
  >up on you long after the offending code corrupted memory,
  >you usually find out right away under OS/2 (your program
  >traps).  When you compile with debug info, and run under a
  >debugger,  it will take you right to the offending line of
  >code -- where your bad/null pointer will be staring you
  >right in the face.

Not always.  I once had this real bad C++ class who's code would
over-write the return pointer (the point where the program would
continue after returning from the function).  It took me several days,
with a debugger to track that puppy down, because it did not trap at the
exact line, it wouldn't trap at all.  It would return to the original
program in the WRONG place!

Michael Douglass
___
 .Mike's Mail Internet:   MICHAEL.DOUGLASS{at}LCHANCE.SAT.TX.US

--- Maximus/2 2.01wb