There seems to be a typo in the table used for the "unknown type" part of the content-type sniffing algorithm, as found at
http://www.whatwg.org/specs/web-apps/current-work/multipage/infrastructure.html#content-type-sniffing:-unknown-type Specifically, the row with the comment "UTF-16LE BOM" lists FF FF as the BOM in question, rather than FF FE. Apologies if this has already been reported. -- Aaron Crane ** http://aaroncrane.co.uk/
