Re: database systems and organizational intelligence

From: Alan <alan_at_erols.com>
Date: Wed, 26 May 2004 16:30:00 -0400
Message-ID: <2hkd28Fe0ou7U1_at_uni-berlin.de>


See inline =====>>>>>

"Dawn M. Wolthuis" <dwolt_at_tincat-group.com> wrote in message news:c92r6d$23i$1_at_news.netins.net...
> "Alan" <alan_at_erols.com> wrote in message
> news:2hk09hFdml8oU1_at_uni-berlin.de...
> > Your comment below is a joke, I hope.
> >
> > A functional dependency is a property of the SEMANTICS or MEANING of the
> > attributes. It occurs at the "business" level. It is something that
> business
> > people can tell you. Like (a not perfect example, but easy to
understand):
> > Each person in our company has a Social Security Number, and each Social
> > Security Number (SSN) identifies one and only one person. So, in the
> > miniworld of our company, we know that the value of an employee's SSN
> > uniquely determines the employee's name. Before any entities or tables
> are
> > created, before any code is written. A functional dependency is a
> constraint
> > between two sets of attributes within the miniworld of the database (our
> > company).
> >
> > "Functional" in this context means "unique". It has nothing to do with
> > functions as you are thinking of them.
>
> When you write down a functional dependency with the little arrow in it,
> just what does that arrow mean? I was thinking it was a mapping from one
to
> the other. And what's another word for a mapping? Function, perhaps?
I've
> never known "functional" to mean "unique" -- is that your own personal
> definition or is there some industry use of the term in this way?

=====>>>>> The "-->" It means "uniquely determines," which is described above. Or, by example:

ssn --> emp_name

>
> > It has zero, nothing, nada, zip to do with code. Nothing at all.
Nothing.
>
> It's all code.
> It's all data.

=====>>>>> Code can be data (when stored in a database, such as the way Oracle stores user-defined stored procedures, but data is not code. Data can be used in code, either as a variable or a constant.

>
> Cheers! --dawn
>
>
Received on Wed May 26 2004 - 22:30:00 CEST

Original text of this message