You might get a laugh out of the fact that Hilbert curves are used to approximate Euclidean TSP. If I remember correctly it yields a log n approximation where n

You might get a laugh out of the fact that Hilbert curves are used to approximate Euclidean TSP. If I remember correctly it yields a log n approximation where n is the number of cities.