In recent years, studies of historical populations have shifted from tracing large-scale processes to analyzing longitudinal micro data in the form of 'life histories'. This approach expands the scope of social history by integrating data on a range of life course events. The complexity of life-course analysis, however, has limited most researchers to working with one specific database. We discuss methodological problems raised by longitudinal historical data and the challenge of converting life histories into rectangular datasets compatible with statistical analysis systems. The logical next step is comparing life courses across local and national databases, and we propose a strategy for sharing historical longitudinal data based on an
intermediate data structure (IDS) that can be adopted by all databases. We describe the benefits of the IDS approach and activities that will advance the goals of simplifying and promoting research with longitudinal historical data.