Just use UTF-8. Works great; less filling. All the other WGs are using it as the replacement for ASCII to meet their IESG requirements on character sets. For us to do anything different is as stupid as saying that we should do EBCDIC (even assuming that EBCDIC met the requirements). Paul