In this paper we address issues related to building a large-scale Chinese corpus. We try to answer four questions: (i) how to speed up annotation, (ii) how to maintain high annotation quality, (iii) for what purposes is the corpus applicable, and finally (iv) what future work we anticipate.