Oracle FAQ Your Portal to the Oracle Knowledge Grid

Home -> Community -> Usenet -> comp.databases.theory -> Re: Pizza Example

Re: Pizza Example

From: Dawn M. Wolthuis <>
Date: Wed, 7 Apr 2004 12:21:02 -0500
Message-ID: <c51das$u$>

"Tony" <> wrote in message
> "Eric Kaun" <> wrote in message
> > "Anthony W. Youngman" <> wrote in message
> > news:j4m$
> > > The problem I see with relational, is that it does not cope with -
> > > indeed, denies the existence of - emergent complexity.
> >
> > I'll address the specific example below, but what does this mean in
> I have learnt from past experience that Wol likes to throw in
> irrelevant but important-sounding scientific concepts from time to
> time. Heisenberg's Uncertainty Priciple comes up quite often, along
> with Godel's Incompleteness Theorem and of course Relativity v.
> Newton! The relational model is deemed wanting because it fails to
> take these into account, though it is far from clear how MV or
> anything else does other, nor why they should.
> It's best to try to ignore these digressions, but I usually fail to do
> so...

I tend to agree, although I have learned that Wol has a point even if obscured by his seeming lack of interest in or respect for mathematical models. I agree with him wholeheartedly that a model is a model -- just that, no more. Pointing out both flaws and usefulness in particular models is helpful. Thinking all mathematical models are hogwash is not, however. I don't need to emperically determine that 1000 + 1000 = 2000 because I have a model, a mathematical theory, that starts with some axioms and builds on that and that convinces me, without any emperical data, that if I have 1000 objects and add in another 1000 objects, then I will have 2000.

So in spite of the shared appreciation that I have for the underdocumented PICK "model" (loose use of the term in this case) along with Wol, he and I have had this mathematics vs science discussion before in other forums. If we have a mathematical model that is useful AND we have some emperical data to back up the usefulness of the model, then that would be great. I think "language" belongs in there somewhere too, but not sure exactly where.

I don't know of any data approaches that have both a solid mathematical model (and no, it will not be both complete and consistent) and emperical data that gives evidence of the usefulness of the model.

But if anyone else does, I'm interested. --dawn Received on Wed Apr 07 2004 - 12:21:02 CDT

Original text of this message