[seqfan] Re: A094627: Sequence whose n-th term digits sum to 2n+1.

[Klaus Brockhaus]

  Am 15.11.2010 09:41, schrieb Joerg Arndt:
> * Charles Greathouse<charles.greathouse at case.edu>  [Nov 15. 2010 08:16]:
>> I don't understand the definition of A094627.  The (decimal) digit sum
>> of a(n) is indeed 2n+1, but it's not clear to me how one is chosen.
>> There is a generating function, but that's not a part of the
>> definition in this case, and of course it would be good to check that
>> it is actually correct.
> Looks OK:
> (x^2 + 2*x + 1)/(10*x^3 - 10*x^2 - x + 1)
> ? (1+x)^2/((1-x)*(1-10*x^2))+O(x^33)
> 1 + 3*x + 14*x^2 + 34*x^3 + 144*x^4 + 344*x^5 + 1444*x^6 + 3444*x^7 + 14444*x^8 + ...
>
> I wish people would use '*' where it belongs.
>
> The choice seems arbitrary and
> IMHO the sequence is not interesting.
>

The %N line gives a property of the sequence, not a definition. There 
are lots of other sequences having that property, e.g. 1, 3, 23, 25, 27, 
... or 1, 111, 11111, 1111111, ...
I don't see a condition that singles out just the given sequence.

The definition could be converted to a comment and the g.f. (or the 
closed formula - which I didn't check yet) taken as definition, but the 
sequence seems still not interesting.

Klaus