Thursday, February 16, 2006
Generative Methodology Glass Bead Games
n-articulated ontological framework
On using RDF to model biological expression
Link back to part of the “solution” to translations between
RDF / OWL and Models of Information
ŕ- [167]
Concurrent Anticipatory web discussion ŕ [30]
Discussion on expressiveness
Communication from Paul
Judith Rosen said:
" Living organisms don't carry complete information about themselves or their environment in those "internal predictive models" my father described in Anticipatory Systems Theory... we can prove this in many ways. "
as part of her communication at:
http://www.bcngroup.org/beadgames/anticipatoryWeb/30.htm
This assertion about biology is specifically counter to the requirement in W3C ontology standards. These standards require that a standard have the ability to describe itself.
Now there are good reasons, in theory and practice, why the W3C standard (and other standards) include this requirement.
I am not precisely sure of all of the reasons.
The discussion that I have placed at
http://www.bcngroup.org/beadgames/generativeMethodology/177.htm
is addressing what is described by ontology logicians as "expressiveness", and related topics described as "decidability" and "tractability".
It may be that, to the degree that one wants to instrument information with an index from a OWL ontology (which is what we seek to do with the OASIS ebSOA-IM to OWL mapping...and related mappings), this self-referential requirement (like the unique ID requirement) may in fact be absolutely necessary.
But this necessity has to do with the nature of finite state machines, and not (necessarily) with the nature of the real world of living system interacting in a natural world? Right?
The definition of the Open World Assumption given by the ontology logicians is not the same as the definition of an open world assumption in general systems theory, since the means in which new information can be added to an OWL ontology is measured with the ruler of "consistency", and logical consistence is not (we conjecture) how the real world (always) measures the emergence of phenomenon [1]. Otherwise the phenomenon of emergence would not require, or make use of, a stratification of layers of interactions. By layer of interactions we mean that there are gaps existed between the layers, and that these gaps restrict the cross scale interactions. (Sorry, this all can be made more precise but it takes a while.)
Simplified example: Individual genes do not interact with my finger strokes to the keyboard.
If I am correct about this difference between a natural system and a formal system, we have great opportunities to develop and use Information Models and mappings to Descriptive Logic based ontology; while at the same time looking to a non-formal means (the human or some aspect of natural world) when the questions of fidelity, expressiveness become an issue (as the OASIS FERA model suggests, these issues become issues through a measurement process).
See [179]
The "solutions" being suggested by both Ulrike and Andrea are deeply informed.
I would like to explore their suggestions further, but I need some help in this conversation.
Andrea and Ulrike, first is the editing at 177 ok? and second .... how do you see the issues at this point in the conversation?