Given for the first time at the london.pm tech-meet at the Fotango offices on 2005/02/24, this was a 5-minute rant about perl, character sets, and why noone can ever get them right. The slides were written in OmniGraffle for some bizarre reason, but I think it worked quite well, and may use the technique again some time.

Other utf-8 resources

• The Absolute Minimum Every Software Developer Absolutely, Positively Must Know About Unicode and Character Sets (Joel) - this is really really good. Not so good for perl specifics (and other languages with ‘real’ string types) - it talks mostly about maintaining byte sequences internally, and not some characters in an abstract ‘string’ type, but good for the history lesson.
• Mark’s utf-8 talk - the specifics of using perl to deal with utf-8.