Pair of digits self-differences

[hv at crypt.org]

"Eric Angelini" <keynews.tv at skynet.be> wrote:
:Could someone compute two dozen more terms ? I would like
:to check if my seq. is ok before submitting it again to the
:OEIS -- after a few tries with a lot of mistakes...
:
:1 20 21 102 103 200 201 211 250 255 330 2000 10201 10206 11102
:
:1) a(n) < a(n+1) and
:2) the n-th digit of the sequence is the absolute difference of 
:the digits forming the n-th pair of digits of the seq.

I think you need to define the sequence more fully.

It seems to require additionally (3) a(1) = 1; (4) a(n) is chosen
to be the smallest number consistent with a(i) (for i < n) and the
other requirements.

But if I assume those additional rules, I get a sequence (by hand)
1 2 3 5 6 9 16 17 90 98 170 181 901 1090 8010 70001 80010 90001 ...

Hugo