I've read single characters off the chat log and broken them down into byte arrays, and seen some interesting results.
The Japanese Hiragana "aa" is the bytes [160,130]. However, the same syllablary in Windows is different bytes when encoded in UTF-7, UTF-8, and UTF-16.
Am I missing something completely, or does XI have it's own character set?
UTF-8 would certainly make the most sense, since it's fact that the standard Ansi text is stored in arrays of single byte characters.