Today I rewrote the complete language detection and the language handling. It is now based on the ISO 639 codes and it should be much easier to support more languages. Besides this the language strings are in one big array instead of hundreds of variables.