RE: case-sensitivity in xml

> What (in that query string) tells me that its case-unaware?
>   OK, its English, I can infer that.
>    Primary strength?
>    (Sounds like an advert for glue :-)

The notion of the "strength" of a collation is explained in the Unicode
Collation Algorithm. The idea is that the distinction between A and B is a
primary difference, the distinction between Z and z is a secondary
difference, the distinction between e and e-acute is tertiary, and so on.
(There are similar distinctions for Asian languages, but don't ask me to
explain them.) A collation defined with strength=primary considers two
characters to be different only if they have a primary difference; a
secondary difference (case) doesn't count. If you want to ignore accents but
not case, specify strength=secondary.
> 
> Next one, I'd certainly be happy to pick up a collation....
>    Where do they grow please?
> Any references where I might look?

The Unicode Collation Algorithm is a good place. Collation machinery based
on this is now built into a number of software environments, such as Java,
Oracle, the IBM ICU toolkit, the Windows platform, and so on. Note however
that the XSLT/XPath specs don't restrict the choice of collations to those
that conform to the UCA.

Micha