\(\alpha\)-flocks and hyperovals (Q1275309): Difference between revisions

This beautiful and important (albeit quite technical) paper is a must read for anyone interested in the theory of hyperobals in finite desarguesian planes of order \(q=2^e\). The usual theory of \(q\)-clans and flocks of quadratic cones is generalized to the setting of flocks of cones over translation ovals (and hyperovals). These are called \(\alpha\)-flocks where \(\alpha\) is the generator of the Galois group of the field \(GF(q)\) associated with the given translation oval. In fact it is shown that every known infinite family of hyperovals is associated with a corresponding family of \(\alpha\)-flocks. Most the definitions and results are too technical for a review such as this, so we state in detail only the final surprising ``pièce de résistance'' that appears as a consequences of the general theory: The ``Cherowitzo ovals'' (first described by the author in 1984 and conjectured by him always to be ovals) really are ovals. Specifically, let \(F=GF(2^e)\) \(e\) odd. Let \(\sigma\) be the unique automorphism of \(F\) for which \(\sigma^2\) is the Frobenius automorphism \(\sigma^2:x\mapsto x^2\). Let \(f:F\to F\) be defined by \(f(x)= x^\sigma +x^{\sigma +2}+ x^{3\sigma+4}\). then the point-set of \(PG(2, 2^e)\) given by \(\Omega= \{(x,f(x),1): x\in F\}\cup \{(1,0,0), (0,1,0)\}\) is a hyperoval in \(PG(2,2^e)\), i.e., a set of \(q+2\) points, no three on a line.