At a recent computer science conference, your friend at U(sic)GA suggested the following algorithm for finding the shortest path from node s to node t in a directed graph with some negative edges: "If we just add some large value N to the weight of each edge to make them all positive, then we can run Dijkstra's on the modified graph and find the shortest path from s tor". Is your friend correct? Either prove that it works or give a counter example