A 3D Unit Sphere?

Dyami started up this thread on Tribe about two weeks ago: http://techpoi.tribe.net/thread/37590299-0963-4197-befb-3d7fbcf007ce

In it, he's taking Alien Jon's concept of the unit circle and pondering whether it can be applied to 3D geometry to create a similar family of moves that work within the path of a sphere.

I've been back and forth on this now for the past couple weeks because I can't for the life of me think of any new concepts this would add to the vocabulary of poi, but also not wanting to dismiss the concept completely because there is always the possibility I've become so focused on a given paradigm I can't see my way out of it. The issue with this concept I keep stopping at is that by definition I think poi spinning is two-dimensional.

What do I mean by this? Well, the head of a poi can only be in one direction from the hand anchoring it at any given time and the influence of gravity will always make it settle in one direction. The same issue exists with all spinning tools I'm aware of, from staff to meteor. The movement of a hand guiding a poi always represents a single two-dimensional slice of a three dimensional shape since it can only explore one extension of the shape at any given time. To describe this another way, imagine if your hand could support spinning three poi at once--one in each of the three planes. In this case, you could move your hand in three dimensions in ways that would drastically change the behavior of all three poi.

Put another way, think of poi spinning as being like watching a movie. While what we see on screen is a two-dimensional rendering of a three-dimensional scene, it doesn't seem in the least bit odd to us. The reason for this is that we really see in only two dimensions, but sort out a third by overlapping two two-dimensional images on top of each other and working out the differences between the two. Likewise, when viewing a poi spinner, no viewer can sort more than two dimensions of poi movement at a given time. Poi being spun parallel to the angle of the viewer is seen for all intents and purposes as a straight line. By the same token, any trick done in three dimensions that I can conceptualize will wind up resembling a two dimensional trick that has likely already been named when seen by an audience.

Thus any three-dimensional family of figures one can map out in a "unit sphere" actually wind up just being two-dimensional figures rendered differently for the audience. Or at least that's the conclusion I keep coming to. Anybody else have any success finding 3D moves that don't have 2D corollaries?