Two activities come to mind. One was when Superballs first hit the market, and we searched for more and more inventive ways to test their capacity. Up to the roof! We stood by the edge, overlooking a parking area, and dropped the ball. It bounced about 60% of the way back. Not good enough.

We leaned farther over the edge and threw them down, but still they didn't bounce all the way back. Perhaps it was wind and air resistance (you scientists in the audience back me on this), or perhaps it was our fear of really throwing with all our might. Even as children, we understood the basic law of inertia -- your body in motion throwing the ball would continue on its path over the top of the parapet, if your forward force overcame the force of gravity holding your sneakers to the tar rooftop. This body would then stay in motion until the force of the pavement far below put a stop to the acceleration of gravity on that body in motion. I believe Archimedes died during such an acceleration experiment with a Superball, thus proving this theory. Kids, don't try this at home.

The other activity involved playing catch, one of us on the roof, the other down below. For added challenge, when I was on the roof I loved to throw the ball straight up so that it had all that extra height before its descent, sending my friends scurrying after it between the cars and bushes and tiny knee high chain fences they posted along the paths that seemed so well designed to break an ankle during such an otherwise healthy activity.

Today if I heard my own kids were up on a roof doing that stuff I would be sending complaints to every safety board in the city, and forbidding them from leaving my sight. In my dotage I have a firm grasp of physics, and a firmer grasp of the concept of "margin for error."