Homogeneous liaison and the sequentially bounded licci property (Q2161174)

Let \(R = k[x_1,\dots,x_N]\) be a homogeneous polynomial ring over a field \(k\). The author considers liaison classes of homogeneous ideals in \(R\), where the links are all assumed to be with homogeneous complete intersections (as opposed to Gorenstein liaison classes). An ideal \(I\) is said to be (homogeneously) licci if there is a sequence of such links taking \(I\) to a complete intersection. A natural question is whether it is possible to start with a licci ideal and arrive at a complete intersection using links that at each step use forms of the smallest possible degree. It was shown by \textit{C. Huneke} et al. [Contemp. Math. 448, 129--139 (2007; Zbl 1132.13304)] that this is not always possible, but it was also shown by \textit{F. Gaeta} [C. R. Acad. Sci., Paris 234, 1833--1835 (1952; Zbl 0046.14801)] and much later by the reviewer and \textit{U. Nagel} [Prog. Math. 280, 103--132 (2010; Zbl 1211.14054)] that there are natural situations where this is possible. Subsequently, \textit{E. Chong} [Face vectors and Hilbert functions. Cornell Univ. (PhD Thesis) (2015)] introduced the weaker sequentially bounded licci property which still allows some control of the degrees used in the links. \textit{K. F. E. Chong} [J. Algebra 445, 221--231 (2016; Zbl 1393.13033)] used this idea to study a very interesting special case of the Eisenbud-Green-Harris (EGH) conjecture. He asked whether sequentially bounded links were possible for all homogeneously licci ideals. The current author addresses this question, as well as a related question of \textit{C. Huneke} and \textit{B. Ulrich} [Bull. Lond. Math. Soc. 39, No. 3, 384--392 (2007; Zbl 1114.13011)]: Are all licci ideals strongly licci after a suitable change of variables and choice of term order? He answers both questions in the negative. His main tool is a careful study of certain Betti tables.