Perhaps, to put a HN spin on it, we could discuss the fact that Twitter is apparently aware of the difference between unicode chars and bytes, but what about SMS? After all, that 140 character limit comes from SMS's limit. Does SMS treat ♺ as 1 against the 140 limit or 3?
In case anyone is interested, I just tested a bunch of UTF-8 Japanese characters in Twitter. It counts each character as 1 on the front-end (countdown from 140 to -infinity), but counts the BYTES on the backend, so the front end will say you are under the limit, but then you click update and get an error message saying you are over the limit.
