In a nutshell, I have the problem that I receive files from many computers, but then when I process the data and store it in another place, I get strange characters now and then, due to the differences in encoding As far as I know, one cannot know the exact encoding of a file created on another pc, only guess it. if one could know it, one could process them in the same encoding to avoid strange characters, doesn't it? How do ETL tools deal with unknown encoding of files, in order to avoid generating strange characters?