

On thinking it over, “proper” spelling of foreign words has done its own share of damage to English spelling. We don’t just have to learn our own spelling conventions, we also have to learn foreign ones. Or not (sent to you from Cairo, Illinois, locally pronounced “care-oh”)
There is one big group of losers from UTF-8: Eastern Europeans. Greek and Cyrillic need two bytes per letter and there’s no way around it in Unicode, while national code pages only used one byte per letter.
On the other hand, unless you have a big strictly monolingual database (or strictly national language/English), it seems worth taking the hit to size for the flexibility.