The Unicode TechnicalCommittee requests that WG2 change its definition of UTF-8 to allow therepresentation of the code points U+FFFF and U+FFFE. These are disallowed inISO/IEC 10646, but are clearly an anomaly: other non-characters (U+1FFFE,U+1FFFF, etc.) as well as the new non-characters U+FDD0..U+FDEFareallowed.
Moreover,these code points are all legal in HTML: see the SGML declaration
(http://www.w3.org/TR/REC-html40/sgml/sgmldecl.html).
The 10646definition of UTF-8 should be amended as soon as possible to allow allnon-characters to be represented in UTF-8.