The paper introduces a relational model of data for large shared data banks, emphasizing the need for data independence and addressing issues of redundancy and consistency. The relational model is based on n-ary relations, a normal form for database relations, and a universal data sublanguage. The author discusses the inadequacies of existing noninferential, formatted data systems, particularly in handling data dependencies such as ordering, indexing, and access path dependence. The relational model is proposed to provide a more flexible and independent approach to data representation, allowing users to interact with relationships rather than specific relations, and to support a universal data sublanguage based on applied predicate calculus. The paper also introduces normalization to eliminate nonsimple domains, making the representation more efficient and easier to communicate between systems. Operations on relations, including permutation, projection, and join, are discussed, along with their applications in deriving relations from other relations. The author emphasizes the importance of symmetric exploitation of relations and the need for efficient translation of user requests into stored representations.The paper introduces a relational model of data for large shared data banks, emphasizing the need for data independence and addressing issues of redundancy and consistency. The relational model is based on n-ary relations, a normal form for database relations, and a universal data sublanguage. The author discusses the inadequacies of existing noninferential, formatted data systems, particularly in handling data dependencies such as ordering, indexing, and access path dependence. The relational model is proposed to provide a more flexible and independent approach to data representation, allowing users to interact with relationships rather than specific relations, and to support a universal data sublanguage based on applied predicate calculus. The paper also introduces normalization to eliminate nonsimple domains, making the representation more efficient and easier to communicate between systems. Operations on relations, including permutation, projection, and join, are discussed, along with their applications in deriving relations from other relations. The author emphasizes the importance of symmetric exploitation of relations and the need for efficient translation of user requests into stored representations.