On the equation \({\mathbf V}*{\mathbf G}={\mathcal E}{\mathbf V}\) (Q5956869): Difference between revisions

Motivated originally by Krohn-Rhodes theory, the study of semidirect products of pseudovarieties of semigroups of the form \({\mathbf V}*{\mathbf G}\), where \(\mathbf G\) is the pseudovariety of finite groups, has a long history. It was found early on that many calculations of such products lead to the equation in the title, where \({\mathcal E}{\mathbf V}\) denotes the pseudovariety consisting of the finite semigroups whose idempotents generate semigroups in \(\mathbf V\). For instance, \textit{J.-C. Birget, S. Margolis} and \textit{J. Rhodes} [Bull. Aust. Math. Soc. 41, No. 2, 161-184 (1990; Zbl 0692.20046)] showed that it is true when \(\mathbf V\) is the pseudovariety of semilattices, based on a fundamental result of Ash. The authors make a systematic study of this general equation, based on syntactic methods. Most of the varieties \(\mathbf V\) under consideration lie within \({\mathcal D}{\mathbf S}\), the pseudovariety consisting of all finite semigroups whose regular \(\mathcal D\)-classes are subsemigroups, since the most powerful such tools are known in this case, based on the first author's results. The equation is completely solved for locally trivial pseudovarieties, and it is shown to hold for various other special subpseudovarieties of \({\mathcal D}{\mathbf S}\), such as \({\mathcal D}{\mathbf S}\) itself and \(\mathbf J\), consisting of all finite semigroups in which each \(\mathcal J\)-class is trivial.