reinforces the authors' point. Nevertheless, the uneven coverage 2D strings, and introduce logical operators and spatial relations. Chapter 6 describes some 2D string-based image database sys-is awkward. In addition, the coverage presents a some alternative techniques are more widely practiced and/or developed tems to demonstrate typical applications of the theory of Symbolic Projection to image information retrieval. than in fact they are-presenting a if understood to represent the state of current system development practices. The second Part II includes four chapters, namely Chapters 7-10. Chapter 7 describes algebraic properties of 2D string manipulation source of frustration is that the entire text suffers from use of difficult language. The entire text is written in dense academese and transformation in the form of an ''image algebra.'' A theoretical concept of ''empty space object'' is defined, followed that is simply inaccessible to mainstream information systems developers and data modelers. A final rewrite with the help of by presentation of various laws of generalized empty space, object manipulation, unification, and operator precedence. a technical editor could have improved (and a subsequent edition can still improve) the writing to make it more readable Chapter 8 deals with transformation of various types of projection strings and representations of object relations into and from and thus more useful to the target audience. This is especially unfortunate in that one of the central themes of the text is the generalized 2D strings. Object rotation based on the image algebra is also discussed in this chapter. Chapter 9 discusses the need for resolving communication difference between information systems developers and users. And in this sense the author's deficiencies of Symbolic Projection in representing line and point objects (direction, and distance). Slope projection, path didn't heed their own advice. The entire text would benefit from this treatment, and the result would be that a much wider projection, and polar projection are presented to demonstrate how Symbolic Projection can be extended to solve these defi-audience would be able to access its important contents.
ciencies. Finally, in Chapter 10, a number of different approaches to qualitative spatial reasoning are described. Although in the introductory section of this chapter the authors Peter Aiken suggest that ''Symbolic Projection also includes other character-Department of Information Systems istics that made it suitable for various forms of spatial reasoning Virginia Commonwealth University and in particular for qualitative spatial reasoning'' (p. 179), Richmond, VA 23284-4000 this aspect is not addressed in any notably substantial form.