Re: some information about anchor modeling

> You are now saying that something is wrong because it is wrong. The
>
> concept of the surrogate is not nonsense, you simply do not understand
>
> it!
 

In this thread I showed why the surrogate key is a bad solution. See my posts from 18 July and 12 September in this thread. There are many other cases, which show that the surrogate key is a bad solution. The cases that I have indicated are very serious. As far as I know, this is the first time that someone has explained why the surrogate key is a bad solution. This explanation is supported with very important examples.

Obviously, RM / T can not support nulls. A person who enters the data has to know all the data of the corresponding entity. So, someone could set the following question: why RM / T uses binary relations, when data entry person has to know the entire entity.

Logical operations that Codd introduced for nulls do not work and do not help.

The surrogate keys cannot be fixed.

There are other serious problems in the RM / T that are not due to surrogate keys. What is much worse, it's that the RM / T has a serious misunderstanding on the theoretical level. There are also serous mistakes on the theoretical level. Let me show you just three of them.

(i) Kurt Gödel 1944:
“By the theory of simple types I mean the doctrine which says that the objects of thought (or, in another interpretation, the symbolic expressions) are divided into types, namely: individuals, properties of individuals, relations between individuals, properties of such relations, etc. (with a similar hierarchy for extensions), and that sentences of the form: " a has the property φ ", " b bears the relation R to c ", etc. are meaningless, if a, b, c, R, φ are not of types fitting together. Mixed types (such as classes containing individuals and classes as elements) and therefore also transfinite types (such as the class of all classes of finite types) are excluded…”

Note that Kurt Gödel uses the term "individuals" rather than “entities”. 35 years later, E. Codd extensively uses the term "entity type" in his work RM / T. What is wrong here?
This “job” is wrong, because no one knows how Codd transmits data from one data model to another data model, using the entity type (from E/R data model to RM data model and vice verse).

Please note that I am completely solved this problem. See my paper “Database design and data model founded on concept and knowledge constructs” from 2008 at http://www.dbdesign11.com , section 1, 4.1, 4.2.7 and 6.4. The mapping between data models for simple databases is defined by identifiers of entities (relationships). The mapping for complex databases is defined by identifiers of states of entities (relationships).

(ii) I distinguish real objects from abstract objects. I defined abstract objects and the identification of each of these objects. See (3.3.3), section 3.3, from the above mentioned paper “Database design and data model founded on concept and knowledge constructs”. In my paper only the "m-attributes" are determined with our perceptual abilities. All other (more complex) objects are defined recursively, according to their complexity (see m-entities, m-relationships and m-states). The complex objects are determined by our mental activities.

We can notice that E. Codd not distinguish these objects. He did not even notice these objects.

On the other hand, we can notice that (3.3.3) is important because it defines the relationship between concepts and identification, that is, it determines the relationship between the relation of satisfy and the corresponding identification.

(iii) E. Codd did not prove the decomposition of a relation into binary relations, that is suggested in his paper RM / T.

Note that it is obvious that the binary relation must have one attribute and the simple key. Obliviously, E. Codd was aware that there was no point in having a binary relation that has a composite key and one attribute. So, he introduced the “surrogate key”.

Many people tried to solve the problem of "binary decomposition". This is perhaps the most important problem in the database theory.

--

In his book "Go Faster! - The TransRelational ™ Approach to DBMS Implementation" C. Date published a letter which was written by E. Codd. This letter E. Codd wrote to Steve Tarin. C. Date represents Steve Tarin as the author of The TransRelational™ Model.  
(This book is free downloaded at bookboon.com). In this letter E. Codd expressed congratulations to S. Tarin for his “revolutionize” work and that this technology can be “extremely effective”. It seems to me that with this letter E. Codd admitted that did not resolve the problem of the mentioned "binary decomposition".

But in his last book "Database Design & Relational Theory – Normal Forms & All That Jazz” C.J. Date writes about “anchor relvars” and “(Does this state of affairs remind you of the RM/T discipline discussed in Chapter 15?) “
I have an impression that this claim about “RM / T discipline” stands in contradiction with the obvious E. Codd assertion that The TransRelational ™ Model superiorly solved the mentioned problems.
 
 


> > I am writing about this, because I have impression that there are people

> 

> > who try to "fix" RM/T paper, by using works which are done by others.

> 

> 

> 

> And you object to this? Why?


 
The following sentence: “An entity is some thing in the modelled universe and is typically identified by a surrogate.” is not true, because Codd’s surrogate does not identify an entity.  (See  http://en.wikipedia.org/wiki/Relational_Model/Tasmania )

In many cases, the surrogate introduced by E. Codd, cannot identify anything. In fact, Codd’s surrogate is serious nonsense. See examples in my posts from 18 July and 12 September in this thread.

Given that the mentioned sentence is posted on Wikipedia, and considering that Wikipedia has a global approach, and considering that a lot of people use Wikipedia as a reference material, I found that it is important to inform the public about these inaccuracies.
Of course, if you think that my examples are not correct, please post it. But please be specific. So, please specify a concrete my example which is not correct, and give an explanation. Given that your comments were not specific, I can not accept them seriously.
 
 


> Eric

> 

> -- 

> 

> ms fnd in a lbry



Vladimir Odrljin