PHP has a lot of trouble with multibyte strings (non-ASCII characters). The entire language was built assuming that each character is a byte. To solve this they invented the mb_strings functions which you can use instead of the standard functions (which work fine).
strlen($str);
mb_strlen($str); // correct
However, this is really a pain since you have to verify that the code you download/find online uses these functions or enable the mb_string_overload which then might break some code that actually needs char = byte calculations.
Does Ruby share this problem?
I think Ruby 1.9 clears up this underlaying assumption