Juan Chamero

Metodología Darwin

  • Aumentar el tamaño de la letra
  • Tamaño de letra predeterminado
  • Reducir el tamaño de la letra

Breadcrumbs

Home Darwin versus W3C Visions - I

Darwin versus W3C Visions - I

Correo electrónico Imprimir PDF

Web Semantic

Going from Mind to Digital or from Digital to Mind?

W3C versus Darwin, two polar approaches that could reinforce each other

Dr. Juan Chamero, Esta dirección electrónica esta protegida contra spambots. Es necesario activar Javascript para visualizarla , Buenos Aires, Argentina February 15th 2009

 

 

The Darwin Vision


Art Tree 

 

 


   
In the figure above is depicted a series of “
semantic paths” highlighted in an ART Map unveiled from the “Web as_it_is” at a given moment by Darwin agents working within its particular
Darwin ontology. This map extends from “root” to “leaves” throughout 13 “semantic levels”. Rigoletto as name of the well known Italian opera is linked to the ART root by a “semantic path” of 11 levels [0.1.2.2.2.2.14.1.6.10.5] meaning the semantic hierarchic embedding:

 

ART => The Art => Performing Arts => main Performing Arts => Theater => genres => Opera => Opera History => Italian Opera => Bel Canto Movement => Rigoletto

 

The paths depicted belong to a prototype built on August 2008. A sample of it may be downloaded from Intag, the Darwin proprietary Website. Once a discipline like ART is mapped all Web documents somehow dealing with ART could be tentatively cataloged over this semantic skeleton. And we say “tentatively” because maps unveiled may evolve along the time differentially in both its content and its structure.

 

"Give to Caesar what is Caesar's, and to God what is God's"

    The ambitious W3C project also known as the World Wide Web Consortium is now showing signs of weakness and tiredness. However we believe that its findings are all valid and with a promising future to cope efficiently with the future Web needs. It was conceived by Tim Berners Lee  the Web father. The best universities, labs, and people worked hard since the beginning of the 90´s under the W3C big umbrella creating all type of abstracts tools and languages to implement TBL central idea: The Web Semantics, its personal and revolutionary vision, an information space to be used by machines rather than by humans. Instead of processing and manipulating Web information, users are encouraged to send agents to catch specific information, eventually forms of intelligence and to solve problems. To conceive its logical skeleton he uses the OSI Seven Layers Model, successfully implemented in the Internet beginnings. See Appendix below. 

 

The imperfect human semantic paradox

    Supposing that all information generated by humans or mechanically generated under their control is readable, fully “understood” and cataloged by machines all W3C languages and tools would work as a Swiss watch soon, let´s say in no more than a decade. The problem is that this information is actually huge, too much unstructured, fuzzy and noisy and perhaps in large extent these last apparent drawbacks are paradoxically necessary!. Concerning this paradox we advise to see our “Semantic Search” White Paper where we discuss about ideas, ideals and concepts, as a philosophical approach to build digital ontologies.

    As we are going to analyze a little deeper below it seems that W3C project proceeded from digital to mind abandoning preliminary and necessary scouting tasks from mind to digital. Among these scouting tasks in the “Web Ocean” we may mention creating procedures to:

 

a) Detect and unveil keywords from existing documents;

b) Discover the main topic/s dealt with in documents:

c) Discover “authorities” and levels of authoritativeness;

d) Discover dominant logical semantic skeletons within the Web Ocean, namely hierarchically interrelated document clusters that share the same “main subject”;

e) Discover the fine hierarchical structure of these apparently -at first sight- homogeneous clusters;

f) Discover invariant meanings within each language (concepts);

g) Discover for each knowledge domain its modal “namespace” (by the way you may find here a deep difference between W3C and Darwin approaches: objects namespace in W3C are fictitious but considered as true if referred to a given Website taken as namespace authority meanwhile Darwin agents try to get modal names statistically that is the “best” name for any meaning at a given time within the Web Ocean);

h) Unveil dominant “well structured documentation patterns”, namely how top authorities document  their ideas. 

 

    In brief we could say that our mind is too complex and that we are also in the beginning of the beginning of its understanding. We have problems to define what an idea is, what a concept, an ideogram or a single keyword. Another problem we have to take into account when dealing with networks of Internet type is that man-machine interactions are performed in two domains instead of one: the Web documents’ side and the people’s side. W3C approach ignores this fact. It works more with metadata that with data and supposing that data is always clean and “well structured”. Darwin on the contrary works more with data than with metadata and it also works on the people´s side data trying to unveil the hidden semantic of real interaction queries.

    In terms of the Theory of Computing Complexity we were performing some experimental computations using the “Blum´s Axiomatic Approach” arriving to some preliminary conclusions that could be expressed as follows: if the OSI Seven Layers Model had a complexity of 10 powered at 15 a similar model trying to approach to the human mind would have a complexity of 10 powered to 19, four more orders.

 

The second best

    The pragmatic IT people who needs to solve hot problems applause and try to use those W3C tools and languages that consider useful now and there like XLM, RDF and some ideas of OWL. The applications people patch most of their semantic needs with programs and scripts written in C++, Java and .NET by “de facto” intertwining at their best wisdom thru the TBL seven abstract layers but not systematically but with punctual approaches far from universal, following by de facto a second best strategy. This circumstantial practical approach shall not invalidate W3C tasks, findings and ideals that should continue but with a different expectancy of observance

    Wrongly used and instrumented, W3C findings, without being fully aware of its actual limitations could be costly and dangerous as long as government agencies, universities, professional associations, and corporations were induced to implement “sine qua non” their approaches. It will imply to give priority to the form instead of to the essence. On the contrary if used as trends and ideals to be met as much as possible it will be like the big lantern of the Web evolution.

    W3C tools and languages used without common sense would resemble Ptolomeic versus Galileo Galilei approaches. Following W3C schemas to describe and solve actual applications would mean to fill thousand of description sheets and forms before start programming even trivial applications. And once everything is ready for programming the debugging task appears in the horizon as a real menace. What happens is that in scenarios of machines trying to understand, emulate and manage other machines, in order to coordinate common work among them, nothing is understood. For example a typical bureaucratic procedure in a government agency will render hundreds of steps thru hundreds of possible office “cubicles”. However human programmers advised and guided by red tape experts will go thru specific shortcuts of a few steps thru a small set of cubicles. It means programs with only a few control points to be checked versus hundreds, thousands for routine applications. That´s the cost of universality many times unnecessary and out of scale.. 

 

 

Última actualización el Martes 13 de Julio de 2010 15:11  

Imágenes Polls

Poll Darwin

Darwin puede ser usado para
 

Newsflash

Poll Semántico I

La Semántica es la
 

Poll Semántico II

La Web Semántica es una
 

Poll Semántico III

El Conocimiento Humano es:
 

Poll Semántico IV

El Tesauro Web es
 

Usuarios Online

Tenemos 20 invitados conectado

Breadcrumbs

Home Darwin versus W3C Visions - I