A remark on two notions of order of contact (Q1725894): Difference between revisions

In the seminal work [Several complex variables and the geometry of real hypersurfaces. Boca Raton, FL: CRC Press (1993; Zbl 0854.32001)], \textit{J. P. D'Angelo} proposed a programme in several complex variables of combining algebraic and geometric ideas to study the geometry of a real boundary by way of its contact with complex analytic objects. An important idea is the notion of measurement of the order of contact with a complex analytic variety. The complex analytic varieties (or subvarieties) at a point are studied by means of an ideal $I$ in an appropriate germ of functions, e.g., holomorphic or smooth. \par Over the last several decades two measures of the order of singularity of an ideal have been proposed for $q$-dimensional complex analytic varieties, the \textit{D'Angelo $q$-type} and the \textit{Catlin $q$-type}. When $q=1$ these measures coincide and it had been asked whether the same might be true for $q>1$. In this nicely written paper, the author shows that they differ in general for $q>1$. The proof is based on the study of an intermediate measure, the \textit{generic value} of an ideal. For $q>1$ the Catlin $q$-type coincides with the generic value but the D'Angelo does not in general. This permits the construction of counterexamples where the two notions of $q$-type can differ by any prescribed magnitude.