Centralizers in mapping class groups and decidability of Thurston equivalence (Q2209758)

``We find a constructive bound for the word length of a generating set for the centralizer of an element of the Mapping Class Group. As a consequence, we show that it is algorithmically decidable whether two postcritically finite branched coverings of the sphere are Thurston equivalent.'' \textit{Thurston maps}, originating in the field of complex dynamics in one variable, are branched coverings of the 2-sphere whose branch points have finite orbits; using triangulations of the 2-sphere, these can be described purely combinatorially, and Thurston defined a natural combinatorial equivalence relation (\textit{Thurston equivalence}). A basic question is then whether Thurston equivalence is combinatorially decidable; this is answered affirmatively in the present paper (elaborating on previous work of \textit{N. Selinger} and \textit{M. Yampolsky} [Arnold Math. J. 1, No. 4, 361--402 (2015; Zbl 1359.37103)]. The whole story started when Thurston described in a topological language which Thurston maps with hyperbolic orbifolds are combinatorially equivalent to an (essentially unique) rational map of \(\mathbb C\) (see a paper by \textit{A. Douady} and \textit{J. H. Hubbard} [Acta Math. 171, No. 2, 263--297 (1993; Zbl 0806.30027)]); such a rational function can then be seen as a canonical geometrization of a Thurston map.