I loved this, since it's a different approach to genetic programming (with the FIFO list) that could work in different kinds of problems. Also the way it is applied to password cracking itself, is an extremely interesting way to exploit the non-random properties of real world passwords.
I really appreciate your enthusiasm. It is one of the core concepts and passwords seemed to be a good way to explore my FIFO idea in a practical manner. It might well have been done before, but from preliminary searches I could not find anything on it(not that I searched really hard). Due to its simplicity, I suspect it will lend itself to some pretty efficient implementations.
I would be interested to hear what other kinds of problems you see it being used for.
Not sure exactly what would be another application, but it's a mental model that looks "general", when we have a few prerequisites, that is, not just a solution, but multiple solutions to a given problem (find a password among many hashes in this case), and where solutions could be related in some way so that using the past solutions could restrict the search space.
