> Otherwise, pick a character set that matches the locale settings

that your

> users will use for input, that is valid for the platform the server

is

> running on: possibly ISO8859_1, with the ES_ES collation sequence

on

> columns that need to be sorted? Or WIN1252 with the PXW_SPAN

collation

> sequence?

>
> What happens with NONE is that it will accept *input* of any

character but

> it will raise transliteration errors when it finds "wrong"

characters for

> operations (search matches, selects, etc.)
>
> heLen

Just did some tests and confirmed this, charset NONE raised
transliteration errors on my "ñ" characters, the only special
character I'm concerned about. Tried charset ISO8859_1 with ES_ES
collation on some data and I believe this is just what I need.

This surely made me understand charsets and collation better.
Thanks a lot, Helen, for the help as always. Can't wait for your
upcoming book :)


Regards,
Jane