I am trying to setup a web page that will have both Czech (ISO-8859-2) and English (ISO-8859-1) users hitting it.

I see how to allow MySql and Apache2 to accept both character sets, but I am wondering what happens if I accept a file upload encoded in Czech and try to save it to the Linux filesystem.

As I understand, Linux uses ISO-8859-1 encoding as the default. I also read a post saying that I can compile UTF-8 in the kernel as the default, but several posts mention that not all apps understand it.

I really don't understand how the different apps convert character sets, or even if they do.

The best thing would be for all apps and the server to use UTF-8. Anyone know of any problems with this?

Any other solutions to make sure that different client encodings will not cause problems?

Thanks in advance...

As far as I know, the kernel doesn't care what encoding a file uses and each program will have it's own story. If the program you want to use supports the Czech encoding then it should display it correctly anyway (You may have to select the encoding somewhere). In general there shouldn't be any major problem, though it all depends on what you want to do with the file.