I read an xml document that contains the entity ” this is defined in the dtd as having unicode code point 0x201d. When I output the same document in utf-8 format with entities expanded the byte sequence 0xe2 0x80 0x9d is written (this can be seen in a hex viewer) which does display as right double quotes. How come the byte sequence 0x201d is not output? where does e2809d come from and why does it display correctly?