Morita ``equivalences'' of equivariant torus spectral triples (Q1935701)

Let \(X\) be a compact surface of genus \(g\geq 1\). It is well known that two surfaces \(X\) and \(X'\) are topologically the same, whenever their genera \(g=g'\) are the same. If one replaces \(X\) by a Riemann (metric) surface and topological equivalence by conformal equivalence, then the condition \(g=g'\) is not sufficient, since there are many (in fact, the continuum \({\mathbb R}^{6g-6}\)) distinct conformal metrics that one can put on the same topological surface. In noncommutative geometry, \(X\) is replaced by a \(C^*\)-algebra \({\mathcal A}\) (noncommutative topological space) and topological equivalence by the so-called Morita equivalence of \(C^*\)-algebras. The metric structure is encoded by a spectral triple \(({\mathcal A}, {\mathcal H}, D)\) attached to the algebra \({\mathcal A}\). The problem is again that topological equivalence does not imply spectral triple equivalence. In this very interesting paper, the author solves the above ``moduli problem'' for spectral triples in the case when \(X\) is an \(n\)-dimensional noncommutative torus (see [\textit{M.~A.~Rieffel} and \textit{A.~Schwarz}, Int. J. Math. 10, No. 2, 289--299 (1999; Zbl 0968.46060)]). For this purpose, he introduces a special class of equivariant torus spectral triples; the main result says that for such triples the Morita equivalence is an equivalence relation, i.e., a true isometry between the noncommutative metric spaces.