The Unicode Consortium Discussion Forum

I believe there is an error in the following statement in page 63 of Chapter3: Conformance

• For example, when using UTF-16LE, pairs of bytes are interpreted as UTF-16 code units using the little-endian byte order convention, and any initial <FF FE> sequence is interpreted as U+FEFF zero width no-break space (part of the text), rather than as a byte order mark (not part of the text). (See D97.)

as the use of FEFF is deprecated as a zero width nobreak space character, even when it's not at the beginning of the file. Clearly when it is a the beginning of the file FEFF should be interpreted as the BOM character.

A) It may be deprecated, but that doesn't change its meaning when it does occur.

B) In the given example, the point is that you already know the text is little-endian (it was declared UTF-16LE), therefore there is no BOM in the text. If the initial sequence is <FF FE>, this is, indeed, interpreted as U+FEFF zero width no-break space (part of the text), rather than as a byte order mark (not part of the text).

The statement you quoted from the spec is correct as written.

You're right. I should have read Chapter 3, up to pages 97 and 98, where this is clearly explained.

Thanks