On Deconstructing the Hype

At the beginning of our study we learnt that the Web is at least distributed, decentralized and an open world.

First, different communities will use different vocabularies, resulting in problems of synonymy (when two different words have the same meaning) and polysemy (when the same word is used with different meanings).

There is no reason to deny this description at least as a starting point. Remember, the description of the weather system sounds very similar. But all these emphasis of the openness and decentralized distributedness of the Web is describing not much more than the very surface structure of the Web. It emphasize the use of the Web by its users not the definition and structure, that is, the functioning of the Web. There are no surprises at all if we discover that the structure of the Web is strictly centralized, hierarchic, non-distributed and totally based on the principle of identity of all its basic concepts. The functioning of the Web is defined by its strict dependence on a "centralized authority".

If we ask about the conditions of the functioning of the Web we are quickly aimed at its reality in the well known arsenal of identity, trees, centrality and hierarchy.

Why? Because the definition of the Web is entirely based on its identification numbers. Without our URIs, DNSs etc. nothing at all is working. And what else are our URIs then centralized, identified, hierarchically organized numbers administrated by a central authority?

"We should stress that the resources in RDF must be identified by resource IDs, which are URIs with optional anchor ID." (Daconta, p. 89)

What is emerging behind the big hype is a new and still hidden demand for a more radical centralized control of the Web than its control by URIs. The control of the use, that is of the content of the Web. Not on its ideological level, this is anyway done by the governments, but structurally as a control over the possibilities of the use of all these different taxonomies, ontologies and logics. And all that in the name of diversity and decentralization.

All the fuss about the freedom of the (Semantic) Web boils down to at least two strictly centralized organizational and definitorial conditions: URI and GOL.

It is not my intention to deny the massive complexity of the Web and the growing Semantic Web on its surface structure. Again, remember:

Some of these systems use relatively simple and straightforward manipulation of well-characterized data, such as an access control system. Others, such as search engines, use wildly heuristic manipulations to reach less clearly justified but often extremely useful conclusions.

Nevertheless, it is important not to confuse the fundamental difference of deep-structure and surface-structure of the Semantic Web. This fundamental difference of deep/surface-structure is used in polycontextural logic not as a metaphysical but as on operational distinction. And all the Semantic Web "cakes" are confirming it.

Here is another one from the W3C, its hidden cards, Unicode and URI, are shown in another game. Unicode and URI are the deepest layer of the Semantic Web Cake.

Beyond the layer of Unicode and URI we have to ad their arithmetical and code theoretical layers. The Semantic Web Cake is accepting the role of logic, down its propositional logic, but is not mentioning arithmetics. As we have seen in Derrida´s Machines, arithmetics and its natural numbers are pre-given and natural. There is not much to add. There are many possible open questions with Unicode and URI, but not with its common arithmetics.

The open question which comes back to my proposal is "Why should the deep structure of the Web be questioned?". At least, it is working. A simple answer, it is not enough. There are to many problems open which cannot be solved properly in the framework of the existing paradigm.

Our media philosophers are still fantasizing about the virtuality of the Web and the new Global Brain and bodiless decentralized sex, but there is no worry, the authority of the URI is controlling the game from the very beginning. And now we are going a step further, still not remarked by the critical media studies, and have to deal with a much more sophisticated attempt to the centralization and control of the Web by the GOL. Without a General Ontology Language there is no Semantic Web at all. GOL maybe made explicit or may remain in the background, as a new cyber-unconsciousness like the URIs, but it is ruling together with the Unicode and URIs the whole game.

The development of an axiomatized and well-established upper-level ontology is an important step towards a foundation for the science of Formal Ontology in Information Systems.

Why do we have to make such a drama about say, polysemy, if the Semantic Web is really in any sense decentralized etc.?

Our global village is dealing with the same, and simple problems, of the old Greek marketplace of discussions, all waiting for a great generalist, Aristotle, to make an end of the semantic chaos by introducing his GOL and Logic.

There is no surprise that the GOL of the Semantic Web is proud to be Aristotelian, it doesn´t change much to be more progressive with Whitehead , Bunge, Kripke or Montague.

All that is not working without conflicts. As we know from Guarani and probably also from the long history of western philosophical, logical and ontological thinking.

With such different interpretations of a term, we can reasonably expect different search and indexing results. Nevertheless, our approach to information integration and ontology building is not that of creating a homogeneous system in the sense of a reduced freedom of interpretation, but in the sense of navigating alternative interpretations, querying alternative systems, and conceiving alternative contexts of use.

To do this, we require a comprehensive set of ontologies that are designed in a way that admits the existence of many possible pathways among concepts under a common conceptual framework.

This framework should reuse domain-independent components, be flexible enough, and be focused on the main reasoning schemes for the domain at hand. Domain-independent, upper ontologies characterise all the general notions needed to talk about economics, biological species, fish production techniques; for example: parts, agents, attribute, aggregates, activities, plans, devices, species, regions of space or time, etc. (emphasis, r.k.)

The conflict between the desire and necessity to "navigate alternative interpretations" and the need of "domain-independent upper ontologies" is obvious and not easy to deal. Its virulence is quickly stopped by the acceptance of GOL, responsible for the definition of such simple things like "parts, agents, attribute, aggregates, activities, plans, devices, species, regions of space or time".

It will turn out that the general theory is not so much an ontology GOL but a theory of translating and mediating different ontologies, first order as well second-order ontologies. A Dynamic Semantic Web would add to the translations some mechanisms of transformation and metamorphosis.

Its main candidate is well known too: category theory, the ultimate theory of translation.

The general language of the Semantic Web is XML. But what is XML? Short: a tree. The same is true for the other languages like RDF.

As developed in Derrida´s Machines the main structure of formal thinking is natural. Everything has an origin and is embedded in a tree. Natural deduction systems, natural number systems and also the limits of this paradigm of thinking is natural. And this is also the way the Semantic Web is organized. XML is a tree. The tree is natural and universal.

"Whether a visitor comes from another place, another planet, or another plane of being we can be sure that he, she, or it will count just as we do: though their symbols vary, the numbers are universal. The history of logic and computing suggests a programming language that is equally natural. The language, called lambda calculus, is in exact correspondence with a formulation of the laws of reason, called natural deduction. Lambda calculus and natural deduction were devised, independently of each other, around 1930, just before the development of the first stored program computer. Yet the correspondence between them was not recognized until decades later, and not published until 1980. Today, languages based on lambda calculus have a few thousand users. Tomorrow, reliable use of the Internet may depend on languages with logical foundations. "

But the Semantic Web is artificial, and nobody until now has given a proof that the nature of artificiality is of the same nature as the concept of nature in all these natural deductions, natural numbers et al. Even to make such a distinction between natural and artificial is considered as obsolet and cranky by the academia.

What have we learnt on our trip around the fascinating perspectives and problems of a Dynamic Semantic Web?

It is all about dynamics and structures. This brings us back to the central topics of DERRIDA´S MACHINES: Interactivity between structures and dynamics, that is, to the interplay of algebras and co-algebras, ruled by category theory and surpassed by the diamond strategies leading to polycontexturality and kenogrammatics.

We arrive back to terms like translation, metamorphosis, polycontexturality, kenogrammatics, algebra and co-algebra, swinging types of algebras and co-algebras, etc.

A new effort has to be undertaken to collect the concepts, problems and methods of the Semantic Web into a more general and formal framework.

Not surprisingly, the main topic of the Semantic Web is translation, in other words a "interchange language". Translation of taxonomies, ontologies and logics. Translation as interaction, merging and transforming different domains, points of view, contexts. The most general approach to translation is given by the methods of category theory and semiotic morphisms (Goguen) not yet applied by the Semantic Web community. In this sense, translation is conservative, keeping the linguistic categories, tectonics and topoi together, that is, saving the meanings during the process of translation.

It seems to be obvious, that the languages of translation, mediation and metamorphosis are not languages of a general ontology as containing the "most general, domain-independent categories of reality" but languages which are neutral to ontologies, describing what happens between ontologies. There purpose is not intra-ontological but inter-ontological, mediating ontologies and not functioning themselves as ontologies.

Dynamics is not only covered by conservative interchange but interwoven in permanent transformations ruled by the play of metamorphosis. Metamorphosis can be understood as an unrestricted interplay of categories disseminated in a polycontextural framework. Metamorphosis is not only preserving but subverting meanings in the process of interactivity. Translation is interchange, metamorphosis is creation of new meanings.

The behavior of the Semantic Web is best modelled in terms of an interplay of algebras and co-algebras in the general framework of category theory. But this is as I have shown enough only a very first step in modeling the interactivity of autonomous systems. This means, that I reject the idea of modeling the structural dynamics/dynamical structure by category theoretical morphisms only.

Interactivity comes with reflectionality, architectonics and positionality. These topics have to enter the game to design a more dynamic Semantic Web as it is considered by the very simple and conservative procedures of merging and integrating ontologies and creating contextual concept spaces.