> And besides, the whole idea of a "species" is dubious at some level, because it's merely a human invented category. I think something like K nearest neighbor

> And besides, the whole idea of a "species" is dubious at some level, because it's merely a human invented category. I think something like K nearest neighbors algorithm[1] describes our understanding of species pretty well. You have a scatter of animals and you try to find groupings. The finer-grained the groupings, the less distinction there is between members on each side of an edge. Or to put it in more simple terms: say you're putting pebbles together. How many do you need before it stops being "1, 2, 3, 5 pebbles" and becomes "a pile of pebbles". [1] https://en.wikipedia.org/wiki/K-nearest_neighbors_algorithm https://en.wikipedia.org/wiki/K-nearest_neighbors_algorithm