>Many years ago, more than I care to remember, I used to program accounting machines. One thing all our "programs" had was a check digit appended at the end of numeric codes so that it was easy to detect misstypings. I cannot remember how those check digits were calculated but it had to be very simple: only the most expensive machines could multiply or divide. Anybody knows how the check digit was calculated?

> I am taking a few days off, the first in several months, and I know that not having the answer risks spoiling the fun.

That we dooooo not want.

for 1 digit they take mod(x,7) where x is the number without check digit. For a 2 digit check they would take mod(x,93).

They may not have been able to multiply or devide, but they could 'shift' 'push' and 'pop' like nobody can today if my memory is correct :)

Marc