Tuesday, February 21, 2006
The Second School of Semantic Science
ontologyMapping Glass Bead Games
On the limits of the OWL standard à [184]
Reading material [1]
Reading material [2]
Reading material [3]
Summary of the discussion up to this point à [186]
Ontology mapping bead game thread
The community would like to have a non-sales pitch description of the differences between information and theory. This difference is not understood, and is in fact confused by standard references to standard meanings. (This is a conjecture.)
Regarding the related question of translatability between:
> 1) RDBMS
> 2) Object Oriented or UML
> 3) descriptive logic based ontology
> 4) n-ary ontology
The second school’s position is that the n-ary is primary to knowledge representation as a semiotic system. For a discussion of semiotic sign systems and ontology, please start with the notional paper, and tutorials on Ontology referential base (Orb) models. The BCNGroup RoadMap for semantic technology adoption is also a good starting point.
Published discussions between Knowledge Foundations Inc, founder, Dr Richard Ballard have been included in the BCNGroup Bead Games. At one point a conceptual index will be made over the bead games. So a summery of the issues will be repeated here, and in this bead thread. At this time, we do not agree with a major part of the Knowledge Foundation’s literature.
How does an n-ary come about? (see also [5] )
The neighborhood of a five tuple, using Orb technology
The answer is through empirical observation. The observation is about the co-occurrence of semantic atoms. (semanticAtoms – future link to a wiki for definition). The n-tuple for a chemical compound is not the full knowledge of the compound available form modern chemistry. So one needs a theory to be applied to a n-ary. (Not all arbitrary n-aries are meaningful.) (again see [5] )
This is a powerful enough solution so that a business value is available for you and me and others to earn a good living over the next decade... , in my opinion. The issue of how to allow a generation of computer scientists to make a living using the best tools, is relevant.
The current complicated-ness of Protégé and the W3C standards has to do with two entailments:
1) the development of professional tools where the use of these tools is very much restricted, by design?, by a requirement to have deep knowledge and experience with, what appears to be, un-natural conventions…. (example of this is to be placed at [6] : date: 2/23/06)
2) the shifting sands of research directions as the descriptive logic (DL) approach to knowledge engineering is explored
a. sometimes this exploration leads to dead ends and the DL standards are refined towards a better description logic
b. the exploration itself is about something that is likely to be complicated even when it is fully figured out. However, it is the opinion of some that DL based ontology has flaws that cannot be overcome, rather these flaws, like the flaws of Codd-type relational database design, have to be understood and the paradigm used in a way that has “eyes wide open” regarding the sometimes over hyping of these paradigms. (Like the AI over hyping that occurred in the 1970s – the present).
Well the paradigm shift from the First School to the Second School?
Convolution theory is what will "bind" the appropriate n-aries together in real time to produce an operational ontology (with "structural indexing or "reasoning").
This technology will produce real time "reification" of information as a situational ontology (ie as a semiotic "system of signs")... with relationship into structural standards developed by various communities.
From this description it should be clear that n-ary representation of ontology is the next "standard".
Of course, the "business problem" is that the n-ary standard is a radical simplification of the other three categories of information standards. The conjecture that there is a radical simplification has to be show (as a precise proof).
THE PROOF HAS TO BE CONSTRUCTIVE, ie no theoretical considerations will do. One has to show the mapping methodology. (see also [3] and [6] )
Proprietary interests would seem to be in the way.... except that the value of this new transformation (I conjecture) is worth far more that is imaginable. There is enough economic rewards for anyone who understands this standard early. So the question now is clarity and transition.
Conjecture: Relational database powerhouses like Oracle will make more money (from the transition), because the new information systems will find persistence in the relational form, while adding huge values to the B-2-B, G-2-G and more generally the Peer to Peer transaction space.
The Second School hopes to develop a new type of discussion and language so that the issue become less hidden (as is now the case with the W3C DL and OWL standards) and more clear.
Paul