For fuck's sake I am not defending UTF16's simplicity, I am defending that: fixed width can count code points (I worded it as "character") faster than variable

For fuck's sake I am not defending UTF16's simplicity, I am defending that: fixed width can count code points (I worded it as "character") faster than variable-length Then his dietrichepp tries to educate me two code points combined should be treated equaly with another single code point, WTF y u no normalization? Downvote me as you like, but you can't change the fact that UCS4 is used internally in Unicode systems. Any reason other than for faster code point counting? ----------------- dietrichepp also offended me that unicode characters should not count or offset. QTF: > Why do you want to count Unicode characters? Why do you care if it is fast to do so? Why would you ever need to use character-based string indexing?