What Richard Gill has stated is wrong. I have revised my simulation, which removes the apparent ambiguity in the directions of angular momentum:
## Note that, despite appearances, u and v specify just one set of directions
## for the angular momentum (and likewise the vectors -u and -v). It is very
## easy to check that there is one-to-one map between the set of directions u
## and the set of directions v. In other words, for every u_i there is a v_j
## such that u_i = v_j, and for every v_i there is a u_j such that v_i = u_j.
The simulation speaks for itself. Further details can be found here.
