>>105592031
>>105592011
Pic is some of the tabulated results of 10k runs of female names with max-len 4.

I wish there was a dataset for full names of my locale so I could train the actual name frequency to represent the population, but there are only lists with popular names or the official data with just firstname by number.

As it stands names are sampled uniformly from a sexp but the word class (first name, surname, article ...) have certain probabilities I made up.