## Not signed in

Want to take part in these discussions? Sign in if you have an account, or apply for one below

## Discussion Tag Cloud

Vanilla 1.1.10 is a product of Lussumo. More Information: Documentation, Community Support.

• CommentRowNumber1.
• CommentAuthordarij grinberg
• CommentTimeOct 20th 2012
• (edited Oct 22nd 2012)

To those who don’t have access to this through springerlink, there is literally too much nonsense to paste in here. The abstract gives a good impression of what is to follow. The text throws around (often defective) definitions and examples of notions like finite fields, wave equations, differential posets, polytopes, Young’s lattice, Hopf algebras (the author apparently can’t tell coalgebras in the algebraic sense apart from computer-science coalgebras; also, the definition of Hopf algebras is missing most of its axioms). Here is probably the most bullshit-intensive paragraph:

For the self-evolvable systems the concept of dual Hopf algebra is of interest. Algebra and coalgebra, integration and differentiation are dual concepts. Fig. 3.6 shows the polytope including algebra and dual algebra. Fig. 3.6 suggests that after the integration or algebraic way S→K1→K2→K3 we need to look at the differentiation or dual algebraic way K3′→K2′→K1′→S′.
Making use of the developments of the direct way may offer in a kind of symmetry-breaking results. This opens the road for dual Hopf algebras interpretation (Hivert et al. 2005, Nzeutchap 2006). If the two ways offer the same results that is, in the case of self-duality the described system may be evolvable but not self-evolvable.

The swinging from algebra to dual algebra is critical since the boundaries where creative research grows and new information is created consist of parallel tendencies of integration and differentiation. The Self describes the interaction of the two algebras in duality relation.

Swinging method based on dual algebras has been applied in model evaluation and software correcting (Padawitz 2000, Jacob and Rutten 1997).

The references in this paragraph are legit but, of course, unrelated to the “context” (and to each other).

The “article” is by some Octavian Iordache who loves self-citation:

Iordache, O.: Evolvable Designs of Experiments Applications for Circuits. J Wiley VCH, Weinheim (2009)

Iordache, O.: Polystochastic Models for Complexity. Springer, Heidelberg (2010)

Iordache, O.: Modeling Multi-Level Systems. Springer, Heidelberg (2011)

Iordache, O., Corriou, J.P., Garrido-Sanchez, L., Fonteix, C., Tondeur, D.: Neural network frames applied for biochemical kinetic diagnosis. Comp. Chem. Engng. 17, 1101–1113 (1993a)

I don’t have the time to sift through these “sources” (particularly as I lack the skills to give a definite judgment on most of the non-combinatorics stuff), but at a first glance the books published at Springer appear to be of the same sort as the one quoted above, whereas the Wiley book at least visually creates a better impression.

I also don’t know whether it’s [url=http://www.polystochastic.com/pages/publications.html]this Octavian Iordache[/url] (notice the lack of these spam texts in his list of publications?) or a false-flag by some competition. Though I must say I have never seen the latter in practice so far.

Does anyone have some spare time to probe into this a bit further? I am hesitating to email anyone so far because these kinds of discussions tend to claim a lot of time at the worst possible moments; probably it would be better to collect some more evidence first, anyway.

• CommentRowNumber2.
• CommentAuthordarij grinberg
• CommentTimeOct 21st 2012
• (edited Oct 21st 2012)

Half an hour of time wasted perusing the Wiley 2009 book has led me to believe that it is the same kind of eyewash as the Springer materials, except somewhat better cloaked (result of “better” editing practices on the publisher’s side or of increased disconnect from reality on the author’s side as time went by?). Some of you might be able to view the whole “work” through Wiley:

http://onlinelibrary.wiley.com/book/10.1002/9783527624027

To me, Chapter 4 is the place where the hogwash is most obvious, with its consideration of a “wave equation” over finite fields (despite no mention of polynomials, power series or any other type constraint for its solutions). While I cannot claim this with any guarantee, I highly suspect that the rest of the text is of a similar quality. Here is a pearl from page 178:

There are some symmetry features in genetic code. These symmetries help in explaining regularities and periodicities as observed in proteins. The relevant algebraic group to describe the symmetries of the bases{C, G, U, A} should be a group of order4. There are only two possibilities for group structure, the cyclic group C(4) and the Galois group associated with the Galois field GF(4) (Findley, Findley and McGlynn, 1982). The genetic code may result as a product of C(4) or GF(4) groups. The product construction appears as a method to obtain multilevel solutions of the wave equation.

Forget Frobenius; the Galois group of GF(4) is now officially the Klein four-group!

Congratulations, Wiley! Keep it up and you won’t have anyone torrenting your books, as noone will want them anymore.

• CommentRowNumber3.
• CommentAuthorzskoda
• CommentTimeOct 22nd 2012

Indexed by Ulrich´s, SCOPUS, MetaPress and Springerlink.

Understanding Complex Systems

Series Editors: Abarbanel, H.D.I., Braha, D., Érdi, P., Friston, K.J., Haken, H., Jirsa, V., Kacprzyk, J., Kaneko, K., Kirkilionis, M., Kurths, J., Nowak, A., Reichl, L.E., Schuster, P., Schweitzer, F., Sornette, D., Thurner, S.

Founded by: Kelso, J.A. Scott

I do not understand such carelessness of such a distiguished editor board to allow Polystochastic Models in Chemical Engineering [Hardcover] Octavian Iordache (Author), O. Iordache (Editor) ever to appear in their series. There is some compilation work which occasionally has correct paragraphs, but there is also lots of total speculation and systematics of author’s own mixture of everything, compatible or incompatible. Look at this typical example of superficial compilation (on the level of a student who barely attended an intro course once per 3-4 lectures)

The efficiency of category theory lies in the possibility of universal constructions as for instance limits, and colimits. The colimit is a formalization of assembly of objects and morphisms. A colimit for a diagram can be thought of as a structure that completes the diagram to a minimal commutative diagram containing it. The colimit puts everything together. It describes gluing or fusion. The tool for describing putting them together is called a cocone. In the category Set the colimit corresponds to the least set. Limits are the dual notion to colimits, which is the one notion obtained from the other by reversing the arrows and interchanging initial and terminal for objects. Intuitively a limit extracts the abstraction. Given a diagram, an element is called a limit if there are morphisms from that element to all vertices of the diagram, and if for any other element satisfying the same property there is a unique morphism from it to the limit. In the category Set the limit corresponds to the biggest set.

which continues with “original” speculation

Limit can be seen as an emergent concept summing up in itself the properties of its constituents. This allows considering a hierarchy where at any level the objects are the limits of objects of the lower level. This is consistent with the opinion that complexity is a relative notion depending on the level of observation. The tool to obtain limits is called a cone.

The author for example does not see that limit is itself a cone, the universal one. For him, the cone is some sort of a tool. He compares the completion hierarchy as the complexity hierarchy what is like comparing height of a building with longevity of marriage; the metaphore of a hi-school essay.

Then he goes on compiling specific facts and gets them entirely wrong

A Cartesian closed category is one which is closed under all kinds of universal constructions for example limits, and colimits.

He should look in $n$Lab to learn closed categories and notice that closed is not the same as complete category which is simulatenously the cocomplete category, the combination which means closedness under small limits and colimits. What is the meaning of closed under “all kinds of universal constructions” beyond those two it is hard to say, as there are too many generalizations to be all there admitted.

Then another untrue statement

To any canonical construction from one type of structures to another, an adjunction between the associated categories, will corresponds.

This was from an appendix which is a compilation. The “original part” has statements of hardly digestable speculation

For PSM framework the conditions K represent the category describing the types of component processes. In this case, the processes types are the objects of category. Interactions among types can be modeled as morphisms. Categorical constructions such as colimits characterize fusion. S is the category, associated to the detailed states of the component processes. The arrows that is, the morphisms describe the transition relations between the states of the component processes.

Where is the composition of morphisms defined in this typical rambling passage ? Mac Lane is certainly turning around in his grave…

Or the following

In category theory, four levels are required to define morphism as unique up to isomorphism. The four levels are the objects, that is, the elements within a category, the category comparing the objects, the functors comparing categories and the natural transformation comparing the functors.

This guy does not understand anything in category theory, but rambles it into a soup of everything to publish in noticable series like this.

• CommentRowNumber4.
• CommentAuthordarij grinberg
• CommentTimeOct 22nd 2012

Did anyone notice something missing from my first two posts?

Of course: Enter Elsevier!

Octavian Iordache, Theoretical frames for smart structures, http://www.sciencedirect.com/science/article/pii/S0928493196001531

I’m not going to fisk this in detail, but let me just make some remarks on Section 4.1. He is trying to model the process of “classification and pattern recognition” by a function y (the answer function) in the variable b (the “degree of classification”, which seems to be something like a time, or progress, variable for our classification process, as far as the text can be deciphered). Initially b takes real values (as I would expect from a time/progress variable), but these magically become vectors over GF(2) by means of base-2 expansion, and the “classification process” is modelled by a differential equation over GF(2). Not a single word is spent justifying this completely chaotic model of machine learning; there is only one paragraph suggesting that the author has thought about it:

The rule of addition \oplus [note: this is how the author denotes addition over GF(2)] , signifies that two identical steps have only contrary and globally annihilating effects. The addition appears rather as a comparison of two degrees of classification.

Exactly. ( http://lolpics.se/pics/16581.jpg )

What can be learned from this story so far? None of three famous “scientific” publishers consistently follows editorial standards. Surely, each of Springer, Wiley and Elsevier does have select journals which do referee and care for the quality of their papers, but on the low end each of them has outlets for random unrefereed nonsense which fly in the face of their own self-conception (Springer: “we provide scientific and professional communities with superior specialist information”; Elsevier: “Elsevier publishes trusted, leading-edge Scientific, Technical and Medical (STM) information”; Wiley: “Offering quality, variety, and value to serve all learners”). The publisher’s name is no longer of any value as a brand on a scientific paper; only along with the name of the journal it provides any measure of quality. (Often, the presence of a preprint on arXiv is better evidence for quality and honesty, though.)

@zskoda: Yes, the comparison to a schoolkid writing his homework essay also appeared to me. This is very much the style in which somebody writes who is scared of accidentally showing his lack of knowledge with a too precise and thus refutable statement. Still, Iordache’s books have a sufficient amount of refutable statements in them, as there is only so far one can get through imprecision in mathematics…

• CommentRowNumber5.
• CommentAuthorJohn Baez
• CommentTimeOct 23rd 2012

Octavian Iordache is a rather famous math crackpot.