----------------

Theory-based semantics software/technologies

"Theory-based semantic technologies possess a common formalism for assigning identifiers to concepts, ideas, thought patterns and their conditional relationships, giving them precise meaning as defined by recognized and well-justified theories. The advantage of theory-based semantic technologies is that they by-pass language ambiguity, the inability to scale, and the reasoning performance limitations of language-based semantic webs. Unlike conventional technologies, theory-based semantic technologies provide the capacity to precisely capture every form of human knowledge, scale to integrate vast knowledge corpuses with minimum encoding, without complexity explosion, and to reason as humans do, operating transparently with all representational forms." - Richard L. Ballard

How are theory-based semantic tools different from conventional technologies?

Architecture (Computational versus Declarative)

Conventional technologies and theory-based semantics offer two distinct architectural models. Traditional machines, software, and databases rely on execution, logic, algorithmic processing, mathematical true/false proofs and structured schemas that require the same data be processed every time an outcome is required. The conventional model provides no means for machines to learn from past computation, requiring greater performance as the demands from growth increase complexity.

The limitations of this conventional model are typical of the database/ computer relationship. Human architects create data structures called database tables that must be programmed before the database can be populated with desired content. Most of this content, such as street, city, state and zip code is duplicated and stored many times over as new customer content is entered into the database. Likewise, as users execute commands to view desired content, the computer processes the same data using the same algorithms - every time. It does not take much thinking to realize how these layers of built-in redundancy can slow a system down - especially when thinking about huge volumes of data generated by organizations and the Internet. The standard solution for this limitation has been to build faster computer chips, but the physical limits of nature have stalled this trend.

Knowledge science dictates a new direction. Theory-based declarative semantic architectures establish "capture once, use forever" computing model that goes beyond first and second order logic, true/false, self-consistency proofs, and eliminates in-advance schema construction and computational redundancy. Theory-based semantic architecture is not built from design, but rather emerges as a self-organizing web structure from its own self-assembly process. Semantic structures do not require outside human architects (though knowledge engineers do assemble content sources), and they grow to unlimited size.

Guided by absolute bit limits of Shannon's Information Theory and Ballard's Quantitative Theory of Knowledge, theory-based semantic structures, or knowledgebases, are pre-computed webs of linked concepts and reasoning paths justified by theory. The addition of information describing a situation or circumstance enables full, precise, constrained representation of the knowledge of that situation or circumstance understandable by both humans and machines.

Complexity (Legacy of early computing models)

Life and the requirements for intelligent systems are n-dimensional. Conventional technologies attempt to manage this complexity through layered abstraction (tiers of metadata, taxonomies & ontologies). This trend exists because conventional computing methods evolved from the early 1940's when simple theories, hardware and primitive computing methods were designed as work-arounds for very expensive memory and storage resources. It made economic sense to continually re-compute rather then store sizable volumes of data.

Today, the requirements and economics of that early model have completely reversed. Theory and data requirements are now massive and complex and memory and storage costs are low. Yet, the conventions of repeat computation and structured data schemas have continued to the point where technology is bumping-up against physical and complexity limits that were unthinkable in the 40's.

Complexity is experienced as:

• Excessive “total cost of ownership”
  
• Explosive growth of networks, systems of systems, and communities
• Machine and software inoperability
  
• Massive data accumulation
  
• Semantic dissonance - the resulting problem of search.
  
  

Structured technologies, to include triple and quad-stores, “hit-the-wall” in terms of their capacity to handle very large knowledge corpuses as well as complex and dynamic relationships. Even chip development, previously the savior of complexity, has reached the 4.33 GHz limit (though there is talk of transistors that operate at 604Ghz?).

Knowledge science shows that the solution to complexity exists, but cannot be achieved through layered abstraction. Rather, the solution is to capture all knowledge used by both humans and machines in n-dimensional theory-based declarative semantic form, at absolute bit limits, leading to new software and computing architectures. If 100% of knowledge exists within a subject domain, databases can capture 15%, semantic web technologies might capture another 5% to 10%, but theory-based semantics has been proven to capture up to 100%. Complexity is decreased in direct proportion to the percent of theory a technology can capture and use. Likewise, Knowledge Foundations' theory-based semantic technologies contain natural limits and efficiency measures that guarantee the absolute smallest bit limits on storage and execution, and the least cost in every situation. These physical limits insure there is no system that can be faster or more efficient, no matter how large or complex the organization, project or job knowledge environment.

Scalability (Structure versus machine behavior)

Scalability to conventional technology means adding more processors, more memory, more algorithms, more machines, more users, and more layers of software to solve the next problem. At a deeper level, the fundamental belief behind computer algorithms and database schemas requiring logic, mathematics, language and grammars, automatically rules out machine learning and the ultimate capacity of machines to change their own behavior - a fundamental characteristic of true scalability.

Knowledge science establishes that both execution and storage should be proportional to information content, so both information and theory storage remain scalable to unlimited size. The semantic codes within the n-dimensional form contain only one copy of every concept or idea required by the knowledgebase, and are found without search by its own unique identifier.

In contrast to conventional technology, self-organizing n-dimensional semantic webs automatically reorganize and scale to meet the demands of newly introduced content.

Human DNA has been relatively unchanged for 100s of 1000s of years. What has changed (the major difference between us and our ancestors), is the number of theories that humankind has learned and passed down from one generation to the next. What do schools and universities teach – theory. What do parents teach their children? Theory. What do people learn on-the-job? Theory. All human endeavor is theory based. For machines to become equal partners with humans, they must have the capacity to capture theory, to invent theory, and to reason with that theory in novel situations. Development is underway leading to a technology that will allow Knowledge Foundations' semantic webs to learn and change their own behaviors from direct experience. These are machines with the capacity to think as well as, or better than humans.

Humans can only hold 3-7 competing ideas in their minds at any one time to solve problems. Knowledge Foundations theory-based semantic systems concurrently consider hundreds and thousands of competing concepts and ideas to solve even the most complex questions and problems.

Security (Advanced declarative security & protection)

Hackers, identity thieves, and system pirates have learned how to make the limitations of conventional technology work for them. They exploit programming holes and processing functions to download invasive programs into systems that monitor user behavior and capture and use proprietary data. Because Knowledge Foundations' theory-based declarative environment is a non-algorithmic, non-processing environment, spyware, adware, viruses and other invaders do not function within its n-dimensional environment. This knowledgebase technology also provides bullet-proof, multi-level, need-to-know, need-to-access security, along with automatic password tampering shut-down functionality and 5/9's-tolerant reverse engineering protection.

What is the added value of theory-based semantic technologies?

Technology Lifecycle: Development costs

When organizations make strategic changes to operations such as adding new divisions, products, services or expanding into new locations, there is most always a need to purchase additional hardware and software. The cost of these software acquisitions include purchase, development & integration, training, maintenance, losses from down-time and lost opportunity due to interruptions during the normal course of business. Though today's machines and software are less costly than a decade ago, development costs for backend integration are 15 times the cost of the software license itself, or higher, and takes several months, or years to completely implement. Enterprise software programs that cost $100,000, cost an additional $1,500,000 to develop and integrate.

Declarative semantics presents a unique strategic purchase option. As an all knowledge system (theory & information), theory-based semantics cost-effectively integrates backend "systems-to-systems" without machine interoperability problems. Development time and costs are significantly less while delivering seamless interoperability of disparate, cross-functional data sources. Of equal or greater value, is that knowledge engineers can effectively capture existing systems data and document content, plus the actual job knowledge of employees, consultants and subject experts. Owning knowledge gives organizations previously unrealized use of vital knowledge to solve operational problems such as:

• The right hand knowing what the left hand is doing.
  
• The ability to track budgetary expenditures through the organization hierarchy with value/expenditure assessments per expenditure.
• Provide employees machine delivered, immersive, on-the-job training.
• Capture and integrate every detail (past, present and future), related to human resources, equipment lifecycle or cradle-to-grave projects.
  
  

Organizations gain even greater value because captured job theory can be stored and used for decades, centuries, even thousands of years. For this reason, KnowledgeAssets that are owned by an organization can be bartered, licensed or sold to a global market. Installed platforms emerge as continuous knowledge commerce where knowledge environments endure for decades and centuries and where content becomes more valuable than the system or the technology itself.

Maintenance and change configuration

Work environments are not static. They require on-going change to accommodate management objectives, evolving markets, technology trends and organizational growth. Structured architectures, computational processing and database schemas, once developed and configured, are costly to change because structural changes often cause daisy-chained malfunction errors within machines and software, or human error due machine forced procedural changes. The semantic answer to this costly problem is that change configuration is an integral part of a semantic system. The theory-based semantic architecture is self-organizing, self-adjusting and self-transcending. It is a "high granularity environment" where every concept is defined by its relationships to other concepts and ideas, and not by a schema, naming convention or linguistic identifiers. In a theory-based semantic web, every idea appears just once and every other relationship attaches to it in a massive n-dimensional space.

Efficiency, performance and profitability

Knowledge decreases uncertainty. To the degree that uncertainty is removed, actual and hidden costs related to time, efficiency, performance and production are decreased, as efficiency, performance, production and profitability is conversely increased. In cases where projects are so large and complex that its managers cannot get their arms around the problems of those projects, theory-based semantics has value well beyond mere cost savings.

The real cost of conventional technology is the drag it imposes on operations. It is estimated that companies lose as much as 25% or more of employee work-time while they search for document sources and content required for their work. Since this time could be dedicated toward customer retention efforts, lost time also represents lost opportunity. When employees, temporaries and consultants cannot find the answers they need to perform their jobs on-the-fly when they need it most, efficiency, effectiveness and performance suffers. The cost, as revealed by a Fortune 100 Human Resource Professional, is both economic and psychological.

N-dimensional, theory-based semantic design is extraordinarily powerful, yet people friendly, because it works the way people naturally think. Because there is no search, but a concept query and navigate function using the K-Browser, people find precisely what they are looking for at the speed of thought. Sales people who need to see all the documents related to a customer, or more precisely, specific content within those documents that is subject specific, can find them within seconds and minutes. Likewise, an HR person can know exactly the job responsibilities, performance rating, supervisor, pay rate and testing scores of an employee at anytime during the employees employment. Through Knowledge Engineering, every form of knowledge can be captured with proven veracity time past, present and yes, as a predictive tool, the future.

© Knowledge Foundations 2004 - 2005