Preface

This book came about because the physical separation of the two authors by the Atlantic Ocean was a surmountable space in the pursuit of a blending of different ideas, approaches and teachings about spatial information systems. Something of a coincidence in our time-space activity paths led us to co-operate in offering an introductory course on geographic information systems for graduate geography students who knew little about information from a computer science perspective. This kind of audience, representative, we believe, of a much larger group of students and practitioners interested in computer processing of spatial data, has an intuitive feel for space, yet an ignorance of relevant principles or formalisms from mathematics and data processing.

Our combination of perspectives from informatics and geography, in the interest of furthering the understanding of spatial information systems, has led to the writing of a book that emphasizes spatial semantics as an organization principle. Our concern is with concepts, principles and ways of organizing; not directly with bits and bytes or the latest algorithm. So we deal with the nature of spatial data and spatial problems. We treat a variety of geometries that are necessary for positioning features in space and representing how they might be related. We present methods of conceptual modelling developed in computer science that provide valuable intellectual aids for sorting out different spatial problems.

Consequently, this book is in the context of geomatics, the fusion of ideas from geosciences and informatics. It is about spatial information systems, the assembly of computer hardware and software, data and intellectual capital that, after only a twenty-five year lifetime, have come to be seen by some people as the combined microscope and telescope for better solving a variety of problems in the world.

We see spatial information systems as toolboxes, as resources, and as a sign of a frame of mind. The practitioner, whether in a government agency for global natural resources monitoring, or working in geographic data processing for a public utility company, or drafting in a house basement with a personal computer to create the latest plan for political districts, can look at a software system as a set of tools for undertaking varied tasks. We see a spatial information system as more than this, as a substantial resource, the tools and the data, for problem solving or for academic research or educational purposes. In time it may be that such a resource will change the way in which some academic disciplines are taught in schools and colleges. In any event, we see the tools and resources, in their current form, as valuable aids leading to a better understanding of a dynamic spatial world.

Yet spatial information systems represent, too, a perspective, a way of looking at problems or thinking about the world. It is interesting to note that the commercial geographic information systems software industry has discovered the value of ‘seeing relationships based on geography’. We believe that the fields of geography and cartography, even with at least a two thousand year history in describing and measuring properties of the earth, and an appreciation already for the spatial relationships point of view, are being enriched by the recent developments in computer software science and engineering.

We also believe that the development of an understanding of how to use computers for spatial data is improved by a knowledge of disciplines that deal with space, not only geography and cartography, but also others like architecture, cognitive science, geology, geometry, geotechnics, meteorology and surveying. So this book is also oriented to the informaticians interested in developing spatial information system products. Indeed, spatial information system engineers often underestimate the complexity of spatial data, the semantics of which are much richer than those found usually in conventional computer use domains like office automation, accounting, student records, library archiving, management, and so on. For this group of people, with this document we can introduce not only some tools and models which can be useful in solving practical problems in geomatics, but which can also serve as foundations for the development of a general data theory.

Serving a mixed audience with one book has meant we have had to bridge several worlds. Not only have we had to blend computing and spatial problem solving, but we have aimed to reach both academician and practitioner, both system designer and database manager, and the intuitive and the formal learner. We have had to deal with different, and at times conflicting, uses of language and particular terminology. Not only are there different word uses for essentially the same phenomena depending on the field or discipline, but there are also differences from the point of linguistics. Absent any standard terminology, establishing a common ground for a French informatician and an originally British geographer living in the USA was part of the effort in putting the book together.

Our approach is conceptual, having a strong pedagogic orientation with many examples presented for edification and assimilation. Moreover, the variety of examples also provides a framework for evaluation, and a realization that there can be a solid systematic conceptual organization for many concepts, tools and perspectives. We have concentrated on showing how to organize and use spatial information rather than on how to process or manage spatial data. Just as an understanding of art is facilitated by looking at many pictures, so we think spatial data organization can be better understood by seeing many examples. The variety is inherent in the cultural and natural realms of this Earth; it is also present in the many viewpoints and needs that users of spatial data have. We see this book as helping others to appreciate the varied world of spatial information systems.

We do not know exactly whom ‘others’ might be; therefore we have not oriented the book to just one group. For example, it is likely that informaticians and software engineers with no background in a spatially oriented science can benefit from the first two parts of the book; for geographers and cartographers much of the first two parts may be a review. For some spatial scientists, Part Three could be the starting point. We have tried to accommodate the reader who may wish to delve only into certain topics by having some repetition of material. Perhaps the practitioners who have different backgrounds and contexts for working with spatial information systems may need to consult only selected chapters. The book has not been written to accompany any particular course (indeed, it has been oriented to people outside as well as in teaching institutions), and we believe it offers an alternative to other introductory books in the field by reason of its conceptual semantic approach, combining information handling and spatial concepts.

In the beginning, we briefly identify components of a spatial information system (Chapter 1, Geomatics), including the reasons for their existence (Chapter 2, Purposes and Types of Spatial Problem). Then we review the nature of spatial data (Chapter 3, Semantics: Objects, Surfaces, Data). In several chapters in Part Two we present details of the foundations for treating and organizing spatial data: different geometries (Chapter 4, Geometries: Position, Representation, Dimensions, and Chapter 5, Topology: Graphs, Areas, Ordering), and structures for organizing and representing (Chapter 6, Tessellations: Regular and Irregular Cells, Hierarchies), and manipulating and processing spatial entities (Chapter 7, on the manipulation and transformation of spatial data and objects, and Chapter 8, on spatial analysis operations for spatial data).

The synthesis in Chapter 8 leads into, in Part Three, Conceptual Modelling for Spatial Data, the presentation of the techniques for conceptual modelling of spatial situations, chiefly the entity-relational modelling approach (Chapter 9, on design methodologies for information systems). Many examples for point and line entities (Chapter 10, Conceptual Modelling of Line-oriented Objects), and for areas and solids (Chapter 11, Conceptual Modelling for Areas and Volumes) are then provided as tastes of spaghetti and pizza, respectively. Chapter 12, Spatial Object Modelling: Views, Integration, Complexities, concludes the third part by a review of the process used in synthesizing different models, and then raises some issues to be dealt with in the fourth part of this book.

Part Four emphasizes techniques and principles for accessing, retrieving and using information from databases containing a variety of spatial entities. The basis of relational algebra is presented (Chapter 13, Algebras: Relational and Peano Tuple), followed by examples of queries (Chapter 14, Spatial Queries). After a discussion in Chapter 15 of the challenging topics of spatial indexing and integrity constraints, the book then moves on to discussions of currently developing fields, at first looking at multimedia and hypermedia concepts (Chapter 16, Multimedia Spatial Information Systems and Hypermaps), and then presents some ideas about intelligent spatial information systems (Chapter 17). These last two chapters (perhaps these are the dessert!) establish how the stage is changing, pointing the way to the anticipated developments of the next few years.

While we have presented this book in traditional form, we wish we could have had the time and resources to prepare it as a...