"Jim Starkey" <jas@...> wrote in message
news:42289F86.5030304@......

>
>>I'm afraid no. What about case-insensitive, accent-insensitive, etc
>>comparisons which are all language-specific?
>>
> Of course they're language specific. But the the collations can be
> written to operator on UTF-8 as well (or better) than a language
> specific character set.

Never mind. Actually I'd better put the respone a sentense up. It's obvious
collations have nothing to do with character sets.

> I think the question of case-insensitivity is an interesting case. In
> general, collation is only necessary for a high/low test -- sorting and
> range retrievals, while equality can be based on raw byte comparison.
> To support case insensitivity, collation is required to determine even
> equality.
>
> Are there other collations where equality is collation specific?

He-he, see CompareString in MSDN.

NORM_IGNORECASE
NORM_IGNOREKANATYPE
NORM_IGNORENONSPACE
NORM_IGNORESYMBOLS
NORM_IGNOREWIDTH

And I suspect Unicode defines more.

Regards,
Aleksey Karyakin