Sorry I was late only reply. >> In , Andreas Schwab wrote: > >> BTW, I think we should convert HELLO to UTF-8. Do you see any symbol > >> in HELLO not supported by UTF-8? > > > > Isn't it currently UTF-8? > It's iso-2022-7bit, and converting it to utf-8 would lose the cjk > variants. `iso-2022-7bit' has been adopted for HELLO file, as it could contain CJK variants differently, but if we are to go into depth of various scripts and symbols that ISO-2022 does not support, we could only use UTF. ISO/IEC 10646 Annex T provides the way to describe the language in plain text by using Language Tags, such as follows: 󠀁󠁺󠁨开发󠀁󠁿、󠀁󠁺󠁨󠀭󠁴󠁷開發󠀁󠁿、󠀁󠁪󠁡開発󠀁󠁿 Language Tags are not recommended by the Unicode (as long as there is other ways to describe language, such as XML lang attributes), but it is the only way to describe language in plain text. Or switch we switch HELLO file from plain text to some XML format? # Emacs still doesn't handle Language Tags, but I think it wouldn't be # difficult in some future time... I'll attach updated HELLO examples as follows, that incorporates all the languages I could find "HELLO" word, plus Language Tag usage examples (someday Emacs may support it?) and fixed the logical order of Arabic language. -- --------------------------------------------------------------------- kawabata.taichi@gmail.com KAWABATA, Taichi