[seqfan] Underlined sequences.

[franktaw at netscape.net]

In https://oeis.org/draft/A190013, Jonathon Vos Post has submitted a 
sequence from a paper. The (finite) sequence in the paper looks like:

0, 1_, 3, 0_, 2, 4_, 1, 3_, 0, 2_, 4, 1_, 2,  3_, 4, 0_, 1, 2_, 3, 4_, 
0_, 0, 2, 2_, 3_, 3, 4, 4_, 1_, 1

where the trailing underscore here represents an underlined number in 
the paper. The question is how to represent this sequence in the 
database.

Note first that this is described as a sequence over a 10-element 
alphabet, so the actual values are not really significant. My 
suggestion was to add 5 to the underlined values, giving

0, 6, 3, 5, 2, 9, 1, 8, 0, 7, 4, 6, 2, 8, 4, 5, 1, 7, 3, 9, 5, 0, 2, 7, 
8, 3, 4, 9, 6, 1

If the values were strictly positive, I would favor making the 
underlined values negative, but that won't work here. One could take n_ 
to -1-n, but that seems even more artificial than the +5 suggestion. 
Likewise for n -> 2n, n_ -> 2n+1.

The main problem with my solution is that someone seeing the sequence 
in the paper and looking for it in the database is probably not going 
to find it.

Maybe we need two sequences, one with the original values ignoring 
underlines, and the other with the +5.

Any suggestions? opinions?

Franklin T. Adams-Watters