Given data in two diasporic languages with a shared ancestor, how can we find words that are linked together?
Step 1: Methodology
Step 2: How can we do this well, portably and fast?