Re: browsers and unicode surrogates

From: Steffen Kamp (
Date: Mon Apr 22 2002 - 12:12:47 EDT


>Section 5.2.1 discusses the BOM. Also see my previous mail.

Ah right. I agree that a BOM is definitely a good idea there but as
0xFFFE0000 could either be a UTF-32LE BOM or a UTF-16LE BOM followed by a
U+0000 (should't happen too often, though) it is not necessesarily easy
to determine the correct encoding from the BOM. Therefore, for me it
seems mandatory to have the encoding defined in the http header if using
UTF-16 or UTF-32.

BTW: On Mac OS X Chimera 0.2.2 <> (uses
Mozilla's Gecko rendering engine) correctly displays the UTF-16 page.


Steffen Kamp

This archive was generated by hypermail 2.1.2 : Mon Apr 22 2002 - 12:56:17 EDT