| |
|
|
|
© 1998 Animation: A.T. Kelemen
"The Semantic Web is an extension of the current web in which information is given well-defined meaning, better enabling computers and people to work in cooperation." -- Tim Berners-Lee, James Hendler, Ora Lassila, The Semantic Web, Scientific American, May 2001
Dynamic Semantic Web (DSW) is based at first on the techniques, methods and paradigms of the emerging Semantic Web movement and its applications. DSW is advancing one fundamental step further from a static to a dynamic concept of the Semantic Web with grater flexibility in the navigation between ontologies and more profound transparency of the informational system. Web Services are now redefinded by Semantic Web. To proof the advantages of DSW, it is the main aim of this project to developed the tools and methods necessary to develop a DSW based Web Service (DSW business application).
The existing framework of the Semantic Web has only very limited possibilities of realizing dynamism. It´s dynamism is reduced to inter-ontological transactions (translations, mappings, navigation) between different local taxonomies and ontologies.
DSW is based on the genuinely dynamic first order ontologies and logics founded in kenogrammatics of the theory of polycontexturality allowing evolution and metamorphosis to create complex interactivity and new domains of interaction.
The Dynamic Semantic Web may be realized by the play of interweaving Computation, Co-Creation, Navigation and Negotiation.
3 The Semantic Web
Today, the Semantic Web is becoming an important reality. Not only in research centres but also in industrial, business and governmental organizations, Semantic Web applications are advancing. Semantic Web is understood as the "Next Web".
"There´s a revolution occurring and it´s all about making the Web meaningful, understandable, and machine-processable, wether it´s based in an intranet, extranet, or Internet. This is called the Semantic Web, and it will transition us toward a knowledge-centric viewpoint of éverything´." Stephen Ibaraki
As the WWW is based on HTML, the Semantic Web is based on XML as its frame language mediated by ontologies. Ontologies are the new key to meaning in information processing. Also deriving from philosophy where ontology is representing the most general theory about being and the formal structure of everything, in the Semantic Web, ontologies are of a very pragmatical value. "Ontologies are about vocabularies and their meanings, with explicit, expressive, and well-defined semantics-possibly machine-interpretable." Daconta
XML is the corner stone of the Semantic Web. "XML is the syntactic foundation layer of the Semantic Web." It is not a language; it is "actually a set of syntax rules for creating semantically rich markup languages in a particular domain. In other words, you apply XML to to create new languages."
"Why is XML so succesful? XML has four primary accomplishments, (...):
XML creates application-independent documents and data.
It has a standard syntax for meta data.
It has a standard structure for both documents and data.
XML is not a new technology (not a 1.0 release)."
More explicit, XML is characterised by following principles:
First: "Markup is separate from content."
Second: "A document is classified as a member of a type by dividing its parts, or elements, into a hierarchical structure known as a tree." Daconta
The Semantic Web is possible today and in reality it is a natural consequence of the fact of the Internet, the WWW, the knowledge about databases and the ubiquity of powerful computing facilities.
Two years ago the Gartner Group has given a marketing projection that "By 2005 lightweight ontologies will be part of 75 percent of application integration projects".
DERI-Centres: Ireland and Insbruck (Austria)
The importance of research in this area is indicated by the recently announced DAML initiative in the USA, under whose aegis projects aimed at developing the Semantic Web will receive DARPA funding totalling $70 million.
The merits of the Semantic Web is that it is in its concepts and in its vision very pragmatically oriented. It is in sharp contrast to the sometimes very speculative aims of Artificial Intelligence.
A sharp distinction between Semantic Web and AI can be made between the relevance and understanding of data and programs. AI is concerned with highly complex programs being at the end able to understand data, e.g. texts and common sense. Semantic Web is more concerned in making its data "smart" and giving them some machine-readable semantics. AI tends to replace human intelligence, Semantic Web asks for human intelligence.
On the other side it seems that Semantic Web is lacking, at least today, strong and complex logics, automated deduction systems and inference machines. Topics which are well developed in AI research and applications.
It is well known that AI has produced a lot of knowledge about Knowledge Representation systems, Concept Analysis and many other semantic based endeavours. Nevertheless, Semantic Web takes a new start on a more pragmatic level, with a more business oriented vision and from an other angle of the whole spectre of "mechanizing" knowledge and interactivity.
The Semantic Web is based on its ontologies. Ontologies are playing the key role in the process of realizing semantic information processing. Ontologies are themselves classified in several types. The most general case is the distinction between core ontologies and upper-level ontology. There are many core ontologies but only one upper-level ontology. The structure of ontology (and ontologies) is strictly hierarchical.
"What are the real values for using ontologies? The real value of using ontologies and the Semantic Web is that you are able to express for the first time the semantics of your data, your document collections, and your systems using the same semantic resource and that resource is machine-interpretable: ontologies. Furthermore, you can reuse what you´ve previously developed, bring in ontologies in different or related domains created by others, extend yours and theirs, make the extensions available to other departments within your company, and really begin to establish enterprise- or community-wide common semantics." Daconta, p. 237
Additional to the link structure of HTML, RDF (Resource Description Framework) comes with a pointer to the resource of the data (object, information) introducing a semantic dimension to the strict syntactic definition of HTML.
A description is a set of statements about the resource.
The RDF model is often called a "triple" because it has three parts: subject, predicate, object.
Subject: This is the resource that is being described by the ensuing predicate and object.
Predicate: This is a function from individuals to truth-values with an arity based on the number of arguments it has.
Object: This is either a resource referred to by the predicate or a literal value.
Statement: This is the combination of the three elements, subject, predicate, and object. (Daconta)
All this is governed by the principle of identity.
"We should stress that the resources in RDF must be identified by resource IDs, which are URIs with optional anchor ID." (Daconta, p. 89)
This linguistic characterization of the RDF triple is defining a statement and adding to its syntax some meaning guarantied by the identifiable IDs. This relation is decidable, that is, the connotation exists or it exists not, therefor it is true or false-TND.
Dov Dori, 2003, http://dori.technion.ac.il/opm/
At this point I would like to mention, that despite of its semantic relation and its foundation in a generally accepted ontology, this RDF triple is defining a statement in isolation, excluding its context. Later, contexts are introduced by ontologies. But the RDF definition is not involving them. As a consequence, all pragmatic points of views have to be introduced secondarily. It would be helpful, if we could introduce this contextual information at the very beginning of our construction. Without this we will simply repeat the paradoxes of knowledge engineering of the AI projects. That is, meaning of a sentence is context-dependent and contexts are defined by meaningful sentences.
In this proposal I will concentrate myself on the basics of Semantic Web as it is proposed by its inventor Tim Berners-Lee and his three-part vision: (collaborative web, Semantic Web, web of trust).
Diagramm 45
The Semantic Web Stack
To begin with the top: trust. Let´s have a look to an example.
Trust or Distrust? Serious or a joke? How serious is the joke? Or is it simply stupidity?
Also there are more serious example today in politics out there, take a look at this funny search result from a proposal by the high church of W3C and MIT.
"An important feature provided by the Semantic Web is the ability to respond to the query "What is the basis for that statement?"; to help a user to answer the question "Do I believe that?", or alternatively, "How much risk is there to my achieving my objectives if I act on the basis of that statement?". The language of the Semantic Web must thus be able to represent information destined for automated processing.
Example: Simple Search Made Sensible
We provide an example illustrating how the Semantic Web permits bridges between the vocabularies developed by independent communities.
Today, we might ask "What is bmw7's power?" and get no answer back because BMW's web page gives the answer in German terminology. With the Semantic Web, we might retrieve an answer based on heuristically driven classical logic:
* What is bmw7's power?
1. search( bmw7, power, x ) yields a nil result
2. search( bmw7, subclass, x ) yields x=beamer, car, thing
3. search( beamer, power, x ) yields a nil result
4. search( power, equivalent, x ) yields x=Macht
5. search( bmw7, Macht, x ) yields x=160kw
6. deduce( bmw7, power, 160kw )
* The bmw7's power is 160kw.
And, of course, the Semantic Web allows us to investigate further:
* Why?
Because power is the same as Macht and the catalog from BMW says the Macht of the bmw7 is 160kw."
http://www.w3.org/2000/01/sw/DevelopmentProposal
This conclusion is a result of the lacking of contexts. Obviously it is simply wrong, if taken as serious. In introducing at least two contexts, one physical and one political, we are able to play between (the political) "Macht" and the (physical) "Kraft". As a neutral observer, equipped with the necessary back-ground knowledge in German and English, we can not decide if the above answer is serious or a joke or part of a different game.
3.1 Hierarchies everywhere
A taxonomy is a semantic hierarchy in which information entities are related by either the subclassification of or the subclass of relation.
One of the basic distinctions of GOL is the distinction between urelements and sets. We assume the existence of both urelements and sets in the world and presuppose that both the impure sets and the pure sets constructed over the urelements belong to the world. This implies, in particular, that the world is closed under all set-theoretical constructions. Urelements are entities which are not sets. They form an ultimative layer of entities without any set-theoretical structure in their build-up. Neither the membership relation nor the subset relation can unfold the internal structure of urelements.
In GOL, urelements are classified into two main categories: individuals and universals. There is no urelement being both an individual and a universal.
Diagramm 46
UML hierarchy diagram of a General Ontology
Conceptual graph of the basic triple (Entity, Urelement, Set) and its uniqueness 1.
Uniqueness means that there is one and only one ontology defined in terms of Urelement, Set and Entity. This also means, there is only one World, and at the end it means, there is only one WWW, too. But this is homogenizing complexity and diversity, and is simply a monstrous nomiminalisation. In other word, it is one and only one way of thematizing the world, the mono-contextural one.
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. Every domain-specific ontology must use as a framework some upper-level ontology which describes the most general, domain-independent categories of reality. For this purpose it is important to understand what an upper-level category means, and we proposed some conditions that every upper- level ontology should satisfy. The development of a well-founded upper-level ontology is a difficult task that requires a cooperative effort to make signicant progress.
Diagramm 47
Axiomatic Foundation of Upper-Level Ontologies
Contributions to the Axiomatic Foundation of Upper-Level Ontologies, Wolfgang Degen, Heinrich Herre
All these axioms of the formal general ontology GOL are not only defining a (probably) consistent framework for all possible applicative, core ontologies, but are also asking a hard price for it: there is no dynamics in this framework of ontology. Everything is what it is, e.g. Urelement or Set. Any dynamics is secondary and localized in "chronoids", "topoids", etc. which are special cases of Individuals. In other words, no Urelement can become a set and vice versa, simply because this ontology is mono-contextural, lacking any fundamental perspectivism and interactivity with diversity.
In contrast to the Aristotelian entity-ontology a more holistic and processual formal first-order ontology for the Semantic Web is accessible by the OPM conception developed by Dov Dori. Therefore, even on a first-order ontology level comparison, translations, conflicts etc. are possible. At least it becomes obvious that despite the fundamentalist desires of GOL to be the ultimate ontological theory for everything, it is also simply only one of the possibilities to conceptualize the topics.
4 How to introduce the Dynamic Semantic Web?
The Semantic Web movement is not only strong and inevitable, it is also open to the future. On a pragmatic level it is open for an increasing multitude of local and personalized systems. It´s general definition is monitored by the W3C, but in encouraging new developments and not restricting its future progress.
In this sense the Semantic Web movement includes without problems a spectre from Aristotelian fundamentalists to Rhizomatic Anarchists.
In other words, it is not in contradiction to the guidelines of the Semantic Web to develop as a new branch the paradigm of DSW.
It is a philosophical question if this branch is well understood as branch and should not be better thematized as something quit different, namely as an interlocking mechanism between core and upper ontologies and their logics distributed over different irreducible upper ontologies.
From a pragmatic point of view, DSW is better localized as a new branch or discipline of the Semantic Web.
The map of the Semantic Web assembles all sorts of theories, methods, implementations from philosophy to hard core programming, including AI and data-base technologies, logics, semantics, context theory, linguistics, neural networks, etc. on all levels of scientifity and scholarship, not excluding some confusions and other cocktail events.
This is allowing a great diversity of different approaches to be involved in the development of the Semantic Web and its extension to the Dynamic Semantic Web, and many other invention, too.
To deal in a flexible and controllable way with decentralized heterogeneities, hierarchies are not delivering the best possibilities. Here is the moment where heterarchies come into the play.
Decentralization and Heterogeneity is obviously in conflict with the strict reglementations of upper-level (first order) ontology as it is formalized in the general ontology GOL.
Two different contexts relating respectively to species and environment point of view.
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 schemas 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.)
http://www.loa-cnr.it/Publications.html
4.1 Heterarchies, in general
In contrast to the Semantic Web with its tree structure, that is, with its fundamental hierarchic organization on all levels of conceptualization and realization, the Dynamic Semantic Web comes with a strong decision for heterarchies.
Heterarchies are not fully understood if we are not studying the interactivity between hierarchies. In this sense heterarchies are the framework of the interactivity of hierarchies. In other words, heterarchies are ruling the interplay between a irreducible multitude of different trees.
One great advantage is, each of these trees is inhereting the well known and proofen methods and technologies of their classical predessor, that is, logics, taxonomies, proof systems etc.
"Whereas hierarchies involve relations of dependence and markets involve relations of independence, heterarchies involve relations of interdependence."
"Stark has proposed "Heterarchy" to characterize social organizations with an enhanced capacity for innovation and adaptability.
Networked or lateral organizations are in direct contrast with the tree-like, vertical chains of control of traditional hierarchies. The second feature means that heterarchies require diversity of components and building blocks." [Stark, 1999, page 159],
http://www.c3.lanl.gov/~rocha/GB0/adapweb_GB0.html
To give a more transparent modeling of the interactivity between hierarchies as it is proposed by the proemial relationship it maybe helpful to set the whole construction and wording into an UML diagram and to use the modeling of heterarchy worked out by Edward Lee as a helpful tool to explicate proemiality in terms of UML modeling.
Also the proemial relationship is not restricted to ontology and the distribution of hierarchical ontologies in a heterarchic framework and despite the fact that UML has no mechanisms of category change, metamorphosis and mediation it seems to be a helpful exercise to find a correspondence between the UML heterarchy diagram and the construction of proemiality which is more based on elementary terms of relationality. The heterarchy diagram is a class diagram which models the static structure of the system. Proemiality has, also it is fundamentally dynamic, its static aspects. It is this static aspect we can model with the help of the UML heterarchy diagram. A further step of UML modeling of proemiality will have to involve more dynamic models like interaction and activity diagrams.
Diagramm 48
UML heterarchy diagram
The conceptual graph of the UML heterarchy diagram may highlighten its structure more directly.
It shouldn't be misleading to read the diagram as a (methodological) hierarchy between the terms Heterarchy, Hierarchy and Entities. The additional terms Model, Frame, Port, Relation and Link are defining the structure of the interaction between the different hierarchies.
Each hierarchy has its own ontology, logic, algebra, proof systems etc. To give an idea of the concept of interactivity between hierarchies let´s introduce the terminology of abstract objects or types or theories.
"First of all, a name is given to the theory so that it becomes an identifiable unit binding together a number of operations and their properties into useful modules.
Keyword sorts opens the theory, listing the sorts or types of objects being defined in the abstract type.
Next we have keyword opns followed by one line for each of the operations or predicates being defined in the abstract type.
Constants are seen as zero-arity operations.
The equations are defining equivalences between strings." (Downward, p.179)
Short, the abstract theory consists of the categories name, sorts, operations, equations which build, again, a strict hierarchy of their tectonics:
The arrows in this diagram represents conceptual dependencies in the notion of name. The notation
opns --> sorts
for example, means that:
the concept of opns varies as the concept of sorts varies.
In particular, it means that the concept of opns, the one that we have in mind, cannot be independent of the concept of sorts and neither can a particular opn be independent of its particular sort.
The notation
sorts --> name
means that the concept of sorts varies as the concept of nat0 varies.
Therefore the notion of opns varies as the notion of nat0 varies:
opns --> name.
In a conceptual diagram, 1 represents the absolute. The notion
name -->1
expresses that the name notion is absolute, for it tells us that the name notion varies as the absolute varies - which is not at all.
Heterarchies are managing distributed hierarchies, therefor we are able to distribute abstract theories as such. This in itself would produce an interesting type of parallelism, architectonic parallelism. But more interesting are the interactions between hierarchies. A very conservative interaction is a one-to-one translation from one abstract theory to another abstract theory, based on morphisms with name to name, sorts to sorts, operations to operations. This form of interaction is basic for a successful realization of DSW applications.
But the advantage of DSW come into play with the possibility of metamorphosis, that is the change of categories. This capability of DSW anables evolution of the system, discovery and creation of new domains, and marks the distinct difference to other architectures of a Semantic Web.
There is an easy way of producing conflicts in a dialogical system, if e.g. L1 declares its object A as a simple object and L2 declares simultaneously A as a complex object, that is a structure. Obviously it is possible, in the polycontextural approach, to model this conflict and to resolve it in another logical system, say L3, this without producing a meta-system subordinating L1 and L2.
Diagramm 49
Tree of data objects
Furthermore, the conflict has a clear structure, it is a metamorphosis of the terms simple object" in L1 and structure" in L2. This metamorphosis is a simple permutation between sorts over two different contextures based on the chiastic structure of the mediation of the systems. But it respects the simultaneous correctness of both points of view in respect of being a simple object" and of being at the same time a structure". In this sense it can be called a symmetrical metamorphosis.
Today computing is often characterized by its interactivity. But the programming languages have not changed to respond to this situation. They are still, in principle, mono-logic.
4.2 Ontology and the Semantic Mapping Problem
Why do we need all these abstract theories of translation and metamorphosis?
"One important issue in understanding and developing ontologies is the ontology or semantic mapping problem. We say "or semantic problem" because this is an issue that affects everything in information technology that must confront semantic problems-that is, the problem or representing meaning for systems, applications, databases, and document collections. You us always consider mappings between whatever representations of semantics you currently have (for systems, applications, databases, and document collections) and some other representation of semantics (within your own enterprise, within your community, across your market, or the world).
"This semantic problem exists within and without ontologies. That means that it exists within any given semantic representation such as an ontology, and it exists between (without) ontologies. Within an ontology, you will need to focus on a specific context (or view). And without (between) ontologies, you will need to focus on the semantic equivalence between different concepts and relations in two or more distinct ontologies." Daconta, p. 218/19
This citation shows us the importance of mappings (translations, morphisms) between distinct ontologies. But don forget, these ontologies are applied, core ontologies, regional, and not general ontologies. They are parts, subsystems, instantiations of the one and only one general ontology, as formulated in GOL. This is an enormous restriction. Because, before we can interact with each other we have to agree to this general and global framework of GOL. But this is not always reasonable at all.
DSW is introducing mappings, morphisms, translations and metamorphosis between first-order ontologies, and is not concerned with only regional, that is core ontologies.
How does it work? The basic framework is given by the proemial relationship (Günther 1970).
"The answer is: we have to introduce an operator (not admissible in classic logic) which exchanges form and content. In order to do so we have to distinguish clearly between three basic concepts. We must not confuse
a relation
a relationship (the relator)
the relatum.
The relata are the entities which are connected by a relationship, the relator, and the total of a relationship and the relata forms a relation. The latter consequently includes both, a relator and the relata.
However, if we let the relator assume the place of a relatum the exchange is not mutual. The relator may become a relatum, not in the relation for which it formerly established the relationship, but only relative to a relationship of higher order. And vice versa the relatum may become a relator, not within the relation in which it has figured as a relational member or relatum but only relative to relata of lower order.
If:
Ri+1(xi, yi) is given and the relaturn (x or y) becomes a relator, we obtain
Ri (xi-1, yi-1) where Ri = xi or yi. But if the relator becomes a relatum, we obtain
Ri+2(xi+1, yi+1) where Ri+1 = xi+1 or yi+1. The subscript i signifies higher or
lower logical orders.
We shall call this connection between relator and relatum the 'proemial' relationship, for it 'pre-faces' the symmetrical exchange relation and the ordered relation and forms, as we shall see, their common basis." Günther
Diagramm 50
Proemial relationship
4.3 What to do with all that for a theory of a Semantic Web?
"No self-reference is possible unless a system acquires a certain degree of freedom. But any system is only free insofar as it is capable of interpreting its environment and choose for the regulation of its own behavior between different interpretations." Gunther, p. 44/p.156
The Internet is not given, its elements are not entities; the Internet has to be read and its elements have to be interpreted. Interpretation involves freedom to chose a thematization, a perspective of cognition, it involves not only an observer but hermeneutical procedures. Otherwise we understand by the Internet a system of being to be studied and classified by means of ontology in the very sense, also modernized and formalized, of the Aristotle-Leibniz tradition.
The project Semantic Web is a challenge for a formalized and operative hermeneutics. Set-theoretical and mereological ontology is mapping only an extremely static and one-sided hierarchical aspect of the "living" tissue of the Web.
A multitude of interacting hierarchies is a question of cognition and volition interpreting the textures of the Web.
Translations from one language to another are not based on a common natural ur-language, but on the co-creative interplay between different languages, natural or artificial.
Ontology in the sense of GOL is "subjectless". It is a theory of being excluding self-referentiality by definition. Therefore it is a monolitical theory of what is, of objectivity without any freedom of interpretability. Again, this is very useful for subjectless domains, but useless, if not dangerous, in all sense of the word, for worlds including subjects. Today it seems to be quite tricky to find such a subjectless world. Especially if we are forced to ask who is producing this ontology of a subjectless world and even our robot are asking for more "subjectivity". Ontology as "the most general possible theory about the world" is fundamentally incomplete. To insist on a realist point of view to build a general ontology in contrast to a conceptualist understanding of ontology allowing some interpretability of the world is a decision which can not be justified easily using scientific and philosophical arguments. At least this decision is not part of the "new" formal ontology.
The introduction of a plurality of first order ontologies, that is of poly-contexturality, opens up an interesting new interplay between the hierarchical order between core and upper level ontologies to a proemial exchange of their functionality. To the many possibilities of metamorphic interplays between basic ontologic terms of different fundamental ontologies, like sets vs. urelements, additionally core ontologies can change to upper and upper to core ontologies, ruled by proemial operators. There is no need for a fixed and single hierarchy between different types of ontologies. Again, this is possible only on the base of polycontexturality and its proemiality.
4.4 Reflectional Computation in the Dynamic Semantic Web
5 Development of a DSW Prototype Business Application
5.1 Attributes of static and evolving DSW Systems
This "killer application" will show a significant increase in flexibility, which goes hand in hand with an increase in speed and transparency of general semantic information processing.
The dynamics of the semantic information processing in DSW opens up thew possibility to create new scenarios, invent new forms of interaction between business partners.
Happily, the Semantic Web community has developed lots of useful tools, free or commercial, to be used to develop the prototype of a DSW business application.
5.2 DSW as Panalogy Problem Solver
The Panalogy Principle: If you 'understand' something in only one way then you scarcely understand it at all-because when something goes wrong, you'll have no place to go. But if you represent something in several ways, then when one of them fails you can switch to another. That way, you can turn things around in your mind to see them from different points of view -until you find one that works well for you now. And that's one of the things that `thinking" means!
ţ
We cannot expect much resourcefulness from a program that uses one single technique-because if that program works in only one way, then it will get stuck when that method fails. However, a program with multiple `ways to think'-the way we'll describe in Chapter §7-could behave more like a person does: whenever you get frustrated enough, then you can switch to a different approach-perhaps through a change in emotional state.
Minsky
Implementing Panalogy
I will use the term Panalogy to refer to a family of techniques for synchronizing and sharing information between different ways of thinking concerned with the same or similar problems. The term derives from `parallel analogy'. By maintaining panalogies between ways of thinking, we can rapidly switch from one way of thinking to another.
We can also make more partial changes like the representation language they are using, the types of assumptions they are making, the methods that are available to them for solution, and so forth. The key idea is to support representing multiple problem solving contexts simultaneously and the links between them. A graphical depiction of panalogy at work is shown below in Figure 6.
5.3 Architectural Parallelsism in DSW Systems
5.4 Web Services and Semantic Web, the classical view
http://www-106.ibm.com/developerworks/xml/library/x-ebxml/
Diagramm 51
Web Service Scenario
Diagramm 52
Semantic Web Services
A metaphor of the internal dynamics of the components Semantic Web, Web Services and RDF, WSDL is given by the chiastic figure of the Ying-Yang-Picture by Wolfgang Dostal and Mario JeckleSemantik, Odem einer Service-orientierten Architektur.
http://www.jeckle.de/semanticWebServices/intro.html
Diagramm 53
Ying-Yang-Picture
5.5 A DSW business application is a DSW Semantic Web Service
THE Internet and THE WWW doesn't exist. THE WWW is a crude and awfully misleading nominalisation and abstraction from the evolving heterogeneous complexity of what we call the WWW.
THE Web Services are not a homogeneous business. They come in different and not homogeneous forms, that is, again, in heterogeneous definitions.
Heterogeneity itself is not a static term, too. It is a nominator for a flexible, loosely coupled evolving complexity of decentralized systems.
The Web is not only defined by its abstract specification but also by its use. The meaning of a sentence is not given by a catalog of administered meanings, but by its pragmatic use. And the administration of meaning is one and only one very special use of sentences and their meaning.
The picture of the situation has to be enlarged from Syntax&Semantics to, at least, Syntax&Semantics&Pragmatics (Hermeneutics).
Pragmatics or Hermeneutics is introducing different points of view, different irreducible contexts, that is, contextures, different approaches, different ways of thinking (Minsky) etc. "The Semantic Web is composed of heterogeneous systems. No real system can answer every possible question about its data (i.e. none can compute the deductive closure of its information). Different systems make different compromises, affording different sets of inferences. For this reason, one application will often be able to reach a particular conclusion when another does not."
Mediation (Proemiality, Chiasm) is introducing the interlocking mechanism, the interactivity of all these different contextures. "The existence of hundreds of subcultures on the Web, and also the existing of millions of legacy systems, requires powerful mediation." Daum, p. 10
Negotiation is realized by human beings. But it is strongly supported by the mechanisms and rules of mediation. Insofar, DSW is not only introducing computer-aided semantics, but also several levels of computer-assisted negotiation.
This is in contrast, or better, in positive addition to Daum´s statement: "Also obvious is that by the default the communication between observers can only be of informal nature. Consistent logical systems are only defined within a given context and, in general, cannot be used for knowledge transfer between different ontologies. The consequence is that some means of informal communication, such as natural language or heuristic mediation systems, is inevitable." Daum, p.185
Diagramm 54
Dynamic Semantic Web as a Pragmatic Web
Maybe that the structure of the metaphoric dynamism of the Ying-Yang-Picture is captured and formalized by the dynamics of distribution and mediation of contextures containing the basic quadruple of its different realizations.
5.6 The Dynamic Semantic Web between Algebras and Co-Algebras
"The chaotic nature of the Internet requires powerful navigation tools." Daum, p. 10
The Web is not a program. It is not solving a problem, starting with a question and ending with an answer. The Web is much more a medium in which all sorts of computational actions can happen. Therefore it is badly described by formal languages with their hierarchic algebraic structure.
Complementary to the programming aspect the dynamism of the Web is better characterized by co-algebraic methods.
5.7 What has do be developed to realize DSW?
DSW is based on a polycontextural philosophy and methodology. This approach has to be transformed to real-world applications.
Dynamic Semantic Web (DSW) consists in general of two main parts:
2. inter-Semantics or Pragmatics of mediation and navigation
Remember the Semantic Web hierarchy:
Trust, Proof, Logic Framework, Rules, Ontology, Contexts and
RDF Schema, RDF M&S, XML; Namespace and
poly-Semantics deals with the decomposition and distribution of different heterogeneous taxonomies, ontologies and their methods.
inter-Semantics deals with the interlocking mechanisms between the different heterogeneous contextures and their methods.
Development of applied polycontextural ontologies
Development of polycontextural logics and proof systems
The existing developments of formal polycontextural logic has to be adapted to the needs of DSW and concrete pcl-applications has to be developed.
5.8 Where do we need DSW?
DSW is everywhere needed where different cultures have to coexist and to interact together.
This happens with merging companies, with cooperation with different business cultures, it happens on a local and on a global scale as multi-cultural interactions.
It mostly doesn't make sense to unify all these distinct simultaneously existing cultures into one big mega-system.
5.9 How to establish a DSW system in a existing company?
It is not necessary to transform at first a business information system into a Semantic Web and in Semantic Web based Web Services. We can directly create a Dynamic Semantic Web transformation of the knowledge management system of an organisation.
1. Discover the heterogeneity of your data base.
Instead of trying to homogenize the different data systems it is more reasonable to understand them as an interacting system of heterogeneous parts. As a mediating tool to the full decomposition of a monolitic data base into its heterogeneous parts, the method of Metapattern introduced by Pieter Wisse maybe a helpful methodology. This methodology goes far beyond the usual techniques of decomposition of database systems because it is involving a new non-hierarchic kind of ontology and semantics. Metapattern can help to move from classical hierarchic ontologies to genuine polycontextural ontologies, semantics and their logics.
The classical Prolog example to prove an "aunt"-relationship can be decomposed from its hierarchical ontology into different situations mapped into different contextures and visualized in the metapattern.
kinship: married/not-married, in-law, aunt
gender: male, female
genealogy: parent, sibling
ontology: different/not-different
It is also possible that there is some over-determination because parent and sibling could also be part of kinship.
In Prolog all the facts belong to one ontology or to one semantic general domain or universe. All the rules are based on this mono-contextural ontology and on the corresponding logical operators AND and OR of the again, mono-contextural logic. Everything therefor is linearized and homogenized to a global or universal domain. This, if corresponding fairly with the real world situation is of great practicality and efficiency in both direction, in the case of the formal system, Prolog, and in the case of its data base.
But often, if not always, real world applications are much more complex than this. Even the fairly classical example is presupposing all sorts of facts which are not mentioned in the definition and which would belong to a different real world situation.
Instead of linearizing the above separated contextures kinship, gender, genealogy, ontology into one universal domain, for the example here represented by kinship, the polycontextural modeling is asking for an interweaving and mediating of these different contextures together to a complex poly-contexturality.
Why should we model a simple situation with highly complex tools into a complex model if we can solve the problem with much simpler tools? Simply because the classical approach lacks any flexibility of modeling a complex world. The truth is, that the simple approach needs an enormous amount of highly complicated strategies to homogenize its domains to make it accessible for its formal languages.
2. Decompose your data jungle into heterogenous contextures.
3. Build your ontologies out of the distinct heterogenous contextures.
4. Discover the interlocking mechanisms between heterogenous systems.
5. Learn to navigate between different contextures and points of view.
With the help of the tools of implemented chisams you have control and transparency about your navigations.
Navigation is more than translations (semantic mapping) or merging of local ontologies it opens up the possibility to access distinct "foreign" ontologies for cooperation which would otherwise be undiscovered.
To make business is not restricted to one business model, like the US american one. Globalization has not to homogenize different other ways of making business. Dynamic Semantic Web opens ways of mediating heterogeneous approaches on all levels of information processing.
6. Find leading metaphors for decomposition, mediation, navigation, negotiation which are accepted by your group and organization.
 ThinkArt Lab |
| |
|
|
|
