Gil and Ty,
Somewhere, recently, there was a detailed discussion on the shape of the back edge of the surfaces.
Short and sweet of it, as I grasped the comments, was:
Rounded TE isn't as good as perfect sharp edge or sharp clean change to a vertical back edge (right angles to the chord line.)
Air 'attaches' to the surfaces it flows over, and a clean edge discourages air from flowing from higher pressure side to lower pressure side. ...Like a squared transom on a speedboat, the water finds its own way to rejoin, somewhere away from the back end.
The 'boundary layer' was mentioned in that discussion, and it gets thicker along the chord of the surface air flows over. The boundary layer is slower moving than the free air further off the surface - it is what could attach and mess things up. There are ways, apparently, to estimate the thickness of the boundary layer at the back edge of whatever surface.
So, a taper to about the thickness of the boundary layer at the end of the flap, or elevator, chord is at least as efficient as a true sharp edge. Estimated 'squared' back end height was around 3/32 to 1/8", if I recall, for a mid/large-size stunter wing chord.
It is also stronger, more warp resistant, and a lot less susceptible to hangar rash...