This paper discusses what is needed to make MIS into a coherent research field. It defines 3 main needs:

1. Clarification of reference disciplines.

2. Definition of the dependent variable.

3. Building a cumulative tradition.

It reviews the relationship of MIS to computer technology and to practice and assess the publishing inlets for MIS research. The author would like to acknowledge with warm gratitude the contribution of Phillip Smith in helping shape the ideas expressed in this paper.