On the structure of the directions not determined by a large affine point set (Q2435046)

Let \(H_\infty\) be a hyperplane of \(\mathrm {PG}(n, q)\) and let \(U\) be a subset of points of \(\mathrm{AG}(n, q) = \mathrm{PG}(n, q)\setminus H_{\infty}\). A point \(t\in H_{\infty}\) is \textit{determined} by \(U\) if there is an affine line through \(t\) containing at least two points of \(U\). For \(n=2\), there are many results on the extendability of affine point sets not determining a given set of directions. For \(n\geq 3\), the following result [\textit{S. Ball}, ``The polynomial method in Galois geometries'', in: J. De Beule (ed.) and L. Storme (ed.), Current research topics in Galois geometry. New York: Nova Science Publishers. 105--130 (2012)] is known. \noindent { Theorem.} Let \(q= p^h\), \(p\) an odd prime and \(h>1\), and let \(U\subseteq \mathrm{AG}(n, q)\), \(n\geq 3\), be a set of affine points of size \(q^{n-1}-2\), which does not determine a set \(\mathcal D\) of at least \(p+2\) directions. Then \(U\) can be extended to a set of size \(q^{n-1}\), not determining the set of directions. In the paper under review, the authors study the structure of the directions of a set \(U\subseteq \mathrm{AG}(n, q)\) of size \(q^{n-1}- \epsilon\), \(\epsilon > 2\), that are not determined by \(U\). The paper ends with a section containing an application of their results to partial ovoids of certain partial geometries.