Code points are not the proper abstraction, since a character can be composed from a variable number of code points. Also, bytes from 0-127 in UTF-8 (MSB is 0)

Code points are not the proper abstraction, since a character can be composed from a variable number of code points. Also, bytes from 0-127 in UTF-8 (MSB is 0) are ASCII characters. In multibyte code points, the MSBs of the first byte are 11 and of continuation bytes 10.