Kjell Rilbe wrote:

<snip>

>
> We recently had a huge discussion on this subject in conjunction with
> the issue of national characters.
>
> I think the consensus (well...) was:
>
> - For unquoted identifiers, store the names in uppercase but ignore case
> whenever the identifier is used. Also, only allow plain 7 bit ascii.
>
> - For quoted identifiers, store the names as specified. Case will be
> significant when the identifier is used. Also, allow all(?) unicode
> characters in the identifier.
>
> Anything else will cause problems.
>
> Kjell

Is unquoted mixed case likely to make it into the SQL standard any time
soon?

Mike