* textset -> data * equalise version number * delete all missing translations * remove unused English texts and their translations