Tame actions of affine group schemes and tame stacks (Q765736): Difference between revisions

In this short note, the author compares two notions of tameness, one being that of the tame action as defined by \textit{T. Chinburg, B. Erez, G. Pappas} and \textit{M. Taylor} [Duke Math. J. 82, No. 2, 269--308 (1996; Zbl 0907.14021)], the other being that of a tame stack as defined by \textit{D. Abramovich, M. Olsson} and \textit{A. Vistoli} [Ann. Inst. Fourier 58, No. 4, 1057--1091 (2008; Zbl 1222.14004)]. The author shows that if \(G\) is a flat affine group scheme of finite type over a Noetherian affine base \(S\), with a tame action on an affine scheme \(X\) of finite type over \(S\), and if all inertia groups are finite, then the quotient stack \([X/G]\) is tame. The author proves there is even an equivalence in the following case: if \(C\) denotes the ring of invariants for the above action, \(Y=\text{Spec}(C)\) is Noetherian, \(G\) is a finite affine group scheme locally free over \(S\) and if the morphism \(X \rightarrow Y\) is flat, then the action is tame if and only if the quotient stack \([X/G]\) is tame. The author then focuses on inertia groups in lieu of assuming that the action is tame and proves that if \(S\) is Noetherian, \(G\) is flat of finite type over \(S\), \(X\) is of finite type over \(S\), and if all inertia groups are finite and flat, then \([X/G]\) is tame if and only if all inertia groups are linearly reductive.