At 10:19 AM 5/11/2003 +0200, you wrote:

>At 07:14 PM 05/11/2003 +1100, you wrote:
>
>
> >If you are working with database default character set NONE
> >(which I as one consider a cardinal sin), automatic charset
> >conversion won't kick in, so you must add an "as ... charset ASCII".
> >
> >Regards,
> >Peter Jacobi
>
>Yeah, I would also like to know. I have never used anything else other than
>NONE, and it hasn't ever caused me problems that I am aware of.

Nor I. But it would cause me problems if I wanted to deploy my databases
to countries that need a character set. In South Africa or the UK, you
could choose an ISO character set and localize using collations. In *our*
part of the world, we're more likely to be needing dual-language
English/Japanese, English/Chinese or one or more of the Indo-Chinese
languages. There's a lot-lot-lot more to that than just choosing a
character set for the database, not least because of the severe
limitations of UNICODE_FSS.

Peter's warning is "on the button" for countries that have locales and
keyboards inputting characters that are not in US ASCII. However, for
English-speaking-only deployment, using a character set doesn't necessarily
solve or forestall any problems. Our keyboards don't come standard with
those extra keys for switching input character sets, either.

IMO, it gets down to good old *requirements*.

h.