[Logician]

Originally posted by hellsatan
So are you saying that we should ban characters that are not latin characters?

Actually I didn't say that.

Technical explanation is long, here is a recap: Different languages reside in different encodings. Most of the encodings are sort of "compatible" with "English" encoding. For example in Turkish encoding if you write letter A, it will be displayed as letter A in English encoding too since their ASCII Codes will be same. The difference of Turkish encoding would be using some special ASCII codes for Turkish letters and these would be assigned to less frequently used chars of English encoding and wouldnt clash with the basic ones.

However if you are using a completely different alphabet like Arabic, Korean, Chinese or Japanese, the number of different letters are so much that they will invade even the basic letters of english char set like A B C D etc. Then problems can begin if the script didnt think about this possibility.

For example if an arabic letter invades the place of backslash \ or $ sign in english char set, the script can regard it as a backslash or $ and try to escape it with "addslashes" command.

Well this was suppose to be recap I guess you get the gist.