
Re: binary search with arbitrary or random split
Posted:
Jul 26, 2012 9:27 AM


On Jul 26, 3:06 pm, Martin Brown <newspam...@nezumi.demon.co.uk> wrote:
> > Analysing the linear fixed split case : > > k : Nk > > 1 < k <= N/2 >
This has already been promptly corrected several posts ago. Ben made an insightful remark about that. We are now dealing with a random splitting which is a function of n. (This has also an intuitive meaning, as if it did not depend on n, we would not be "dividing" anything.)
> > Why did you delete the algebra that makes my meaning clear?
First of all, i can't delete your posts, and everyone can see them. Second you have a rather peculiar idea of what "clear" means.

