John Schoenfeld wrote:

[snip]

Um, yes, it is.

No it is not. Imagine a stationary black-box floating in space. One
wall of the box is hard iron and the opposite side is ellastic. If a
ball is thrown from the middle at the hard iron wall there will be a
high-impulse transfer of momentum from the ball to the box. Relative
from the center of the box (which at this point is moving), the ball
now approaches the opposite ellastic wall in which it inevitably
collides with and transfers the same momentum but in the opposite
direction bringing the box to rest again. However, the elastic wall
collision was low-impulse and took longer for the momentum to be
conservered. Irrespective of momentum conservation, there is an
overall displacement.

At this point we have the box at rest yet it is displaced from its
original position, however in future time this same effect will occur
but in the opposite direction and thus the overal motion of this
contraption would be to OSCILLATE about the original position. So
technically speaking, its not inertial propulsion yet as the center of
mass is constant.

So the third and final requirement would to have a constant stream of
balls colliding just as the first one thus always staying one step
ahead of the "backwards oscillation phase".

I think some past claims of reactionless drives that allegedly
reduced their weight (though never to zero, it seems) on scales, had
more to do with a similar phenomenon in the springs of the scale, then
actually providing a net upward force. Time your oscillations right, and
you can fool the scale, but not Mother Nature....