Re: The term "theory" as in "database theory"

dawn wrote:

>
> Agreed. This gets at one of my reasons for asking. I have read Date,
> Pascal and others refer to Occam's razor related to relational theory.
> For example, see http://www.dbazine.com/ofinterest/oi-articles/pascal12
> and search the page for Occam. Although Occam's razor cannot be
> proven, it is in the belief system of many. So, let's take it as a
> given (even if it isn't). Doesn't Occam's razor refer to the above two
> meanings of "theory" rather than to anything related to "database
> theory"?
>
> Maybe this has all been dispelled before, but I heard it again recently
> and really want to put this one to rest, if feasible. Are people in
> this forum at least agreed that Occam's razor has nothing to do with
> relational theory?
>

A long time ago I "made peace" with the fact that most of our precious theories fall in the category of Euclid's fifth axiom -- it's only true if you believe it. I stress as well that this is different from the standard post-modern nonsense that your opinion is as true as mine. What I have observed simply is that people grasp these bodies of theory, hold them to be laws of nature, and generally write some pretty good systems.

I can "prove" my db-centric theory works by implementing it in a working system that meets its goals. Another programmer "proves" that ORM works by implementing it in a working system that meets its goals.

All we've proven is that it is a matter of human taste and judgement.

Which, BTW, I believe your preference for MVL is just such a thing, a taste or preference. If no MVL system existed you would still have a CV of successful projects, and if all of us were forced to use it we would all have a CV of successful projects. This idea that you can "pick a religion" is extremely distasteful to some theorists.

-- 
Kenneth Downs
Secure Data Software, Inc.
(Ken)nneth_at_(Sec)ure(Dat)a(.com)