Hi Nando,

>> as Martijn discovered (much to my surprise and joy) there are
>> collations for the ISO8859_1 charset that are case insensitive. I
>> checked du_nl, en_uk, en_us and de_de with positive results.
>
> not completely, as it seems, since Martijn reported they aren't
> working in the where clause

Indeed they aren't.

> (which is their main field of application I suppose)

Not for me, or rather: not for my customers, in most situations. They
want to be able to sort table columns alphabetically as opposed to
asciibetically. And they don't want to find all the Turkish names
starting with O-umlaut below the Z when they expect them among the
regular O's.

For case-insensitive comparisons I use UPPER(). I must admit however
that my databases don't contain millions of records; if they would,
the performance hit might be painful...

Greetings,
Paul Vinkenoog