Re: Is relational theory irrelevant?

"Serge Rielau" <srielau_at_ca.eye-bee-m.com> wrote in message news:bpgef8$gp6$1_at_hanover.torolab.ibm.com...
> Mikito Harakiri wrote:
>
> > "Serge Rielau" <srielau_at_ca.eye-bee-m.com> wrote in message
> > news:bpfunr$dic$1_at_hanover.torolab.ibm.com...
> >
> >>The syntax is as follows:
> >>SELECT .. FROM OLD TABLE(DELETE FROM ...)
> >>
> >>OLD TABLE being the set of deleted rows
> >
> >
> > May I ask you how is your single statement it different from the
following
> > sequence:
> >
> > SELECT * FROM T;
> > DELETE FROM T;
> Depending on your isolation level it will give you different results or
> deadlocks.

Agreed, and to mention the point again, if you want to get rid of logical transactions (and so make the relational model a decent model of reality - you can't back out in real life, the arrow of time moves forward only), then you need this stuff. That academics might not see it, is due to the overly narrow scope that they allow the relational model (such as claiming that transactions are orthogonal to the model. sesh!)

Regards
Paul Vernon
Business Intelligence, IBM Global Services Received on Wed Nov 19 2003 - 20:42:27 CET