RKC-Re: Re: RKC-MORE

<Bob Clark (940809.1508 EDT)>

[Bill Leach 940807.00:42 EST(EDT)] Subject: Re: RKC-MORE CNTRL,DISTRB

Your using the general definition of "disturbance" is not a problem
-- when the purpose is a "general" one. For application to The Test,
it needs to be limited. I think your suggestion of
"Test-disturbance" has merit.

But more important, I think, is recognition of the need for
"calibration." _Two_ observations are necessary, one with the system
connected and one with it disconnected. In some cases, where
reliable computations are possible, the second observation may be
unnecessary.

I remember thinking that is was "necessary" to "lie" a little when
teaching basic electronics. I don't necessarily believe that one
has to do that but I do believe that many will "give up" if you do
not at least use a few "over-simplifications" in the early stages of
the learning experience.

The problem with "over-simplifications" at any stage in the learning
experience is that they can hardly avoid being, at least partly,
incorrect -- hence misleading. Commonly, "over-simplifications" are
in the form of analogies, similes, or metaphors.

"analogy, n. 1. a partial similarity between like features of two
things, on which a comparison may be based."

"simile, n. 1. a figure of speech in which two unlike things are
explicitly compared."

"metaphor, n. 1. the application of a word or phrase to an object
or concept which it does not literally denote, in order to suggest
comparison with another object or concept."

These definitions demonstrate that they are loaded with possible
misunderstanding. Indeed, they are effective only when both objects
are already known. For definition of new terminology, this type of
definition is guaranteed to produce misunderstanding!

The use of these concepts tends to be revealed by such terms as "a is
like b," "x resembles y," and similar "warnings."

These verbal forms are excellent for literary writing. But for
accurate communication, especially in technical matters, they are
likely to spell disaster.

"How do you define the 'taste of chocolate'?" "It is like this ..,
but a bit more .., with some .. added" -- NO! It can only be
defined by sensory experience.

"How do you define a 'number'?" It the result of the procedure of
"counting," which can be demonstrated.

After quoting from my comments, you state:

Your first sentence tells me that even you perceive your "limit" as
artificial.

Of course. All words, as words, are "artificial." They are defined
in many ways for many purposes.

Later:

The term "function" has both a specific and general meaning and I
was using it in the general sense.

Yes -- but in discussing PCT it is used in the specific sense.

Regarding "triggering" of the reorganizing system:

I would say that the jury is still out on that one but indeed the
"activation " of a control action from the reorganizing system is a
result of a perceptual error.

This is not a "jury" matter. "Reorganization" and "reorganizing
system" are _defined_ in B:CP. If you want to discuss something else,
ok, but don't apply these terms.

Personally, I prefer "n-system," the original label for this concept.
There were misunderstandings with it, but I think they were much less
important than the shifting from the original definitions of
reorganizing system that develops from the general concept of
"reorganizing."

Finally:

I don't really think that we are arguing about a disagreement in
belief here. All control system failure should contribute to
intrinsic error.

How did "belief" get into the act?

Why should "all control system failure contribute to intrinsic
error?" Look at the definition of intrinsic error -- it pertains
exclusively to built-in physiological requirements. When I hit the
wrong key on my key board, is there an intrinsic error? When I find I
took the wrong exit on the Freeway, is there an intrinsic error?

Even if we select our words carefully, there will still be
misunderstandings.

Regards, Bob Clark

<[Bill Leach 940809.22:15 EST(EDT)]

<Bob Clark (940809.1508 EDT)>

I agree that analogy, metaphor and the like have serious problems.
However, they often are an appropriate "starting point". I would tend to
think that in this subject area, such methods will sometimes have value.
The rigorous approach to theory of control systems does not necessarily
ensure understanding of the significance of the idea that living systems
are control systems.

There is no doubt in my mind that what understanding that I do have is
due largely to the painstaking efforts (most especially on Rick's part
but indeed of many) to use nearly every literary technique at their
disposal to keep "driving me in the correct direction." Not that I am
in any way a PCT "expert" but at least for my own purposes, I believe
that I understand PCT IN A WAY that none of my previous engineered control
systems work would have likely guided WITHOUT significant efforts of
others.

Virtually all of the "stories", examples and the like seriously lack
precision and were "loaded" with assumptions but they "sparked" questions
and lead to discussion after discussion about seemingly trivial (as it
seemed to me at the time) points.

I suppose that my conviction is that there are no "short-cuts", there is
no easy way and maybe even more importantly... there is no ONE correct
way to teach PCT. The path to understanding is probably somewhat unique
for every individual and the learning process for two people may differ
so markedly as to make one wonder if the subject matter is even the same
(though the ultimate fundamental concepts must be the same of course).

My comment about "over-simplification" is probably largely related to the
idea that often "new comers" to any field are overwhelmed by the scope of
the field as they begin to study. Over-simplification frequently allows
a person to begin to grasp a "global" understanding and indeed once the
"big picture" is somewhat comprehended then the "detail" information that
in the early stages of learning seemed so overwhelming often makes sense
to the point of even being "obvious".

I believe that many people that "give up" on a "new" subject are quite
often not "lazy" or lacking desire but rather are almost "driven away" by
those that are unwilling to allow the "new person" a chance for any
measure of "incremental" progress. I am not saying this to refute what
Mary stated so succinctly (there are no "baby steps"). I agree with
this. One can not really understand PCT partially. That is there is
sort of a "threshold" below which "one just doesn't quite 'have' it."
However, it takes effort to get over that "threshold"... effort on the
part of the person studying and most emphatically, effort on the part of
those trying to help.

I don't know the solution except that I am certain that there is not a
"one size fits all" type solution. Thus, on reflection, I am sure that
the effort that you are making will help some people even if not others
and is therefore worth the effort.

The "problem" with PCT is that the Control System Theory defines quite
precisely in mathematical terms what closed loop negative feedback control
is all about. Even when someone "understands" this definition of
behaviour of control systems, one does not understand PCT. The real
understanding of PCT is somehow related to the idea that one begins to
appreciate that this precise mathetical definition actually is defining
one's own behaviour.

I also tend to think that understanding PCT includes realization that
some of the massive detail of human motion that borders on SuperComputer
power of computation to "control output" is instead the "natural" result
of perceptual control system operation. That is, control systems don't
"plan" output in the sense that other attempts at "duplicating" human
motion have always used.

Your first sentence tells me that even you perceive your "limit" as
artificial.

Of course. All words, as words, are "artificial." They are defined
in many ways for many purposes.

My point was that you worded justification for your definition in such a
fashion that you appeared to be acknowledging the more general
definition.

Function

OK, I will repent though I still do not think that there should have been
a missunderstanding in the context in which I used the word.

This is not a "jury" matter. "Reorganization" and "reorganizing
system" are _defined_ in B:CP. If you want to discuss something else,
ok, but don't apply these terms.

There is a "party line" on reorganization. However, unless I completely
missunderstood Bill P. (and others), the precise details of how and when
reorganization "kicks" in is untested. In particular, matters such as
does any amount of intrinsic error result in corrective action with this
system? Is sustained error required? Will very high error levels from
some systems result in an immediate error output from the reorganization
system?

Further, How or intrinsic errors sensed (what is the perceptual
function)?

How did "belief" get into the act?

The same way it does for any other "truth" that we think is true.
Hopefully, in this case it is based upon some measure of application of
"scientific method".

Why should "all control system failure contribute to intrinsic
error?" Look at the definition of intrinsic error -- it pertains
exclusively to built-in physiological requirements.

And what are these physiological requirements? Failure to achieve
control does not contribute to intrinsic error? I'm afraid that is not
my understanding. Certain errors seem to have pretty strong indication
that there is a close link to reorganization but I have not read of any
proof either way. The reorganizing system exists in the theory (as I
understand it) because it is essential to have something of that sort to
explain behaviour that "pure" control theory will not explain. That
modeling of simple biological system behaviour strongly supports the
reorganizing concept lends great support for the general concept but
hardly defines its scope or operational details.

When I hit the wrong key on my key board, is there an intrinsic error?
When I find I took the wrong exit on the Freeway, is there an intrinsic
error?

Yes, there may well be -- sometimes and at other times to "contribution"
to somekind of "general error level" may be insignificant. I don't think
that anyone really knows but "clinical depression" could be the "feeling"
associated with "high" general error level.

Even if we select our words carefully, there will still be
misunderstandings.

And here with your closing remark, there is no disagreement. In this
forum, only through effort on the part of all involved will reasonable
understanding be achieved. We certainly have misunderstood each other
enough times and there is no doubt in my mind that neither of us intended
to misunderstand the other nor mislead the other.

-bill