Cube tilings as contributions of algebra to geometry (Q1261217)

This paper is an excellent survey on the Hajós-Minkowski theorem and some related results of G. Hajós, L. Rédei, Ph. Furthwängler, R. M. Robinson, K. Corradi and S. Szabó. The author also surveys the very actual problem on O. H. Keller's conjecture, and describes the method that is based on the factorization of finite abelian groups. In his paper [Period. Math. Hung. 17, 265-277 (1986; Zbl 0613.52006)] the author sharpened the Hajós' method, and opened a new combinatorial attack against the problem. The author and \textit{K. Corradi} [Period. Math. Hung. 21, No. 2, 95-100 (1990; Zbl 0718.52017)] found some interesting extreme constructions. In this way, by further clever construction, the Keller conjecture has recently been disproved by \textit{J. C. Lagarias} and \textit{P. W. Shor} [Bull. Am. Math. Soc., New Ser. 27, No. 2, 279-283 (1992; Zbl 0759.52013)]. For dimension \(n \geq 10\) there is a cube tiling which does not contain twins, i.e. cube pairs where two cubes join along full \(n-1\)-dimensional hyperface. The \(7 \leq n \leq 9\) cases are undecided. This survey helps in understanding the problem; the author also sketches some elegant simplified proofs, e.g. that of Hajós for Minkowski's conjecture on lattice-like cube tilings.