Data warehouses are a primary means
for a consolidated view on the data within an enterprise and
frequently a first step in integrating enterprise information
systems. Above all, data warehouses are used for analyzing
enterprise data online, giving the possibility to aggregate and
compare data along dimensions relevant in the application domain.
Typically time is one of the dimensions we find in data warehouses
allowing comparisons of different periods. The instances of
dimensions, however, change over time - countries unite and
separate, products emerge and vanish, organizational structures
evolve. In current data warehouse technology these changes cannot
be represented adequately since all dimensions are (implicitly)
considered as orthogonal, putting heavy restrictions on the
validity of OLAP queries spanning several periods.

In this paper we briefly propose an architecture for
temporal data warehouse systems which allows the registration of
temporal versions of dimension data and the transfer of data
between different temporal versions.