[seqfan] Re: seqs whose |differences| are 1,2,3,4,...

[Robert Munafo]

If you start with anything other than 1, then it's no longer the
"lexicographically
earliest", no matter what the second or subsequent terms are, with holes or
without.

Not a complaint as such, I'm just pointing out that you're diverged from the
problem statement at the beginning of this thread,

On Sun, Mar 14, 2010 at 10:58, Ron Hardin <rhhardin at att.net> wrote:

> Using Paul's algorithm starting with val instead of 1, the first infinitely
> deferred (never filled) hole is
>
> val hole
> 1 5
> 2 7
> 3 11
> 4 15
> 5 19
> 6 23
>
> so apparently 4*val-1 except for 1.
>
> The series always fills downwards to 1 first.
>
> --
 Robert Munafo  --  mrob.com