[seqfan] Incorrect b-files

Charles Greathouse charles.greathouse at case.edu
Mon Jan 25 15:38:16 CET 2010

I recently (Thursday/Friday) sent njas nine b-files that I
re-computed, after having found errors.  Most of these errors seemed
to be caused by a numbering script or other tool used in creating the
sequence, like in this (fake) example:

390 390390390390390390390390390390390390390390390390390390390
391 3913913913913913913913913913913913913913913913913913913913
392 39239239239239239239239239239239239239239239239239239239239
393 39339339393393
394 9339339393393393933933939339339393393393933933
395 3953953953953953953953953953953953953953953953953953953953953

Others were typos or just mistakes of some kind.

(I feel obliged at this point to repeat Neil's warning
IMPORTANT: Thousands of people use the sequence database every day.
Please take great care that what you send is absolutely correct.
The standards are those of a mathematics reference work.)

First: Along those lines, would someone recompute A130228? a(148) and
on are presumably wrong.
Second: Any ideas on how to prevent these errors in the future? Of
course more metadata would be helpful: these truncations would be easy
to catch if they were in sequences marked ,incr, or ,monotone,. But I
wonder what other kinds of errors are common.  (Experiences?)

Charles Greathouse
Case Western Reserve University

More information about the SeqFan mailing list