A suggestion concerning the future of OEIS.

karttu at megabaud.fi karttu at megabaud.fi
Wed Aug 1 12:18:46 CEST 2001

Jud McCranie wrote:

> Subject: Re: OEIS gripes
> Date: Thu, 26 Jul 2001 09:48:25 -0400
> From: Jud McCranie <jud.mccranie at mindspring.com>
> To:   seqfan at ext.jussieu.fr
> At 02:56 PM 7/26/2001 +0000, Labos Elemer wrote:
>>A remark which was or still it is hidden somewhere in OEIS web pages
>>promizes to contributors that they might become immortal if
>>sending sequences to EIS...
> I expect that the sequence database will be used long after we are gone.

I would guess that even in our lifetimes the volume of submissions
might grow so large that no single person can take care of it all
(even if Neil spent all his waking time with it).
So some kind of collaborative effort will be needed one day.

Here's my suggestion, taking hint from how usenet is organized.

We could have hierarchical subsections like:

oeis.base.2.encodings (for various combinatorial & other structures)
oeis.base.2.word-problems (like Thue-Morse or Rabbit sequence related infinite words
                           or maybe the concept of k-regular sequences gives better


oeis.tables (any sequence which is a triangular array would be cross-submitted also here)

or the hierarchy might be strictly modeled after
AMS 2000 Mathematics Subject Classification 

Each top-level section (oeis.combinatorics, oeis.number-theory)
and also the subgroups like oeis.base.10 (if the volume is high)
would have its own "moderator(s)" or "referee(s)", who would
handle all the submissions to that section, and gently correct
the newbies in case their submissions were incorrect or
directed to the wrong subsection. "Cross-submissions" to
more than one section would be allowed.

The format for submissions would be as strict as it is now
(no free format text-messages, and maybe even stricter
with regards to formulas, g.f.'s and program code, possibly allowing
auto-generation of the terms on the fly), and the system would still
allot a globally unique A-number for each accepted submission.

If each sequence would be kept under a source control system
like SCCS or such, the system would have a log of all the
changes made to each entry. (Or maybe in a special database
allowing easy storing and retrieval of the integer sequence data).

The most important advantage of this kind of organization
would be that the users of the database could directly home
to the section they are interested in, and the serious
mathematicians would not need to care at all about
the thousands of submissions arriving under base.10 section.


Antti Karttunen

PS. Here's the AMS section 11-XX = Number theory as a sample
from http://www.ams.org/msc/11-XX.html

     11Axx Elementary number theory {For analogues in number fields, see 11R04} 
     11Bxx Sequences and sets 
     11Cxx Polynomials and matrices 
     11Dxx Diophantine equations [See also 11Gxx, 14Gxx] 
     11Exx Forms and linear algebraic groups [See also 19Gxx] {For quadratic forms in linear algebra, see 15A63} 
     11Fxx Discontinuous groups and automorphic forms [See also 11R39, 11S37, 14Gxx, 14Kxx, 22E50, 22E55, 30F35, 32Nxx] {For relations with quadratic forms, see 11E45} 
     11Gxx Arithmetic algebraic geometry (Diophantine geometry) [See also 11Dxx, 14-XX, 14Gxx, 14Kxx] 
     11Hxx Geometry of numbers {For applications in coding theory, see 94B75} 
     11Jxx Diophantine approximation, transcendental number theory [See also 11K60] 
     11Kxx Probabilistic theory: distribution modulo $1$; metric theory of algorithms 
     11Lxx Exponential sums and character sums {For finite fields, see 11Txx} 
     11Mxx Zeta and $L$-functions: analytic theory 
     11Nxx Multiplicative number theory 
     11Pxx Additive number theory; partitions 
     11Rxx Algebraic number theory: global fields {For complex multiplication, see 11G15} 
     11Sxx Algebraic number theory: local and $p$-adic fields 
     11Txx Finite fields and commutative rings (number-theoretic aspects) 
     11Uxx Connections with logic 
     11Yxx Computational number theory [See also 11-04] 
     11Z05 Miscellaneous applications of number theory 

And 11Axx: from http://www.ams.org/msc/11Axx.html


 Elementary number theory

 {For analogues in number fields, see 11R04} 

11A05 Multiplicative structure; Euclidean algorithm; greatest common divisors
11A07 Congruences; primitive roots; residue systems
11A15 Power residues, reciprocity
11A25 Arithmetic functions; related numbers; inversion formulas
11A41 Primes
11A51 Factorization; primality
11A55 Continued fractions {For approximation results, see 11J70} [See also 11K50, 30B70, 40A15]
11A63 Radix representation; digital problems {For metric results, see 11K16}
11A67 Other representations
11A99 None of the above, but in this section

I guess the section 11A63 means our beloved base-sequences.

And 11Bxx = Sequences and sets from http://www.ams.org/msc/11Bxx.html

11B05 Density, gaps, topology
11B13 Additive bases [See also 05B10]
11B25 Arithmetic progressions [See also 11N13]
11B34 Representation functions
11B37 Recurrences {For applications to special functions, see 33-XX}
11B39 Fibonacci and Lucas numbers and polynomials and generalizations
11B50 Sequences (mod $m$)
11B57 Farey sequences; the sequences ${1^k, 2^k, \cdots}$
11B65 Binomial coefficients; factorials; $q$-identities [See also 05A10, 05A30]
11B68 Bernoulli and Euler numbers and polynomials
11B73 Bell and Stirling numbers
11B75 Other combinatorial number theory
11B83 Special sequences and polynomials
11B85 Automata sequences
11B99 None of the above, but in this section

More information about the SeqFan mailing list