Re: Pizza Example

"Chris Hoess" <choess_at_stwing.upenn.edu> wrote in message news:slrnc7umpn.l4q.choess_at_force.stwing.upenn.edu...
> In article <TVoj+7HXgvfAFwwt_at_thewolery.demon.co.uk>, Anthony W. Youngman
wrote:
> >
> > (And I'm not against maths. It's just that science is all about fuzzy
> > logic, uncertainty, and statistics. But (a) my grasp of the fundamental
> > theory behind that is very poor, and (b) it seems to be a lot better
> > than anybody else who might be inclined to debate with me. I'm just
> > looking for *scientific* proof, not mathematical, that relational theory
> > works, and nobody seems willing to try to prove it.)
>
> I'm not sure asking for "scientific" proof of something that's not a
> scientific theory makes sense.

If relational theory is just for the purpose of playing with mathematical models, then there is nothing scientific to be tested. But if the idea is that using this theory in the development and ongoing support of software applications will make such software better quality, less expensive to build or maintain, last longer, or whatever else, then there is a hypothesis that we ought to be able to test in some way.

> Newtownian mechanics offers rules by which we
> can describe the world. The relational model offers rules for the
> manipulation of propositions, whose mapping to the real world is arbitrary
> (but should be consistent for each individual database). I suppose I might
> try to put the premises of the relational model thus:
>
> Axiomatic statements about the world form the conditions of the database.
> Data consists of statements which are evaluated as logical propositions
> against the constraints.
> Queries against the database consist of the application of additional
> constraints (axioms) to the propositions in the database, and return all
> logical propositions which are true in the new axiomatic system.
>
> So statements in a relational database are guaranteed to be true to the
> degree that truth can be defined by the constraints expressed in the
> database.
>
> As I see it, the main two lines of "attack" are the idea that all data can
> be represented as statements (certainly the world is fuzzy, but how
> better to *automatically* evaluate truth than logic?) and the expressive
> capabilities of the relational model as regards conditions (but here
perhaps
> we run into Russell's Paradox).

Any "attacks" that I make are based completely on the usefulness or lack thereof in applying the relational model to software development or in claiming that it is the best or even only approach for doing xyz. --dawn

> I am neither by training nor practice a logician, so I'm not entirely sure
> this description is correct, but I think it's a good starting point for
> discussion of the issue.
>
> --
> Chris Hoess
Received on Fri Apr 16 2004 - 06:38:26 CEST