It would appear that his aim is to design a frictionless bearing using magnets to levitate a generator rotor. My guess is that he thinks that a frictionless bearing system equates to ... uh, the (ever elusive) perpetual motion machine. What seems to escape his sparsely furnished mind is that the "electrical load" on the generator contributes to the vast majority of the drag, which can also be translated into friction. Good bearings contribute very little to rotor friction.
Chris
yeah. when connected up, an alternator, as if by magic, takes a fair bit of force to turn, while a disconnected alternator can be turned relatively easily (you gently turn it, and is spins freely for a little while...).
then if course, there is all the magic goes into making induction motors work...
had seen a video where someone had gone and drilled some holes and put neodymium magnets in a shaded pole motor, and used it as a generator. had a random idea that maybe this could be done without the magnets, provided the right electronics (you would feed a little current into the system while the motor is being spun, such as by wind, and possibly you could get slightly more power back from the induced back-emf).
a possible application would be making a wall of box fans (mounted on fence posts), with a small computer fan used as a wind-detector (wind will induce a voltage in the computer fan windings, which may trigger the control electronics to start pulsing the big motors).
granted, conversion efficiency would likely be pretty low vs the use of magnetic rotors.