Hello Math-Fun, I'm investigating, together with Carole Dubois, a fascinating (for us) sequence: Say the fraction a/b has quotient q and the fraction b/a has quotient r. We want that the set of digits used by a U b has no common element with the set of digits used by q U r. Example 1: 2/3 = 0,6666... (q) 3/2 = 1,5 (r) The sets {2;3} and {0;1;5;6} are disjoint. Example 2: 41/148 = 0,2770270270270... (q) 148/41 = 3,609756097560975... (r) The sets {1;4;8} and {0;2;3;5;6;7;9} are disjoint. We have turned this idea into a draft: https://oeis.org/draft/A333437 Are there strong programmers who could extend our 50-term file? The last two terms have respectively 4 and 6 digits (!): 5412/178596 = 0.03030303030... (q) 178596/5412 = 33 (r) {1;2;4;5;6;7;8;9} vs {0;3} And this: Is this seq infinite? Thank you, Best, É. (and C.)