[seqfan] Re: Help me understand these sequences

[Douglas McNeil]

> The definitions elsewhere suggest that this means numbers n with n+2
nonprime, but
> then why are 8 and 63 not in the sequence?  (And others beside: I don't
think it's a case
> of an omission but my misunderstanding,)

Given my catastrophic March Madness bracket, I should know better than to
gamble, but I'd be willing to lay good money that you're not
misunderstanding anything.

I don't know about this particular sequence, but it shares a property with
some of the other sequences recently discussed.  There were ~160 such
sequences in recent.txt on the weekend, and I looked at 53 of the unedited
sequences.  Of those, only 15 had values I could reproduce.  The remaining
38 all had errors, for an error rate of 70%.  There were ~20 previous
corrections by other people, which means that even assuming that every other
sequence is correct, there's at least a 35% error rate just among the ones
listed there.

A144565 ("a(n)=p-th digit of 2^(1/2) where p = n-th prime."):
> By the description, this is a duplicate of A090204.  But the values
> don't match at all.  What is this?
>

I tried and bounced, but my search wasn't very fault-tolerant.


Doug

-- 
Department of Earth Sciences
University of Hong Kong