2 ms·
”Even if we are discussing a pair of points on adjacent polygons?” Yes. Consider a pointy triangular wedge of two 1 by 100 rectangles and a third one that’s 1
by Someone 6y ago
”Even if we are discussing a pair of points on adjacent polygons?”
Yes. Consider a pointy triangular wedge of two 1 by 100 rectangles and a third one that’s 1 by 1, and pick points on the two large rectangles that are close to the edge they share with the small one. The shortest path will visit all three rectangles.
Also: I was only refuting the claim that such a line is the shortest line between the two points. I happened to know an (IMO) interesting counterexample, so I mentioned it. I think the conclusion that such a line is a geodesic is valid, but not because of the argument given.