Part I begins with surfaces and introduces some key contrasts: geometry versus topology (whether smooth deformations matter or not), intrinsic versus extrinsic (what properties can someone on a surface or in a space ascertain without leaving it), local versus global, homogeneous versus nonhomogeneous (the same everywhere?), and closed versus open. Other concepts include orientability and connected sums and products and bundles. Weeks also looks here at the simple (flat) three-manifolds.
Part II considers geometry on the sphere and the hyperbolic plane, and general geometries on surfaces: "Except for the sphere and the projective plane, which have elliptic geometry, and the torus and the Klein bottle, which have Euclidean (flat) geometry, all surfaces can be given hyperbolic geometry." This leads to an explanation of the Gauss-Bonnet formula and the Euler number.
Part III proceeds to geometry on three-manifolds, looking at four-dimensional space, hyperspheres, hyperbolic space, fibre bundles, and the full range of geometries on three-manifolds. Attempts are made to prime our intuitions for these, subject to the limitations of the printed page:
"The geometry of H² x E is also homogeneous but not isotropic. Like S² x E, it has different sectional curvatures in different directions. Vertical slices have zero curvature, while horizontal slices have negative curvature. Figure 18.2 provides a rough illustration of H² x E geometry."
The mathematical conclusion of The Shape of Space comes here, with a statement of Thurston's Geometrization Conjecture (which was not yet proven in 2002), classifying three-manifolds into eight kinds. This second edition has a new Part IV, "The Universe", discussing applications to cosmology and the possibility of determining from observation something of the universe's topology or geometry, but that's really a digression.
The Shape of Space is moderately involved, but nothing in it should be beyond engaged high school students. On the other hand, it could profitably be read by undergraduate mathematics students, since it treats material poorly covered in most university curricula, and even research students who have studied differential geometry might find something new in it, perhaps some novel intuitions for visualising spaces.
The cosmology is now a bit dated, but the mathematics is not, the proof of the Geometrization Conjecture notwithstanding. There is also a web site with supporting videos and games and visualisation software, some of it pitched at younger children (tic-tac-toe and mazes on toruses) and some at undergraduates and researchers:
"The Curved Spaces software includes flat spaces as well, in spite of its name. It has no games, but the graphics are good. Includes all ten flat 3-manifolds, as well as several geometrically different versions of the 3-torus, some of which are quite beautiful and surprising. You can see the manifolds in stereoscopic 3D if you have red-blue glasses."
- External links:
- buy from Amazon.com or Amazon.co.uk
- information from Jeffrey R. Weeks
- share this review on Facebook or Twitter