I remebered I used the legendary 'notepad' to modify and save the Xml file, so I don't think there such an encoding problem.
That's exactly why.
I think that's why too. Remember that XML headers define the encoding (usually UTF-8), so the parser will switch to different decoding schemes regardless of what the file originally is. Notepad can save in UTF-8 format, but on my English machine it saves as ASCII by default. See what your Notepad saved it as.
Office can say "XML file appears to be corrupted" but a standard user will not know what that is all about. ("I'm opening a Word file! What XML file!?!")