Re: A question for Mr. Celko

Jan Hidders <jan.hidders_at_REMOVETHIS.pandora.be> wrote in message news:<pan.2004.07.19.22.57.16.709896_at_REMOVETHIS.pandora.be>...
> On Mon, 19 Jul 2004 11:13:53 -0700, John Jacob wrote:
> >
> > In fact, *any* formal system at least as powerful as typographical
> > number theory (which a relational language certainly is, even without
> > relation-valued attributes) is subject to the Godelian attack, it says
> > *nothing whatsoever* about whether or not the system is useful.
>
> Not so fast. The relational algebra corresponds to first-order logic
> without function symbols, which is a strict subset of FOL with function
> symbols, and for that Godel is actually our friend because he has shown
> that we have a complete axiomatization for that, known as Godels
> completeness theorem. Which, by the way, doesn't mean things get suddenly
> easy because FOL, even without function symbols, is still undecidable.
>
> -- Jan Hidders

Agreed, that's why I was careful to say relational language, which, in order to be useful, is extended with scalar computation. This makes it at least as powerful as arithmetic, and therefore subject to the Godelian attack.

Regards,
Bryn Received on Tue Jul 20 2004 - 06:37:35 CEST