To whom it may concern:
At this time, I want to retract my usage of "The F-bomb" in
an earlier posting that I made to this newsgroup. In all fairness,
perhaps my posting in regard to "wisdomics" was slightly on the
vague side, leaving me open to some of the "internet personalities"
I have seen in other academic and technical newsgroups on the inter-
net who seem to look for breaches of unwritten newsgroup protocols
as an excuse to "ride their high horses off to the flame wars".
However passionate I may have felt and still feel about my studies,
two wrongs do not equal a right and I am as bad as Torsten for
engaging in "gutter tactices". Being an amateur, independent
researcher with cognitive gifts but a diagnosed emotional disability
is still not an excuse here.
Now I shall explain how my hypothesis (not theory) should have
originally been stated. Oz, in my mind, is a special language
for reasons that are expounded upon in Van Roy and Haridi's book
and the Mozart/OZ website. Not much more has to be done to sell
me on this point based upon a year's worth of (but not certainly
not constant) time of learning about different programming languages
and paradigms off the internet.
Neutrosophics, encompassing neutrosophic logic
and neutrosophic cognitive maps among other topics, is a means of
expressing neutralities between contradictory statements in
mathematical form. It is based on a dynamical systems approach to
computation (complexity theory stuff), and this is "The Force", in
Torsten's terminology, that makes "things come alive", so to speak.
I think most would agree that complexity theory is a new discipline
that offers alternative explanations for the way nature carries out
its processes. When I read about neutrosophics, I abandoned efforts
of my own to settle upon a working definition of wisdom in
computational systems. I believe that neutrosophics is a start in
the right direction (http://www.gallup.unm.edu/~smarandache/philos.htm)
based upon what I had experienced previously in life.
Before neutrosophics, however, I was very interested in the
topic of software agents because experiments have shown that they
exhibit (to some degree) humanlike traits of trust, communication,
collaboration, negotiation, etc. These traits seem "wise" to me.
In my studies, I picked out a system named AKIRA (which is currently
under development). Its project description reads as follows: AKIRA
is a run-time C++ multithreading and clusterable environment able to
execute software agents and a development platform to model their
behavior. The system core is made up of a server daemon that answers
network programming and AI computation requests and that executes agent
instances. It also includes a programming interface based on a C++
macro language and some automated scripts that allow the creation of
new agents. Two soft computing technologies are provided (Fuzzy Logic
and Fuzzy Cognitive Maps), together with a high-level, psychologically
valid Goal Oriented Programming Language: BDI (Belief Desire
Intention) Model. Neutrosophic cognitive maps are an extension of
fuzzy cognitive maps. In the begninning, my experimental idea was to
observe if AKIRA processed better using "FCMs" or "NCMs".
By way of extension, I found an article about using the Oz constraint
language for natural language processing
(http://www.ps.uni-sb.de/~niehren/Web/Vorlesungen/Oz-NL-SS01/vorlesung/
).
I speculated that perhaps if NLP was subject to constraint language
(optimization), there could be a possible tie in between an "optimizing
system of agents" like AKIRA (augmented by NCMs) and agents having
"more intelligent conversations". What I mean by the preceeding phrase
is that agents would start out by transmitting garbled bits of words and
phrases by NLG (natural language generation) and the receivers of other
agents would use NLU (natural language understanding) to "critique" the
degree of understability of what had been transmitted. As agents began
to communicate, per satisifying a system fitness function, the
transmitters and receivers would be appropriately tweaked to make the
communication more understandable. Maybe, over time, the agents might
start to have conversations that would be considered "interesting" to
human observers in terms of the "wisdom" -- a human judgement call --
they communicate.
It is interesting to consider how the model I have described above
could additionally be extended -- affective computing, ontologies, common
sense, etc. Since AKIRA is middleware and highly extensible, this may
be entirely possible. This model is what I mean by the term "wisdomics".
However, a may be useful to port AKIRA to Mozart/Oz and write the NCM
module in Mozart/Oz because it offers, I think, possibility more flexibility
and expressiveness in the future due to its multiparadigm programming
language features.
I am putting the idea up on the internet because although my emotional
disability (interestingly enough) sometimes enhances my ability to come
up with many novel ideas, execution suffers because my ability to
concentrate and organize is more limited compared to other people. I am
proud of my ideas and I don't want them to "die on the vine" so to speak,
so I am looking for the support of others. I would even care to dream
that they may fulfill a "social good", since all technology should enhance
man's way of life as opposed to making it more difficult (in the case of
pollution, detachment from society due to "cyber immersion" in children,
for example). I don't suggest that computers replace human wisdom, but I
think most would agree that perhaps the process of designing the beginnings
of a wise computer system could shed some light on our own ability to
reason effectively.
Craig
_________________________________________________________________________________ mozart-users mailing list [email protected] http://www.mozart-oz.org/mailman/listinfo/mozart-users
