The vast majority of mathematical puzzles ask for the existence of a solution. It is merely an exercise when the method is known and it is more of a puzzle when the method is not clear. An algorithmic puzzle takes this further by only asking for the method itself or a property of the method. It is in this sense that much of computer science is puzzle solving. We discuss the theory behind this in the context of material taken from Martin Gardner’s Scientific American column. The answer to the following puzzle will be given:
There are five pirates dividing up 100 gold coins. Pirates are strictly ordered by seniority, are very logical and wish to live. The rule pirates use to divide gold is: (1) the most senior pirate suggests a division, (2) all pirates vote on it, (3) if at least half vote for it then it is done, otherwise the senior pirate is killed and the process starts over. What happens?
Dana Richards is an associate professor of Computer Science at George Mason University. His research is on theoretical and algorithmic topics. He has been a friend of Martin Gardner for nearly four decades and has edited Gardner’s book
The Colossal Book of Short Puzzles and Problems.