It appears that Singmaster, Thistlethwaite, and just about all the cube
hackers I know at and around Stanford consider anything you can do
with one wrist motion to be a single twist. Since this gives a more
accurate measure of how complicated a sequence is, I'm happy with it.
Why do you folks at MIT insist that you're right and the world is wrong?
(I admit it complicates the notation. My own cube notation uses two chars
per twist, one being the face and the other being the direction: left-arrow
for counterclockwise, right-arrow for clockwise, down-arrow for 180 -- isn't
extended ASCII wonderful?)