[hiisukun]

Perhaps because I'm new to this stuff, I enjoyed the writeup. I wonder if I'm out of place expecting a single run through of a-z 0-9 to determine the range of chars present in the password?

It turns out (due to repeated chars) to only have 14 unique chars. This single run through would have reduced the alphabet size (A, in the article) from 36 to 14. The 432 iterations becomes 168.

I'm sure there are other optimisations I'm missing!