On Oct 1, 2008, at 8:59 AM, Austin Ziegler wrote: > I think that some libraries (including possibly CSV) should make > that effort, while others won't need it. To be honest, I doubt I would have made the effort if I had known this change was coming. It was challenging and I'm a wimp. ;) I originally decided to add m17n support because the most common question people email me with for FasterCSV amounts to it choking on various encodings. I always just show them how to transcode their data to UTF-8 using iconv. Ironically though, I didn't expect my m17n changes to cut down on these emails. The kind of people who email me with encoding issues aren't generally aware they are facing encoding issues, so it's unlikely they will set the encoding properly for their IO/String, which is what CSV now requires. The upside was that I could skip having to explain iconv to them though. :) James Edward Gray II