central river for automatic stabilization?

> I had been thinking that the accumulation of extra mass (the water)
> would pull the system even further out of whack, but the key is that as
> the CoM moves in ONE direction the water
> flows in the OTHER direction. The extra mass to one side pulls the CoM
> in it's own direction, in the process pulling the CoM through the
> geometric center. In a severe case, it might pull the CoM through the
> geometric center and a little beyond, but if that happens then water
> will flow the other way and pull it right back again. And since the
> water doesn't teleport, it flows at a finite speed, the whole effect is
> to dampen out the offset.

Yes, that's how it seems to me too.

Of course all this sloshing around consumes energy, which would come out
of the spin, and have to be made up somehow. This is another reason why
a good design should probably have two counter-rotating sections, so you
can spin them up with simple electric motors. But even if it's a single
section, and you have to use rockets, you could use some high-efficiency
thrusters (e.g. ion engines), and it's probably not a big deal.

> It takes some brain-wrapping, but I see it now. This is really quite
> brilliant and I'm amazed at how simple it is. This is almost certainly
> how it will be done, and all it'll take to keep people safe is an
> advisory to stay away from the river bank (or beach) for a while.

In reality I would expect these currents to be rather subtle. Only if
you suddenly dropped a very large mass on one side of the habitat would
there be any significant waves (and in that case, somebody should get
fired!).

Best,
- Joe