I've asked this question few time and honestly it's no where close to "the one question" author makes out to be. Lot of candidates drifts on to something very c

I've asked this question few time and honestly it's no where close to "the one question" author makes out to be. Lot of candidates drifts on to something very complex, or describe other people's work as theirs, or make up stuff as they go. Plus it totally fails on campus hiring, one-man projects that evolves slowly over time and has legal risk of candidate disclosing sensitive information. The best interview question, IMO, is still the one that Sergey Brin used to ask: "Describe to me something that I already don't know".