I just tested samba on two computers.

W98
Suse 10

For the letters ÄÖÜ, and others, Kate writes twobytes-codes on txt-files in the W98 machine, and W98 writes them as 1 Bytecode on similar files the Susemachine. The result is chaotic.

The problem only appears in simple textfiles. The fullygrown texteditors seem to translate automatically, and when I change filenames it also seems to work perfectly.

Will this be a real problem in the practical life ?
Is there any easily installed solution, or a readymade translationscript, that I can use on mixed-up textfiles ?