Peter,

> > At 06:39 AM 6/15/2004, johnsparrowuk wrote:
> > >According to the quote from Ann (peace be upon her, grin) she also
> > >believes check constraints should operate in the system transaction

I wrote

> > Right. And they don't. And they should. And it won't be
> > trivial to make them work. Damn!

You asked

>Would it be easier to implement DEFERRED constraints?

Maybe, but getting an error sooner rather than later is
a benefit. And we ought to have a mechanism that works.

Regards,


Ann