Re: [math-fun] Wilson's sequence prime-cover problem

Heuristically speaking large "k" should occur often enough so that even when p=nonresidue mod 17, we should get noncoverage in in infinite number of cases with prime p. This is due to the divergence of the sum (1/k) where k is summed over those prime p, which in turn is due to the divergence of the sum 1/(N*logN). However, the same heuristic argument would expect that noncoverage cases should be pretty rarely encountered, despite presumably being infinite in number. -- Warren D. Smith http://RangeVoting.org <-- add your endorsement (by clicking "endorse" as 1st step)