Algorithmic topology and classification of 3-manifolds (Q5898700)

The aim of this fine book is to study the algorithmic topology of low-dimensional manifolds: a branch of mathematics which has received a lot of attention, expecially in virtue of the exponential advancement of computer software. Almost all parts of the book are based upon courses that the author gave in different universities. However, the book contains a lot of new results and covers material not found elsewhere. One of the most important new result is the algorithmic classification theorem of Haken manifolds. Chapter 1 contains a comprehensive overview of the theory of simple and special polyhedra. Special spines, up to elementary moves (also known as Matveev-Piergallini moves), give a combinatorial representation of triangulated 3-manifolds which is easily turned into a computer presentation. This chapter also contains an existence result giving a criterion for two special polyhedra to be spines of the same manifold. Chapter 2 is devoted to the complexity theory of 3-manifolds. This notion is defined in terms of the almost simple spines of a 3-manifold. The complexity permits to introduce an ordering into the whole set of 3-manifolds. More precisely, the set of 3-manifolds is supplied with a filtration by finite subsets (that is, the subsets of 3-manifolds up to a fixed complexity), and this allows to break up the classification problem for all 3-manifolds into an infinite number of classification problems for some finite subsets. This approach is implemented in Chapter 7 where the author describes a way for enumerating 3-manifolds of a given complexity. It is also explained how to recognize such manifolds of smaller complexity by using the first integral homology groups and the Turaev-Viro invariants (the latter are described in Chapter 8). In particular, the author shows that all the 3-manifolds with complexity less than or equal 8 are graph manifolds and describes the first hyperbolic 3-manifold of complexity 9. Chapter 3 contains the first complete exposition of the Haken theory of normal surfaces. As remarked by the author, almost all known nontrivial algorithms in low-dimensional topology use the Haken theory or at least are derived from it. Chapter 4 is devoted to discussing several applications of the theory of normal surfaces. In particular, one can find many examples of algorithms based on the Haken theory as the algorithm for recognizing the (boundary) incompressibility of a given surface, the algorithms for recognizing the splittable links and the unknot, and the algorithm to detect when a 3-manifold is sufficiently large. Chapter 5 contains a new presentation of the Rubinstein-Thompson algorithm for recognizing the 3-sphere. Chapter 6 is the central part of the book in which the author proves the algorithmic classification theorem for Haken manifolds, including the author's original contribution. In particular, he proves the existence and the uniqueness of JSJ-decompositions and gives a detailed discussion of Stallings manifolds and their generalizations. Here, simple polyhedra are used to encode the hierarchies in 3-manifolds. Nowadays, for closed Haken manifolds other proofs of the algorithmic classification theorem can also be obtained by using the Thurston hyperbolization theorem for Haken manifolds containing no essential tori and annuli and by Sela's results on algorithmic recognition of hyperbolic manifolds (see the references of the book). Chapters 7 and 8 contain computer implementations of the recognition theorem and the Turaev-Viro invariants, respectively. This serves to describe an enumeration procedure for 3-manifolds of low complexity. In particular, the resulting 3-manifolds, up to complexity 6, are listed in the Appendix of Chapter 9 (together with pictures of minimal spines for some manifolds of complexity 7). The author is one of the major experts in the topology of 3-manifolds, and he has written a lot of interesting papers on the topic (for more references see the bibliography of the book under review, and the book indicated below). The contents of the book and the point of view adopted in their exposition fill a gap in the existing literature on the topic, and it will become a standard reference for algorithmic topology of 3-manifolds. The book is written in a very accurate and readable style. A lot of nice figures illustrate the contents of the book. It looks like as a selfcontained and fundamental work on the algorithmic 3-dimensional topology both for graduate students and researchers. I suggest to accomplish the reading of the book with the other fine monograph written by the author together with \textit{A. T. Fomenko} on the topic [Algorithmic and computer methods for three-manifolds. Mathematics and its Applications. 42 S. Kluwer Academic Publishers (1997; Zbl 0885.57009)]. This is a book I am very happy to have, and it is heartily recommended to anyone interested in studying topology of 3-manifolds from an algorithmic point of view.