Throughout an entire decade, the Internet has brought unmanageable amounts of information to the average user’s fingertips. Since this growth will only continue, it is vital that users are supported in converting this universe of information into improved productivity and opportunity instead of being swamped and paralyzed. Failing to address information overload will cost enterprises and individuals money, often in ways that are not easily measured: Costs that result from lowered productivity and from mislead business decisions. To really satisfy user needs and restricted budgets, the myriads of information need to be structured and organized in an intelligent and user-oriented way. Technically, appropriate architectures to integrate existing archives with an intelligent news retrieval engine are to be developed. The research approach in the discussed OmniPaper project is investigating ways for drastically enhancing access to many different types of distributed information resources. The key objective is the creation of a multilingual navigation and linking layer on top of distributed information resources in a self-learning environment, thus providing a sophisticated approach to manage multinational news archives with strong semantic coupling, delivering to the user more than the sum of the individual service features.