Character Encoding in the Trenches
Good timing for this discussion -- I've been struggling with the right way to do character encoding just recently. The interaction between the HTTP Content-Type header and the encoding attribute of the XML declaration (and even the machine's default locale!) creates a confusing environment.
Wouldn't it be nice if there was no ambiguity and the HTTP charset parameter always matched the encoding of the XML feed? Stay tuned ... (evil grin!)
Source: Dare Obasanjo aka Carnage4Life - Sex, Lies and XML MIME Types