My earlier paper TRI/EFC-4 from The Relational Institute on this subject was published in ACM SIGMOD Record, December 1986. As a result, I have received several comments and proposed changes. The purpose of this paper is to respond to these items, while continuing to place heavy emphasis on the semantic aspects of missing information. The response is principally further explanation, but a few changes of a quite minor character are proposed. In this paper the reader is assumed to have read TRI/EFC-4.