The type of a good semigroup and the almost symmetric condition (Q2295435)

The concept of a \textit{good semigroup} was formally stated in a paper by \textit{V. Barucci} et al. [J. Pure Appl. Algebra 147, No. 3, 215--254 (2000; Zbl 0963.13021)] in order to study value semigroups of noetherian, analytically unramified, one-dimensional, semilocal, reduced rings, arising from curve singularities with more than one branch. Properties of such semigroups were also considered by many other authors, e.g. \textit{A. Campillo} et al. [J. Lond. Math. Soc., II. Ser. 60, No. 2, 420--430 (1999; Zbl 0974.14020)]. The class of good semigroups is larger than the class of value semigroups; in particular, they are not finitely generated as a monoid. Thus, in order to study such semigroups, one must work only with semigroup techniques. In the paper under review, the authors restrict to the case of good \textit{{local}} subsemigroups $S$ of $\mathbb{N}^2$. Following their usage, in this review we omit the word local. In their paper [``The Apéry set of a good semigroup'', Preprint, \url{arXiv:1812.02064}], the authors introduced, for any $\boldsymbol \beta \in S$, the notion of Apéry set $\mathrm{Ap}(S,\boldsymbol \beta):=\{\boldsymbol \alpha\in S\mid \boldsymbol\alpha-\boldsymbol \beta\notin S\}$ of $\boldsymbol \beta$. (These sets were studied by \textit{R. Apéry} in his paper ``Sur les branches superlinéaires des courbes algébriques'' [C. R. Acad. Sci., Paris 222, 1198--1200 (1946; Zbl 0061.35404)].) They showed, even as they work with infinite sets, sometimes, it is possible to produce partitions of them in a finite number of subsets, that they call levels, and the number and the nature of the levels give key informations on the semigroups $S$ they are dealing with. Let $A\subseteq S$ be a subset for which there exists $\boldsymbol{c}\in S$ such that $\boldsymbol{c}+\mathbb N^2\subseteq S\setminus A$; for such a set $A$ the authors construct in Definition 2.1 a partition $A=\bigcup _{i=1}^N A_i $ with subsets $A_1,\ldots,A_N$, $A_i\cap A_j=\emptyset$ for $i\neq j$. For good ideals $E\subset F$ in $S$ the authors define a distance $d(E\setminus F)$; if $E$ is a good ideal in $S$, then $A:=S\setminus E$ is a subset as above; the authors prove that $N=d(S\setminus E)$. In Definition 3.3 the \textit{type} $t(S)$ of a good semigroup $S$ is defined; the definition of type is coherent with the corresponding definitions for numerical semigroups and one-dimensional CM-rings. Let $M$ be the maximal ideal in $S$; if $S-M$ is a good relative ideal in $S$, then $t(S)=d((S-M)\setminus S)$. One important criterion: $S$ is symmetric iff the type of $S$ is $1$. Let $K$ be the canonical ideal of a good semigroup; $K\subset\mathbb N^2$; $K$ is a good relative ideal of $S$. Since $S\subset K$, $S$ is symmetric iff $K=S$. A good semigroup $S$ is said to be almost symmetric if $M=KM$. Clearly any symmetric semigroup is almost symmetric. In the second part of the paper, the authors study the class of almost symmetric good semigroups. In Section 4 they study good semigroups. In Section 5 they study duality for almost symmetric good semigroups.