From: Hans Aberg (haberg@math.su.se)
Date: Tue Jan 18 2005 - 18:09:33 CST
On 2005/01/18 22:58, D. Starner at shalesller@writeme.com wrote:
> "Jon Hanna" writes:
>
>>> In , the use of BOM is
>>> discouraged for use on UNIX platforms. So if endianness may appear to
>>> becomes a problem, it might be better to use UTF-8 externally, and then
>>> convert it to UTF-32/H/L internally in the program.
>>
>> Discouraged or not, it's in the standard, you have to support it.
>
> That may or may not be true in a standards conformance sense, but
> it's defintely not true in the real world. UTF-8 is a minimal
> change for easy conversion in the Unix world, and nobody is going
> to change the low-level tools to recognize the BOM, especially the
> ones that are used for byte-streams as well as text. Stuff that
> supports a dozen different formats may as well support UTF-8 BOMs,
> but a lot of stuff doesn't and won't.
UTF-8 BOM's seem pointless.
Hans Aberg
This archive was generated by hypermail 2.1.5 : Tue Jan 18 2005 - 18:13:12 CST