ON WHAT A R-TABLE IS

From: Randy Ferrer

Date: 3 Mar 2006

I have one question if I may. Joe Celko in SQL FOR SMARTIES Ch.1 claims in a discussion about duplicate rows that Dr. Codd and Mr. Date claim that a table "is a collection of facts." Is this correct? I have scoured my 6th edition of AN INTRODUCTION TO DATABASE SYSTEMS and I can find no such reference. I have not bothered to look in the mentioned reference to Dr. Codd since I have the feeling that Celko's claim is baloney like most of that section. I thought I'd ask you to get your point of view on this. I have gotten into a couple of arguments recently regarding this issue with some fellow database practitioners ... never mind explaining to them that a table and a relation really are different things ... sigh ... The infamous "Can of Cat Food" and Mr. Beech keep coming up. I won't get into that here since you and Mr. Date have beat that one to death already. I would just like to know if Mr. Date ever referred to a table as "a collection of facts" and if so where.

You know, being dedicated to learning my craft to the best of my capabilities, it really becomes increasingly frustrating to have to deal with the lack of fundamental knowledge that is out there. But, you know all that.... ;-) I'm a peaceful guy, but today I actually got called a "db fundamentalist subversive", go figure. :-)) I guess I should be proud of that!

From: Fabian Pascal

You are wise to stay away from Celko. For those without a thorough grasp of the fundamentals and a fair ability to reason, he is an extremely risky proposition.

The correct way to state it is that a R-table (a database table which obeys the discipline that gives it relational properties, namely single-valued cells, no duplicates, no missing values, unordered rows and columns) has a set of rows interpretable as representing in the database a set of propositions that are assumed by convention to be true. True propositions are facts. [For further details see Truly Relational: What It Really Means, and Business Modeling for Database Design].

Of course you should be proud. And, as you can see from my writings, I know only too well the appalling ignorance out there; the site is intended for thinking practitioners, to preserve their sanity, which is extremely hard these days. Unfortunately, ignorance is increasingly characteristic of the Western society as a whole, led by the US, and database problems are an integral part of that, with hardly the worst consequences. As I keep saying, any society that dismisses knowledge and reason does not survive.

Posted 4/28/06