Proper handling of unicode strings

[LCID Fire <lcid-fire gmx net>]

I'm currently in the process of writing an application which needs to
support unicode - but I'm still a little confused of how to properly
handle it. Maybe someone can help me out here.

First of is it valid for e.g. utf8 strings to assume they are NULL
terminated? Would it be valid to call g_strdup on a utf8 string?

If not (and this is done quite often in the unicode glib part) I assume
I have to add the byte length of a string, right (which will bloat
function declarations)?