Maybe, but almost everything outside the basic multilingual plane is so infrequently used that it's a waste of space to use a fixed-width format. But I could s

Maybe, but almost everything outside the basic multilingual plane is so infrequently used that it's a waste of space to use a fixed-width format. But I could see it being a useful replacement for UTF-32... If you need to be able to calculate code points in constant time, and you don't mind wasting one bit out of every 64. Self synchronizing, constant time code point count from buffer size, and close to as space efficient as UTF-16, but with none of the surrogate mess.