They are formally defined in RFC 2781 (e.g. ftp://ftp.rfc-editor.org/in-notes/rfc2781.txt)
UTF-32* are defined in UAX #19, as Doug wrote, and are also IANA-registered charset names.
markus
Doug Ewell wrote:
Jungshik Shin <jshin at mailaps dot org> wrote:Note that "UTF-16 little-endian" is not technically the same as "UTF-16LE"; the former implies the presence of a BOM while the latter implies that none is present.)Where does this distinction come from?The sources I checked were UTR #17, "Character Encoding Model," and UAX #19, "UTF-32." The latter does not specifically talk about UTF-16BE or UTF-16LE, but uses the same definitions to distinguish UTF-32, UTF-32BE, and UTF-32LE that we are using here.Mark Davis can probably point you to other sources as well.
