The Second School, ontologyMapping thread

Community centric service methodology à [home]

This thread is preserving initial discussions on ontology based modeling of complex social processes.

Paul S Prueitt

Janet Kopec

NCOIC communications

Janet,

NCOIC uses a number of branding phrases such as found in the following;

"Industry working together with our customers to provide a Network Centric environment where all classes of information systems interoperate by integrating existing and emerging open standards into a common evolving global framework that employs a common set of principles and processes."

My comments in brief.... (I ask other scholars to make a comment)

Since the founding of NCOIC there has been some progress on the core issue of data interoperability. However, there continues to be reports and other evidence that interoperability breaks down when novelty arises within any deployment situation.

A common set, of terms and phrases with semantics, has been hard to find, and common agreements about open standards has been seen to continue to have deep proprietary roots. The discussions at the CIO Council meetings and other meetings have shown that un-recognized difficulties continue to exist. For example: John Sowa and I and others have expressed concern that RDF and OWL standards are built on a poor technical foundation. Another example: The treatment of the topic map standard by the IT industry has been problematic.

A practical issue seems to be core to both the advocacy of RDF/OWL and the dis-crediting of topic maps. This issue is the issue of the origin of control, as discussed at www.secondschool.net .

Novelty detection is achieved in plants and animals using several systems, including the immune system and the neural system. There are shared principles in the design and functioning of these systems. The science on these designs and functioning is and has been available in the natural science. This science continues to be beyond the selective attention of IT industry management and government procurement officers.

I will review my understanding of the science of perception, and ask that NFS program manager (neuro-engineering) Paul Werbos make a comment.

On going category formation required of a natural recognition system creates an interaction between two types of sensors, one type that is localized and one type that is measuring a field potential. Karl Pribram's work in quantum cognitive neuroscience is but one small part of the scientific literature related to the role of locality and non-locality in perceptual acts.

What one may observe is that the common interests between computer science and business creates a separate reality, the reality of business interests, not fully connected to natural reality as seen by the primary natural science. The emergence of concepts in humans depends on natural law in ways that the computer scientists are generally not tuned into.

This observation is controversial, but increasing evidence indicates that a shift in the natural science is not informing the on going development of IT systems. The problem is deeper, given the focus of funding by NSF in the old IT paradigms, almost all of which are reductionist in nature. The problem extends even further due to the capture of political and economic processes by a particularly sever form of ideology. This ideology is resilient but not sustainable. [1]

In ontological modeling, using either topic maps or web ontology language, the key element is the development of classes, subclasses and membership in a way that the membership defines the class. This is often called "reification". When reification is done well, one can use one of the various description logics to produce inferences.

Can class-subclass reification always capture the pragmatics of situations? The answer may simply be in the negative.

The usefulness of reified class hierarchy is easy to see, when dealing with simple cases. The break down of this usefulness takes many by surprise and has placed the industry into a state of denial.

With this reaction it becomes impossible to understand that on going new contracts and applications of systems and standards, from the battlefield to FAA threat detection for example, will create serious vulnerabilities. Both human enemies and natural phenomenon will exploit these vulnerabilities over time. The expenditure of great wealth will be largely wasted.

[65] home [67]