IRAIA will be an information retrieval system realised as a set of information repositories, where each of several communities maintains a collection of their own knowledge. Its service establishes an information space, where users navigate abstract spaces to perform their search for information. The approach of IRAIA focuses on a highly interactive environment to assist the users by presenting the information space by a semantic coordinate system that provides them with a clear orientation during their retrieval process. Within this project a technical infrastructure will be created to support the construction of an advanced and user-centred approach to very large information spaces. In our view, this infrastructure is not only a technical challenge but also the result of a corporate effort of various domain-specific and content-related organisations together with the users. In defining an infrastructure, we consider all aspects, from low-level transport layers to user interface. It is hard to evaluate new technology in the absence of real users and large collections. The best way to approve new technology is by real-world electronic testbeds used by actual users. IRAIA goes this way by providing a representative user group with information on the economy. The data come from huge databases of Economic Research Institutes and National Statistical Institutions. To live up with the expectations in our approach, IRAIA will provide access to an information space that contains time series and related text documents concerning various information domains of the economic analysis.