Informational Transparency
using
post OWL and post XML Ontology referential bases (Orbs)
Friday,
July 23, 2004
Short “Systems AS-IS”
Paper
Short “Finding the Balance” Paper
Friday, July 23, 2004
previous discussion on
“structural holonomy” -> .
An exploratory talk by Dr. Paul S. Prueitt
Talk given at: Institute for Defense Analysis July 28th, 2004
Informational transparency is available by using a stratification of structure from function. The management of this stratification is doable using an extension of Soviet era “applied semiotics” and control theory related to automatic development of information about patterns (what Lev Goldfarb calls Inductive Informatics). When this management of stratification is done using action-perception cycles (J. J. Gibson), we have Human-centric Information Production (HIP) within anticipatory webs of information.
The currently available Orb technology includes the measurement of co-occurrence of terms within word-level n-grams moved over filtered web harvests:
http://www.bcngroup.org/area2/KSF/Notation/notation.htm
This encoding involves two, or three, startling innovations. One of these involves the CCM (Contiguous Connection Model) formalism, developed by Applied Technical Systems Inc and deployed at INSCOM and at Army Intelligence (General Alexander). The extension to CCM that my group has made is non-trivial, and yet explainable in simple terms. An additional extension, in theory, makes Orbs interoperable with both Cyc and OWL ontologies as well as produces differential ontology having ontology primitives (as in John Sowa’s work on “cognitive graphs).
My group has also extended a post-relational/post-object-oriented development environment based on a 2003 patent by Prementia Inc, called the Hilbert Engine. The Hilbert encoding simply treats any ASCII string as a base 64 number in positional notation. The result starts as a key-less hash table, having no collisions and no need for empty hash containers. With this engine, one creates a data schema independent means to clean and integrate any relational database and then re-express the data completely into any other relational schema.
Standard set theoretic operations are reduced to a small number, < 60, machine cycles even within massive databases. The extension my group has made is non-trivial and is explained in a proposal I have made to NASA on the analysis synthesis and storing of all Earth Observational Data.
http://www.bcngroup.org/beadgames/InOrb/theoryOfInformation.htm
Informational transparency requires a complete encoding of all data patterns and then a type of selective attention, using associative memories. Complete encoding of all data patterns is possible because the natural world always has a , time dependant, structure. Using the compression of data into a small finite set of pattern structures, the human mind is primed by data of a specific sort in a just-in-time fashion. Novelty detection is then essential, as well as issues related to incorrectly assessing whether or not something makes sense.
http://www.ontologystream.com/area2/REAL/innovativeClaims.htm
These issues are sophisticated. However, it is clear that the business case has not yet been made that transforming information technology might precede the transformation of the intelligence community. I would like to make this case.