Tuesday, December 13, 2005
Proposal: To identify a common, widely
used, and available concept ontology to be used with UDEF (Universal Date Element Framework) and other
finite coding systems underneath. (
see [298] )
BCNGroup December 12, 2005
John and members of the ONTAC working group
To summarize John's paper,
http://www.jfsowa.com/logic/theories.htm
Theories, Models, Reasoning, Language, and Truth
from my point of view:
Your central conclusions are
· that natural language has properties that are not present in, at least, those formal systems that are derived from the Greek traditions.
· that the interpretant has a great role in localizing experience both via the cognitive as well as the emotive centers. (but you make no operational suggestion – in this paper - as to how to move forward on core-hub-common ontology plus basic structural data interoperability).
· and that a lattice of theories might be used to support analogous computations (but this is unproved)
Am I close?
I am sure I have slanted these conclusions, but I tried for quite a while last evening to find some way to operationalize what you were saying in the context of approaches that have been discussed here in this forum.
I must say that I utterly failed to find some path forward based on your conclusions. I am renewed in the conjecture that "all" logic should be !! separate-able !! from what ever "common" interoperability mechanisms are promoted and deployed. This means that the concept of "formal semantics" of which Tarski was one of the causes of, should be publicly set aside, in much the way that it has recently become politically correct to set aside AI (of course this political correctness is not universal).
I am not advocating that logic not be integrate-able with (controlled vocabularies)!!
I am advocating that approaches be extended along the line made in the founding topic maps community (I forget the detail here – but the key word is “Grove”) related to modeling roles, aspects, lines of possible entailment, and open-ended referential anchors (in the URI). The key end goal is to actually use the OASIS standard that Rex Brooks and his group has developed on a human behavioral mark up language. Rosette net and CoreTalk indicated a similar process markup language where process is “standard business processes”.
Robert Rosen handles the category of all models in quite a different way than did Alfred Tarski. Rosen split entailment into logical entrainment and referential entailment. The logical entailment was not tied to classical predicate logic, but was tied to the notion of coherence. There is no parallel to Rosen’s notions about referential entailment in the works of Tarski. (At least this is what I gather. Tarski is not so easy for ME to read, because I feel he has a lot of things “ontologically” incorrect, as did a lot of other Philosophers.) It is a judgment call on my part that others clearly disagree with.
Rosen referential entailment is related to those things that exist structurally in the world at the time of an emergence. There was not an assumption that the world was closed and thus referential entailment was something, from this point of view, that could not be formalized (at all). The work on quasi axiomatic theory by Pospelov and Finn, that I talk about in my book; addressed this absence of formalization – but according to Peter Kugler (and I agree) the QAT work failed to take into account the perceptual measurement problem. Again, I talk about this extensively in my book and make part of the required scholarly references from that work.
http://www.bcngroup.org/area3/pprueitt/kmbook/Reference.htm
The formation of categories, the definitions of set membership, were all developed with the proposition that formal systems have, in some undefined sense, less of an existence than did the natural system. In Finn and Pospelov’s work, this formation of categories in the natural world can only be modeled (in a formalism) if at each step (in deduction chains for example) there is a “checking” process where by a human uses human awareness, memory and anticipation to check the results of computations.
etc...
Are you saying in this short paper that the approach using logic is at an impasse?