[seqfan] Re: Mystery of A049476

[Allan Wechsler]

Here is a proposed interpretation of A049476. I have only checked it by
hand that it agrees with 1, 2, 5, 13, 14, so somebody should write some
code to verify it.

The rows of the triangular table described at A088643 all appear to
terminate in copies of earlier rows. For example, row 7 is (7,6,5,2,3,4,1),
while row 11 is (11,8,9,10,7,6,5,2,3,4,1), which ends in a copy of row 7.

If this always happened, it would imply the conjecture that motivated
A088643 in the first place.

Row 11 exhibits a "delay" of 11-7=4 steps before the copy of row 7 begins.

We can't meaningfully assign a "delay" to row 1, but the delays for
subsequent rows form the sequence

1,1,1,4,1,1,4,1,1,4,1,6,11,1...

This sequence is not in OEIS, but it should be, because (I think) A049476
gives the row numbers that generate record values for it.

Please, somebody double-check me.

On Mon, Aug 2, 2021 at 2:23 PM Sean A. Irvine <sairvin at gmail.com> wrote:

> Thank you for all the responses.   Neil and I both wrote to the authors of
> the paper and there is still some chance we will track down a copy.
>
> Doug, it seems you tried a similar set of possibilities to what I tried.
> Every so often, I would find a way to produce a few terms and think this
> was it, but alas no, another dead end.
>
> Sean.
>
> --
> Seqfan Mailing list - http://list.seqfan.eu/
>