[seqfan] Re: Can we create a workable taxonomy for classifying all the various kinds of sequences in the OEIS? - not manually

[Georg.Fischer]

Dear Wayne,

being also "nowhere even close to as familiar with the OEIS
as many others" I still dare to throw in my thoughts on your
proposal.

So far the main classifications are by the keywords, and by
the OEIS index. When I search for a sequence, I do it mostly by
entering some numbers in the search field. I used keyword
catalogs in libraries in my youth, but today, I mainly use
freetext search (with Google). They even go into the b-files.

I think it could be worthwhile to lean back and ponder over the
nature of the 300k+ OEIS sequences. But I think it is totally
impossible to *manually* introduce any additional classification
of the OEIS sequences (because of their huge number).

Any such classification or categorization must, if ever, be
done *programmatically*. And long before such an automatic 
categorization I personallay would wish some standardized
mechanism to run the program(s) for some sequence, and
compare their results with the b-files.

Moreover, the categories you mentioned would contain so many
members that they are probably not useful for searching for
a specific sequence without additional criteria.

There may be areas where a generalized overview might be helpful,
see for example my list of 48 sequences of Clark Kimberling:
<http://www.teherba.org/index.php/OEIS/Negative-Positive>.
But this already implied quite some work.

Best regards - Georg Fischer

Am 07.04.2018 um 08:35 schrieb Wayne VanWeerthuizen:
> Neil wrote, "The Index to the OEIS does a pretty good job."
...