On the containment problem for fat points (Q2074391)

Let \(R=\mathbb{K}[\mathbb{P}^N]\), where \(\mathbb{K}\) is an algebraically closed field of any characteristic and \(N\geq 2\). Let \(I=\bigcap_i I(P_i)^{m_i}\), where \(P_i\) are distinct points in \(\mathbb{P}^N\), \(I(P_i)\) is the ideal of all forms that vanish at \(P_i\) and the multiplicity \(m_i\) is a nonnegative integer. For this type of ideals we define the \(m\)-th symbolic power as \[I^{(m)}=\bigcap_i I(P_i)^{mm_i}.\] The containment problem is to decide for which \(m\) and \(r\) the symbolic power \(I^{(m)}\) is contained in the ordinary power \(I^r\). \textit{C. Bocci} and \textit{B. Harbourne} [J. Algebr. Geom. 19, No. 3, 399--417 (2010; Zbl 1198.14001); Proc. Am. Math. Soc. 138, No. 4, 1175--1190 (2010; Zbl 1200.14018)] introduced an asymptotic quantity, knows as the resurgence, whose computation is clearly linked to the containment problem.\\ Definiton.\\ Given a nonzero proper homogeneous ideal \(I\) in \(R\), the resurgence of \(I\), denoted by \(\varrho (I)\), is defined as the quantity \[\varrho (I)=\sup \{\frac{m}{r} : I^{(m)}\nsubseteq I^r\}.\] In general, directly computing \(\varrho (I)\) is quite difficult and it has been determined only in very special cases. In the paper authors compute the resurgence of \(I(Z)\) for two classes of fat point subschemes. First they study the subscheme \[Z=\sum_{i=1}^n m_iP_i\] in \(\mathbb{P}^N\), where the points \(P_i\) are distinct and collinear and they prove that in this case \[I(Z)^{(m)}=I(Z)^m\] for all \(m\in \mathbb{N}\), thus \(\varrho(I(Z))=1\). Then they consider the subscheme \[Z=m_0P_0+m_1P_1+m_2P_2,\] where \(P_i\) are noncollinear points in \(\mathbb{P}^N\) and \(m_0\leq m_1\leq m_2\) are nonnegative integer. In this situation they classify all fat point ideals with \(m\)-symbolic defect zero for all \(m\) (and hence that \(\varrho(I(Z))=1\)). Let's recall that the \(m\)-symbolic defect of a homogeneous ideal \(I\) of \(R\) is the minimal number of generators of the \(R\)-module \(I^{(m)}/I^m\). The authors formulate the following statement. Theorem. For \(I\) as before, \(\mathrm{sdefect}(I(Z),m)=0\) forall \(m\in\mathbb{N}\) if and only if one of the following conditions holds: \begin{itemize} \item[1.] \(m_0+m_1\leq m_2\), \item[2.] \(m_0+m_1>m_2\) and \(m_0+m_1+m_2\) is even. \end{itemize} In other cases they compute resurgence of the ideal. More precisely, if \(m_0+m_1>m_2\) and \(m_0+m_1+m_2\) is odd, then \[\varrho (I(Z))=\frac{m_0+m_1+m_2+1}{m_0+m_1+m_2}.\]