Adieu (fwd)

Henry Gould gould at math.wvu.edu
Thu Jan 11 18:41:58 CET 2007


Fellow Seqfans!

While I have *not* decided to unsubscribe, I agree with the complaints 
from Antti and Richard. The sheer volume of junk email I get every day 
is fifty percent Viagra/Cialis/Nigerian-Millions-inhertance plus fifty 
percent of what Antti and Richard arfe complaining about. It is most 
disturbing and counterproductive.

I am hanging in  but reluctantly.

Regards,

Henry Gould

= = = = = = =

Jonathan Post wrote:
> First we lost Antti from seqfans, and now -- tragedy! -- we lose 
> Richard Guy.
>
> I am one of the offenders.  I have submitted hundreds of sequences, 
> none of them officially "nice."  I have made apparently annoying 
> emails to seqfans.  I have, since gentle hints were made, cut back 
> drastically on both OEIS submsissions and seqfans gmails.  I did so to 
> improve the signal-to-noise ratio.  I have only submit, in the past 
> month or more, seqs from published papers, and those pre-edited by 
> Associate Editors of OEIS, as coauthored. I would rather hear what 
> Richard Guy has to say than say anything myself.
>
> Can any other of "the usual suspects" voluntarily embargo their 
> submissions and emails, as I did, and with respect to Neil J. A. 
> Sloane's "vacation"?  That vacation has not stopped some people.
>
> Please, isn't there an urgent need to respond to what Antti and 
> Richard Guy have said, in hopes of luring them back?  Please?
>
> -- prof. Jonathan Vos Post
>
> On 1/11/07, *Richard Guy* <rkg at cpsc.ucalgary.ca 
> <mailto:rkg at cpsc.ucalgary.ca>> wrote:
>
>     I haven't received a copy of this, so perhaps
>     my request to be cut off was implemented even
>     before I made it!  I send again, also using
>     the old (?) address.       R.
>
>     ---------- Forwarded message ----------
>     Date: Wed, 10 Jan 2007 19:59:49 -0700 (MST)
>     From: Richard Guy <rkg at cpsc.ucalgary.ca <mailto:rkg at cpsc.ucalgary.ca>>
>     To: seqfans at seqfan.net <mailto:seqfans at seqfan.net>
>     Subject: Adieu
>
>     Sequence fans,
>                    Swan song.
>
>     1.  Helena Verrill, in a talk at
>     New Orleans on   Series for  1/pi
>     mentioned  ``Almkvist - Zudilin
>     numbers''
>
>     1  -3  9  -3  -279  2997  -19431
>
>     which I don't find in OEIS, but
>     then, as my Mother said, I'm not
>     a good looker.
>
>     2.  In a paper written with Alex Fink & Mark Krusemeyer there
>     is the following table.  These sequences have a good deal in
>     common, but what is common is not always recorded at each
>     sequence.  I will elaborate on this in a message to Neil's
>     Dream Team before much more water has flowed under the
>     bridge.
>
>     \begin{center}
>     \begin{tabular}{cc|cc|cc|cc|cc}
>       $r$ & OEIS \# & $r$ & OEIS \# & $r$ & OEIS \# & $r$ & OEIS \# &
>     $r$ & OEIS \#
>     \\
>     --14 & \ldots  & --8 & A070998 & --2 & A001519 & 4 & A002878 & 10
>     & A057081 \\
>     --13 & A001570 & --7 & A070997 & --1 & A000012 & 5 & A001834 & 11
>     & A054320 \\
>     --12 & A085260 & --6 & A049685 &  0  & A011655 & 6 & A030221 & 12
>     & A097783 \\
>     --11 & A077417 & --5 & A001653 &  1  & \ldots  & 7 & A002315 & 13
>     & A077416 \\
>     --10 & A078922 & --4 & A004253 &  2  & A057079 & 8 & A033890 & 14
>     & \ldots  \\
>       --9 & A072256 & --3 & A001835 &  3  & A005408 & 9 & A057080 & 15
>     & A028230
>     \end{tabular}
>     \end{center}
>
>     3.  Also arising from this paper is an iterative process,
>     which may be familiar to most seqfans, but which I don't
>     seem to be able to tie up with anything in OEIS.  Here's
>     a not very good example, because the degrees of the
>     polynomials go up too fast.  Start with an array, say
>     the Omar Khayyam triangle
>
>                               1
>                             1   1
>                           1   2   1
>                         1   3   3   1
>                       1   4   6   4   1
>                     1   5  10  10   5   1
>                   .........................
>
>     and then write the diagonals as polynomials, whose
>     coefficients, after normalization by  (-1)^r * r!
>     form the array
>
>                               1
>                             0   1
>                           0   1   1
>                         0   2   3   1
>                       0   6  11   6   1
>                     0  24  50  35  10   1
>                   0  120 274 225 85  15   1
>                 .............................
>
>     which, in this case, I believe to be Stirling
>     numbers of the first kind.  The diagonals are
>     A000012, A000217, A000914 (or A115057 -- is
>     this different?), A001303, ... .   Repeat the
>     process, yielding
>                               1
>                           0   1   1
>                       0  10  21  14   3
>                   0   1  11  47  97  96  36
>                ..............................
>
>     (are these in OEIS ?  Schroeder numbers ???
>     this done by hand, and probably containing
>     errors)  and repeat the process ad lib ...
>
>     The array formed by the sequences listed
>     under 2. above form a similar, but in some
>     ways more interesting example, and presumably
>     many of the arrays in OEIS will also yield
>     sequences of arrays which will be of interest.
>
>     4.  I reluctantly request that I be removed
>     from the seqfan list since the messages have
>     reached a volume, and have often a content,
>     matched only by the spam that I receive.  I
>     will send to Neil, or to his Dream Team, if
>     I have any serious comments or queries about
>     OEIS, a beautiful project which I have seen
>     grow since I first met Neil over 40 years ago.
>
>     Best wishes to all serious contributors --
>     they know who they are.     R.
>
>






More information about the SeqFan mailing list