Re: godel-like incompleteness of relational model

From: Paul <paul_at_test.com>
Date: Thu, 20 May 2004 10:37:17 +0100
Message-ID: <6%_qc.6458$NK4.656343@stones.force9.net>

mountain man wrote:

>>>>And what is the problem with The Relational Model?
>>>
>>>It has a Godel-like incompleteness:

>
> http://www.mountainman.com.au/software/history/relational_model_incomplete.htm

I don't quite understand what you mean here. Even if you think that relational theory is missing something, I don't think it is a "Godel-like" incompleteness.

>>I'm no mathematician, but didn't Godel prove that 'any' formal system
>>is incomplete?

>
> Yes, he did. But I am being specific about provision of one specific
> instance in which the incompleness of the RM is comprehendable.

Well, Godel acually proved that first-order predicate logic (upon which the relational model is based) is complete in some sense. The Incompleteness theorem only applies to theories that are above a certain complexity. To add to the confusion, there are slightly different meanings of "complete" here. See this page for more details: http://www.sm.luth.se/~torkel/eget/godel/completeness.html

I think essentially the difference is that you need to use logic to show that some other theories are incomplete, but to show the completeness of   logic itself you've got a bit of a self-referential paradox. I could be completely wrong here though. Very interesting though.

Paul. Received on Thu May 20 2004 - 04:37:17 CDT