NB It's not a simple learning question. It's actually from my ORM software pluralizing entities' names unexpectedly. For curious and/or programming readers' reference, see [the source code].
So, when I declare a class Activity and run my softie on it, it produces a statement to the database called Activities. Great, so we know it pluralizes. I can dig that. In fact it's kind of neat when we look into the schema later on.
Then, I created a class Human and the junky ORM mapper produced a statement called Humen. Is that just a faulty algorithm (since Woman and Man should be Women and Men)? It seems to be a bit far fetched.
Although there are a lot of man suffixed words, there are examples when the ending man originates from elsewhere (roman, human etc.) so generalizing the pluralization like so is very odd. Also, the developers might be using a different language altogether, where man means donkey and the plural of it is hazaa.
So my question is whether it's a viable pluralization (in English, that is).