On sets of vectors of a finite vector space in which every subset of basis size is a basis. II (Q438886): Difference between revisions

The MDS conjecture is one of the central problems in coding theory and in Galois geometries. In the language of coding theory, it is a conjecture on the maximal length for linear MDS codes of dimension \(k\) over a finite field of order \(q\), and in the language of Galois geometries, it is a conjecture on the maximal size of \(n\)-arcs in the finite projective space PG\((k-1,q)\) of dimension \(k\) over the finite field of order \(q\). Recently, via an ingenious new technique involving polynomial methods, \textit{S. Ball} [J. Eur. Math. Soc. (JEMS) 14, No. 3, 733--748 (2012; Zbl 1241.15002)] proved the MDS conjecture for \(q\) a prime. One of the new ideas he developed was a coordinate-free version of Segre's lemma of tangents. Now, the techniques are extended to discuss the MDS conjecture for \(q=p^h\), \(p\) prime, \(h>1\). The authors here prove the MDS conjecture for all dimensions \(k\leq 2p-2\), and reprove the conjecture for \(q\) prime.