[mmcnickle]

The same reason int(-1.999) is -1; The operation is different to integer division. I think of it as taking the "integer" part of the float.

[Izkata]

As the article mentions, this is called truncation. It truncates (cuts off) the value at the decimal point.