Let me start this discussion before filing an enhancement request :)
From time to time, I would open PN2 and type (or paste) directly into it. Sometimes there are non-ANSI characters (ie Unicode characters).
It would be nice if PN could accept these characters without having to switch encoding to UTF-x first. The same behaviour with NotePad would be nice: it accepts characters (Unicode or not) and only warns user to change encoding only if the content can't be saved in ANSI.
Actually, currently there's a bug related to UTF-8 (http://code.google.com/p/pnotepad/issues/detail?id=1018) which might be used to allow this behaviour:
make PN2 go through internally encoding to UTF-8 then back to ANSI without user intervention :D