That makes some sense to me. The closer the connection between the rods and the center of the crank, the less stroke you would see. That could be set on a hydraulic system that would move them (connection points) inwards and outwards. The tough part of this would be that it would be damn near impossible to wire those hydraulics up because the crank is constantly spinning.

I say we all try to think of a theoretical solution to this instead of just saying "impossible".... It's a very cool topic. I'm a frosh in mechanical engineering, so its fun to talk about this stuff.

Think about this here... Instead of using hydraulics, have the connection between rods and crank be on tracks with some extremely stong springs (tracks that run from near center of crank towards the standard position) As the crank speeds up, the connection point would naturally move outward (causing larger stroke) This isn't what we want however, but it would cause a change in stroke, which is what we are looking for. If there is anyway to have a counter balance that pulls the connection pointer inward as the cranks rpm speeds up, we would be in great shape.

I'm still a little tipsy, so I'm just letting ideas flow. Let me know what you think.