Re: The anatomy of plagiarism that was made by authors of "Anchor Modeling"
Date: Wed, 27 May 2015 13:14:04 -0700 (PDT)
Message-ID: <a59088b7-4ea2-4756-a85b-7ad316a8ee3d_at_googlegroups.com>
With this post I will finish this thread. Of course, if someone wishes to
comment or criticize my arguments in this thread, he is welcome. This is the
user group for the free expression of thinking in relation to the theory of
databases.
In this post I will present serious errors and plagiarism that are related
to "sixth normal form" at the level of db design in award-winning paper on
ER09.
I will show, also that "decomposition" into atomic structures can not be
done by using "6NF".
In "Anchor Modeling" many important cases have been omitted. Here are some examples:
- "Metadata"
Basic data-structure in "Anchor Modeling" is Hatt (C, D, T). But this
structure is presented with only one "metadata", it is "T". Note that
"metadata" often have 6, 7 members. The regular four members are "(DateFrom,
DateTo)" for the real world date, and "(DateFrom, DateTo)" for the "database date". In my posts from 23 April and from 11 May in this thread, I showed that
"Anchor
Modeling" can not correctly work time intervals, and that is why they have only one "T" in the Hatt (C, D, T). Thus, the structure Hatt (C, D, T) is tuned according to "6NF". Of course, this kind of technique does not belong to a scientific domain. - Relationships
Relationships in "Anchor Modeling" do not have attributes, although P. Chen
presented relationships with attributes. It is obvious that "6NF" does not
work for relationships that have "metadata". Note that there are
relationships between relationships.
In this thread is presented that "6NF" can not be used for general databases, and therefore neither for "maintenance history". 6NF does not work well in the "Anchor Modeling" for the following relationship's cases: a) 6NF does not work well with relationships in Anchor Modeling. b) 6NF does not work well with relationships with "metadata" in Anchor Modeling.
c) 6NF does not work good in Anchor Modeling with history of intreval's data.
Note that these are the most difficult cases in the database theory.
3. Decomposition into atomic structures
This "decomposition" into atomic structures can not be done by using "6NF".
Let me mention just a few reasons:
(i) General db theory is opposite from existing "RM" (RM is a short cut for
Relational Model). "6NF" is based on"RM". In "RM" there is no redundancy,
General db keeps redundancy.
(ii) Authors of "6NF" use delete and update operations. General db does not
use these operations
(iii) Current "RM" is not entirely based on G. Frege's theory.
(iv) "6NF" does not define a procedure, which leads relations into "6NF".
Now the following text "using sixth normal form" which the authors of
"Anchor Modeling" put in the title of their work, does not make sense.
So far, the decomposition into atomic structures is done, in only one way. This was done by using my theory about states. It has been published in 2005 year on my website and presented on this user group.
In their paper "Analysis of normal forms for anchor-tables" authors of
"Anchor Modeling" prove that their main structure Hatt (C, D, T) is in "6NF"
???
Here C is the anchor-surrogate key, D is an attribute, and T is time. Note
that here the authors of "Anchor Modeling" have only one "metadata," it is
time "T".
Note that here the issue is not "6NF" at all. The point is the introduction
of a completely new type of relations - it is a relation that maintain
history. Such kind of relations do not exist in "RM". In 2005, I defined the
new General db theory. Hatt (C, D, T) is the plagiarism of the main ideas of
my General db theory. As I showed in this thread, "6NF" does not solve
history. "6NF" is just a cover for this plagiarism.
In my thread "The original version", in August 2010, I started writing that
that authors of "Anchor Modeling" are working in "Anchor model" and that
"6NF" is in "RM". I also wrote that the problem of mapping between data
models is solved only in my paper and in my data model.
The authors of "Anchor Modeling", in September 2010, announced on their private website, a series of papers in which they corrected this huge mistake. In November 2010, their second work was presented in DKE, editor Peter Chen. In this paper they plagiarized the main part of my data model - that's the theory about states. In this way, they corrected the mistakes of the mapping between two data models, proclaim themselves as the authors of the solution for mapping between data models.
Most importantly they plagiarized my theory about the states. In my post of March 17, 2015 in this thread, I published the first time some of the solutions in Logic, semantics, model theory and some other areas, which in my opinion can be directly solved by applying my theory of states of objects and relationships.
Vladimir Odrljin Received on Wed May 27 2015 - 22:14:04 CEST