I mean, I've ridden bicycles; their wheels weigh, at a guess, less than a pound (racing bikes, half that?), so any gyroscopic forces would have to be pretty small compared to your (or especially, *my*! :-) body mass... And bicycles still keep their balance pretty well.
Therefore, it seems to me as if this gyroscopic effect, while apparently beneficial, is not strictly necessary for a "balancing" vehicle.
And yes, I *was* actually thinking handlebar steering.
Thing is, I've heard or read somewhere -- yeah, typical, but isn't that the way it is with a lot of what we claim to "know"? (can't even recall if it was in a bicycle or motorbike context) -- that handlebar-steering kind of *is* steering-by-leaning: If you turn the bars, and *don't* compensate the imbalance by leaning to the inside of the turn, you'll fall to the outside. Now, the claim went, we actually use this effect all the time -- the very *way* we start that inside-leaning, is by (sub-consciously) turning the handlebars the *wrong* way. This starts us "falling" toward the outside of this wrong-turn, which we then utilize as leaning toward the *in-*side of the *correct* turn -- we just follow up by turning the handlebars in the now-appropriate direction, the one we originally *intended* to turn to.
Weird, isn't it? I tried to confirm it a couple years ago, by paying close attention to how I actually rode my bicycle, and by trying to do it this way *consciously*. The results were inconclusive -- it's *damn* hard to consciously monitor or guide your ingrained reflexes! -- but I certainly couldn't *dis-*prove it, anyway.
So, what does this mean, in practical terms, for this particular project? I dunno; could be that I've just ass-u-me'd handlebar steering out of inertia... But maybe it means, at least, that it won't actually actively *hurt* to do it that way.
Apart from the little keel-fins I'm also assuming under each ski (like the ones on regular water-skis and surf-boards -- that is, in a couple of the propulsion scenarios, the rear one would consist of the one on the propeller housing), I've been thinking that by just shaping the skis appropriately, we'd automatically get a correlation between our lean angle, and the angle between the lines they each cut in the surface of the water:
They gotta be rather strongly tapered (or, in extermis, triangular) -- the front one with the thinner end (or the point) forward; the rear one, rearward. Just thinking about it, at first it seemed to me that nothing would stop such an edge from cutting down into the water... But then, that doesn't happen with water skis, does it? (Why not?) Anyway, *if* there is a danger of that, then we can always add vertical up-standing edges to the skis (turning them into something like "trays"), which would then automagically become the second leg of a 'V-shape' when the ski is leaned.
It'd be "a planing c-oracle", eh? Well colour me green and call me Bede, but maybe I can get it sponsored from work... :-)