Résumé
Today, in the direction of research we can observe a trend from the investigation and development of "pure" systems to activities usually tailored to specific applications. Concurrently, computers have penetrated daily life in a degree which had not been forecast even in optimistic estimation scenarios 20 or 30 years ago. This trend has not yet ended; actually we can expect the emergence of many new "intelligent products"; i.e. digital computing systems are more and more integrated or merged into many other systems of different physical domains and characteristics. In this paper, the impact of these general trends on the current state and future directions of fault tolerance activities is discussed.