Re: Towards a definition of atomic

From: Jan Hidders <hidders_at_gmail.com>
Date: Tue, 5 Feb 2008 02:54:00 -0800 (PST)
Message-ID: <77e2d3da-29fb-4e55-9d89-b2f8b37fb0bc_at_e6g2000prf.googlegroups.com>


On 5 feb, 02:38, David BL <davi..._at_iinet.net.au> wrote:
> On Feb 4, 11:29 pm, Jan Hidders <hidd..._at_gmail.com> wrote:
>
>
>
> > On 4 feb, 03:09, David BL <davi..._at_iinet.net.au> wrote:
>
> > > What do you think of the idea to use existence of a bijection as a
> > > definition of information equivalence between databases (expressed
> > > with alternative DB schema)?
>
> > It depends a bit on what intuitive concept of equivalence you want to
> > formalize. Under this definition, for example, all schemas that have
> > countably infinite many instances are equivalent. That is probably too
> > crude.
>
> Way too crude!   Yes, I can see the bijection needs to be constrained
> somehow.
>
> Presumably the bijection exists if and only if the set of instances
> for each schema have the same cardinality.   I gather that is in fact
> the definition of equal cardinality.

Yes, it is. An important restriction that is often added is that of genericity as defined in the context of database query and transformation language research. It is probably the single most important concept for understanding that theory, but something that people often have difficulty getting their head around.

  • Jan Hidders
Received on Tue Feb 05 2008 - 11:54:00 CET

Original text of this message