Introduction

This monograph is written with two purposes: to help bridge the gap between theory and practice in databases, and to help bridge the gap between research in query language and research in other areas (outside databases, and some outside Computer Science) that are clearly related to querying. The motiva- tion behind the book, then, is a belief that the gaps exist, and that this is a bad situation.

Databases are a strange area. On the one hand, they are clearly an applied field. Databases (and related services) are a multi-billion, world-wide software industry. Some advances (for instance, in query optimization) make it to mar- ket in relatively short time. Changes in the industry (for instance, the move to data warehouses) provoke comparable changes in the field, creating whole subareas of research. On the other hand, databases are a theoretical field. Since the definition of the relational model by Codd, its main concepts have been tied to logic ideas, and logician’s methods have been used to study the model abstractly. This study has blossomed in entirely new subareas, like descriptive complexity ([56]) and finite model theory ([27, 69]). Thus, databases is, at the same time, a clearly applied and a clearly theoretical field of study.

This would be a good thing if theory and practice walked hand in hand. However, this does not always happen. To quote a familiar dictum, “In the- ory, theory and practice are the same. In practice, they are not.”1. While occasionally theory occupies itself with questions of practical interest, and practice does generate questions of theoretical interest, the mainstream body of theory and practice remain separated. This is not due to the will of the re- searchers -many theoreticians find delight in seeing theoretical developments put to good use in practice; and they are also eager to apply their skills to problems motivated by new areas of practice. It is also the case, in my opinion, that many practitioners consider the establishment of a solid, formal founda- tion for a practical problem a positive development. In fact, most database people (certainly this author) adhere to another familiar dictum, “There is nothing as practical as a good theory”. But theory and practice have different goals, demand different approaches and produce different results. Most of the time, the disconnect is present. It is for this reason that it is a pleasure to work on a subject that gives an opportunity to bridge this gap.