Supported character sets?

First, I want to thank everybody who provided responses to my question wrt a complement charset. Next issue...one that I think I asked a year ago and I either a) can't find the answer or b) the answer wasn't clear. I'll greatly appreciate any help I can get with the questions below. What character set (or sets) do Rebol characters and strings belong to? Given Rebol's cross platform nature I'm betting that it depends somewhat on the platform. Is there a default character set? There doesn't appear to be any immediate evidence that I can construct characters beyond the 255 boundary. So here's some scenarios: - can I construct strings encoded in UTF-8 and/or UTF-16 - can I construct strings of UCS-2 or UCS-4 characters - does the REBOL i/o support reading UTF-8/16? If I'm on a WinNT box I live in a world occupied by both Unicode (really UCS-2) and a currently selected ANSI codepage like Latin-1 or Shift-JIS. So imagine my currently selected ANSI codepage is Latin-1 -- will Rebol read in chars in Latin-1 or a Unicode based encoding? Sorry for all the questions but I build apps that handle global data and I'm struggling to figure out whether I can use Rebol for non-ASCII data. Working for a commercial software firm I've had first hand experience building cross-platform products that can handle many different encoded character streams; I know it isn't child's play, but in a world of global commerce it's becoming a necessity. Best Regards, Gavin.