[seqfan] Re: A104851

[Harvey P. Dale]

Neil:

	I did fix A104851 as you requested, but when I searched the OIES
with the (now corrected) terms of that sequence it turns out that
A198176 is the same as the corrected A104851.  It also turns out that
A198176 was submitted by me!  Should A104851 be deleted?
	Best,
	Harvey

-----Original Message-----
From: N. J. A. Sloane [mailto:njas at research.att.com] 
Sent: Sunday, February 05, 2012 2:54 PM
To: seqfan at list.seqfan.eu
Cc: Harvey P. Dale; njas at research.att.com; WestA at wlu.edu
Subject: Re: [seqfan] A104851

Harvey said:

(start)
A104851 lists "Primes from merging of 10 successive digits in decimal
expansion of e."  It seems obvious that the primes listed will depend on
how many digits of e are generated before looking for 10-digit primes.
I generated 10 million digits of e and could not find the first term of
the above sequence in them.  On the other hand, using only 10,000 digits
of e I found 398 10-digit primes.  Only 10 of those
398 primes appear in the above sequence, out of a total of 18 terms
included in it.
(end)

Me: this kind of sequence only makes sense if we scan from the left, and
record the primes as they appear. (Otherwise every 10-digit prime will
be there.)

So I added this comment:

Scan decimal expansion of e from left to right, recording any 10-digit
primes seen. - N. J. A. Sloane, Feb 05 2012

I added a similar comment to A104843, where it is easier to see what is
going on.

It looks like the sequence is simply wrong. Harvey, can you correct it
please? You said "using only 10,000 digits of e I found 398 10-digit
primes."
That should be the real beginning of the sequence!

For A104843 I checked the first few terms by hand, and they look
correct.

In A104843 there is this comment:

Cf. A104843 - A104862.

I guess all them should be regarded with suspicion.

Neil