I was reading some BSD code (reverse.c) and came across this: ... bytes(fp, off); ... The man on bytes() defaults to Perl which says something about "changing character semantics to byte semantics". I'm guessing this is a BSD function somewhere (um-manned) that does this sort of thing. Now why would they want to do that? What is meant by character v. byte semantics, and why would I be interested in changing them? Olwe __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com