Re: RM's Canonical database

From: Marshall <marshall.spight_at_gmail.com>
Date: 3 Jul 2006 10:55:05 -0700
Message-ID: <1151949304.991749.122630_at_b68g2000cwa.googlegroups.com>


Dan wrote:

>

> Sure. I will accept that. I recognize it as a valid definition, but
> not as the only definition.

Definitions are funny things; anyone can make one. For example, within the context of this post only, I will use the following definitions:

data: a starbuck's coffee
information: small tasks done throughout the day; errands

We can see from this that definitions are much like name binding in programming languages. They have a scope and a lifetime, can easily be created by anyone anywhere, and refer to whatever the author wants them to refer to. For them to be useful, it is only necessary to get two modules or two people to agree on them. Furthermore, it is possible to have an authoritative name or definition when one registers with an authorizing body, such as when one creates the Customers table in a dbms or has a standard definition endorsed by a standards body.

Anyway, I've finished my morning's data, and now I've got to go run some information.

Marshall

PS. I still think the ISO definitions are the best ones. Received on Mon Jul 03 2006 - 19:55:05 CEST

Original text of this message