After some rounds of emails, the trip wound up having three simple parameters:
I also included Washington, D.C. and added two stops in California to get us to an even 50 stops.
Next, Randy did the hard stuff. He details the challenges of optimizing such a route in this blog. To get around some of those challenges, he used information freely available from Google Maps API and wrote a bit of code to calculate the distance and time it would take to drive between all 50 landmarks, which represents 2,500 individual routes.
8 Far-Flung Walking Sojourns
It would take a computer millions of years to account for every twist and turn along a continental drive and give us the absolute best route - one that backtracked as little as possible. But hey, we just don't have that kind of time. So Randy pulled the genetic algorithm card again, which he used for the Where's Waldo solution.
He writes: "Instead of exhaustively looking at every possible solution, genetic algorithms start with a handful of random solutions and continually tinker with these solutions - always trying something slightly different from the current solution and keeping the best one - until they can't find a better solution any more."