Bill P., are you there?

Last week I sent a private note to Bill Powers asking for
          some details about the ARM1 simulation. As I have not yet
          received a reply, I'm beginning to think that perhaps my
          message did not get through, so I will repeat part of it
          here. Within the subroutines that compute the output of a
          given control mechanism (e.g., azimuth1), you employ a
          temporary variable, which is set to its previous value
          plus the computed change. This variable is not initialized
          within the routine itself. When is this done and to what