[boost] [serialization][documentation] character encoding oddity

Hi, http://boost.org/libs/serialization/doc/contents.html shows three strange characters at the top of the page (). Apparently, those characters have the values EF BB BF in hexadecimal notation. Interpreted as UTF-8 sequence, they encode the Unicode codepoint FFFE. This codepoint is used as a marker for the endianness in encodings other than UTF-8. I doubt such a marker would be legal in or would make sense for an UTF-8 encoded document. Also, the document's content-type meta-tag claims the document uses ISO-8859-1. I suspect there's a problem in the toolchain used to create the document (oherwise, I would have fixed it in the repository). Regards, m Send instant messages to your online friends http://au.messenger.yahoo.com