Re: Entity and Identity

From: paul c <>
Date: Sun, 27 Sep 2009 13:49:03 GMT
Message-ID: <j3Kvm.45451$Db2.3538_at_edtnps83>

Bob Badour wrote:
> What case, Clifford? You are full of shit.
> What specific defect in Nijssen's work absolutely forces you to invent
> yet another stupid fucking marketing term? I can see the utility for
> self-promotion, which is exactly what I expect from a self-aggrandizing
> ignorant, but little else.
> Have you even heard of Nijssen? With all due respect to Terry Halpin,
> did you even look at his bibliography? Do you have even a single clue
> about the nature of Halpin's work?

I stopped reading when it was posted that tuples have location and are "confined" by ACID! It seems many, many people have not been able to separate the many gizmos that encumber relational implementations from the very few concepts of the theory. Although it is hard to separate them, it is essential for coherent discussion. As much as I like writers like Date, I've (very) gradually learned that they have to be read extremely carefully in order to separate concepts of the theory from the various implementation compromises. This remains a constant struggle for me, obviously I'm not alone, and there is way too much quoting and not enough critical thinking in this field. Until they appreciate this, I would say that most people will not profit at all from reading more than two or three "authorities", because most of the writers have the same conceptual confusions as the readers.

Another part of the problem is that most people start to learn the theory effectively handcuffed because so much of the theory is mangled by the typical dbms products. Those products usually have extremely fractional implementations of essential theory such as constraints and the languages suggest that the implementors are oblivious to much else, such as the application of predicates. Received on Sun Sep 27 2009 - 15:49:03 CEST

Original text of this message