Sunday, June 11, 2006
Generative Methodology Glass Bead Games
On ontological modeling of expression
The metaphor between gene, cell and social
expression à [217]
On Formal verses Natural systems à [206]
Link to education commons à [***]
Regarding the excellent paper at:
http://www.twason.
Tom,
I think you agree, that the process you describe, in this paper, is a measurement (knowledge elicitation) of the information flow and the structure of language used by a (any) community of interest. Your previous use of this methodology in distance learning communities is of great interest to me and to others who are beginning work on a specification for service measurement and instrumentation/facilitation in Higher Education.
What we are looking for is an OASIS standard that specifies how this measurement should occur, in cases where there is a social information exchange and the object is to
1) measure the information flow (create an information flow diagram)
2) create standardized information structure that aligns with the information content observed in this flow
(See also a paper I wrote for Admiral Poindexture at TIA DARPA - in 2002 on harvesting the social discourse.)
Your methodology is strong on the use of multiple viewpoints to create the higher dimensions (tensors in the language of Hilbert Technology - ie as in keyless hash tables) of this information space. The encoding into "n"-ary structure using these keyless hash tables allows (one to two orders of magnitude faster data "mining" using simple convolution theory as discussed at:
No logic is imposed for the purpose of inference, rather a sign system (semiotics) is created as an "KOS" (knowledge operating system) interface between a user and the mobile (Grid) operating system nodes using Hilbert mathematics (not sting processing) and n-ary representation. The reason that there is no string processing is that all strings are regarded ("hashed") as a number in base 64.
The shifting of viewpoints can be done (has been shown to be done) in the Hilbert "n"-ary space (using convolutions) and "reconciliation of terminological differences" as has been done by the SchemaLogic SchemaServer 2.0 for several years. This move from XML to "n"-ary processing removes the search and find issues with XML and by-passes the need for XML accelerators. Provably, same results however, only 100 times faster with 10 - 20 times less in memory memory requirements. (no indexes are needed)
Moreover the methodology shifts the origin of information design from the IT community to the community of interest (by allowing a measurement of normal conversation {information exchanges} and using this measurement as the cause of SOA Blueprints with reuse and normalization occurring as per the newly adopted BCM standard.
www.businesscentricmethodology.com
As you noted, the information space has to be weakly bound so that managed vocabulary evolves.... as new realities impact the community - such as a new policy or some event like Katrina or 9-11-2001.
I have developed a simplification of the difference between
1) OWL/RDF specification of the information structure needed to match the information exchanges in a community of interest
2) Topic Map use to connect with and let go of interpretations that are made of real time information structure flowing within a community of interest.
This simplification highlights the maturity of the OWL/RDF (web-ontology) and a more full recognition of its limitations, and the potential new development in using Topic Maps as a process model for merging OWL (Full) ontology as considered by the www.BioPAX.org activity over gene and cell expression bioinformatics data interoperability.
This simplification is at:
http://www.bcngroup.org/beadgames/generativeMethodology/complexOntology.ppt