>> 	- for non-western languages how should fmt behave?
>> 	  for example, japanese/chinese/korean text does not have space between
>> 	  chars.
>Presumably a sufficiently robust i18n library would include
>character-set-specific functions which fmt could use to know if a line
>break was appropriate between two characters, and, if so, how to do
>it..

	are there any hints in Solaris manpage?  i don't remember
	correctly...

>What is the convention for line breaks in japanese, chinese, and
>korean?  (Japanese has multi-character words, correct?  Are
>multi-character words broken across lines in printed text?)

	it is okay to break lines during multi-character words (no hyphen :-P).
	in some cases, we try to avoid it though.

itojun