Crowds and Content: Crowd-Sourcing Primitives for Digital Libraries