In Codd's relational model, the relation name groups together a family of functional dependencies over a set of attributes. For integrity and for maintenance purposes it is important to eliminate inherent redundancy within a relation due to the repetition of instances of a functional connection between attributes. This led Codd to propose a series of three normalizations. The manipulation of normal forms is governed by functional dependencies that are explicitly declared to exist within the relation. Since functional dependencies completely govern the decomposition rules of normalization, perhaps it is more sensible to take them as the elementary notions to be later synthesized into more complex structures, such as relations. Our goals, then are twofold. First, we will discuss how the use of functional dependencies lends itself to a rigorous and correct, yet clear and simple description of complex data relationships. Second, we will outline some new computational techniques to map functional dependencies into normal form relations algorithmically, thus allowing us to use the dependencies as a basic unit on the implementation as well as conceptual level.

The relational model for data base organization introduced clearly defined basic algebraic concepts whose properties are well understood. As a consequence, questions of a semantic nature arise. Due to the mathematical nature of the relational model, these questions cannot be answered completely by it. Therefore, a data model is discussed that takes into account basic semantic properties that object types and relationships among them apparently have in the real world. The model permits a data base to be partitioned into independent object types which are natural insertion and deletion units, and associations among them. Independent object types are described by characteristics which are meaningful modification units.This data model permits us to analyze the ideas behind the relational theory more precisely, in particular the meaning of functional dependency and normalization. Insertion-deletion rules for relations in third normal form are developed, and proposals are made to avoid some drawbacks that exist even when relations in third normal form are used as the user interface to a data base.