Educational changes may be required in order to realize the benefits that computer assisted learning (CAL) can offer geography instruction. CAL could make the learning and teaching of geography truly efficient and practically eliminate failures of instruction. An institutional constraint that discourages instructional innovations is the conflict between the demands on professors that are inherent in the research/publish model and the time and effort needed for significant improvements in instructional practice. CAL requires large amounts of time and effort for the development of effective programs. This time is not available to professors because of the time demands of the research/publish model. In addition, the research/publish model encourages the excessive intellectualization of all subjects. Significant improvements in university instruction in geography are likely to be produced only in the less intellectualized aspects of the subject, such as applied geography and environmental analysis. Institutions should permit professors to devote time and effort to instructional development without suffering penalties in career advancement and remuneration. (RM)