No production code here, so that's not a problem. I guess the idea stems from my having taken a year to understand how your N-Queens thing worked and maybe because I'd like to be able to say "because I can too":)

The reality is that the unbounded nature of the datasets involved probably means that I will need to keeps stack usage at each recursion as small as possible, which I wouldn't have control of using the RE. It was a nice thought though.