<Bob Clark (940731.1514 EDT)>
Bill Powers 940727.0715 MDT) Subject: Definitions; PCT and IT; the
argument
I hope the Convention was enjoyed by all. We were disappointed that
we were unable to attend this year. Better luck in '95!
Back to the discussion. For clarity, I'll quote from your post:
Definitions are tricky; as you say, one has to use judgment about
which words really need definitions and which you can assume have
adequately similar meanings in the recipient's mind. The biggest
problem is with important words like control, which in PCT are
defined in a very specific way and have a single clear meaning.
Using the right meaning is important enough that some effort must be
taken to get it across.
As a general approach to communication, I certainly agree with this.
As a professor of physics, etc, definitions of words, procedures, etc
have been my primary focus. One thing I try to show my students is
how to discover which word (phrase, etc) has been misunderstood. Too
often I find that they have been misled by attempts to "simplify" the
subject by using poorly chosen examples and analogies.
Such attempts to simplify are, perhaps, especially true for derived
concepts, such as "mass" as in f=m*a. This is not a directly
perceptible property of an object, rather, the definition is: m==f/a.
(I am using "==" to mean "by definition.") One needs, in physics, to
distinguish between "mass," "weight" and "density," which are often
confused by use of unsuitable examples. Force and acceleration are
both perceivable, or defined in perceptual terms. Thus "mass"
becomes, in a sense, a perceptual variable.
A similar procedure is needed in any presentation of ideas to an
audience lacking background in the field. One problem in such a
situation is, "Where to begin?" What language can one assume as the
basis for presentation? This is especially tricky, as you note,
Bill, when the needed words are already familiar to other people, but
use different definitions.
Such a procedure is not a matter of "baby steps" -- rather a matter
of "how to develop a new subject from old, familiar ideas." One must
learn some minimum of arithmetic before algebra, algebra before
analytic geometry, and both before calculus. [NOTE: I have taught
physics both with and without using calculus. Much easier when
calculus concepts and methods are available!]
In the paragraph quoted above, you mention "control." In B:CP's
Glossary, p 283:
"Control: Achievement and maintenance of a preselected perceptual
"state in the controlling system, through actions on the environment
"that also cancel the effects of disturbances."
Contrast this with, B:CP, p 47:
"The subject can be said to control a variable with respect to a
"reference condition, if every disturbance tending to cause a
"deviation from the reference condition calls forth a behavior which
"results in opposition to the disturbance."
And, with my definition, Bob Clark (940704.1355 EDT) Subject:
RKC-MORE PROG CONTROL:
"Control is the production of actions in opposition to the acts of an
"external entity."
I think these three definitions are essentially equivalent, but which
is easier for the newcomer? I am sure my offering can be improved,
but it uses words consistent with general usage. Focussing on the
concept of "control" rather than "the subject" as above, seems more
likely to communicate the concept.
I proceed to define (same post) a "control system" as:
"A system demonstrating such actions is a "Control System.""
And then I point out (same post):
"Such a system must be sensitive to the magnitude of the variable
"being opposed, must have a pre-set magnitude for the variable, and
"must have a means of affecting the external situation."
Since this includes the feedback function (and signal), the reference
signal, and the output function, I probably should have included the
comparator in this group. The quoted statement is very close to a
definition of a "negative feedback control system" with a minimum of
specialized technical terms.
Thus this approach leads rather directly and easily to the elements
of a PCT "control system."
However, that post also included the terms "flow chart" and "system
diagram." I used these terms in accordance with general practice. I
wanted to be understood by people with limited knowledge of
engineering, physics, and math. At the same time, I presumed that
those who are knowledgeable in those fields would have little or no
difficulty with these definitions.
In your post:
The other approach, which I prefer, is to say that we are choosing
a very precise meaning for an existing term because other existing
usages amount to false claims about the phenomenon to which the
terms refer.
Generally, I agree. I am not so sure about "false claims,"
misleading, or confusing, perhaps. "False" may be a bit extreme.
And "claims?"
In any case, when choosing such a precise meaning, it is necessary to
state that meaning with precision, to explain the difference(s) with
the other meaning(s), and the reasons for proposing the new meaning.
It is incumbent on the inventor of the term to elucidate its meaning.
Thus I pointed out the engineer's use of "control system" in
reference to a system used by a person to help that person control
(PCT meaning) the same person's perceptual variables. I hoped that
this might clarify the distinction between engineering terms,
"control system," and PCT usage of "control system."
By the way, _all_ words in the dictionary lead to circularities,
often multiple circularities.
I commented on this observation in my post to Bill Leach, Bob Clark
(940726.1516) Subject: RKC-MORE ABOUT WORDS.
*Definitions, particularly dictionary definitions, are often pictured
*as useless since one definition in words leads to additional
*definitions in more words. For some words this becomes a circular
*situation.
Ultimately, of course, like all definitions, dictionary definitions
must be based on sensory experiences. A quick scan of familiar names
for: sensory experiences, objects, adjectives, adverbs, common nouns,
transitive verbs and the like, demonstrates ways in which words can
be related to sensory experience.
Dictionary definitions are based on observed common usage -- and
common usage is certainly based, ultimately, on sensory experience.
"Taste the chocolate!" "Smell the roses!" "See the dog!" "See the dog
run!"
As to your definition of "flow chart/diagram," I find none in B:CP.
My only source of your definition is in your recent posts. In Bill
Powers (940507.1430 MDT) [--> (940705.1430)], I find;
A flow diagram ... describes how a system behaves.
Later, after much discussion of control systems and their properties,
you present (same post):
To sum up:
A flow chart is a _description_ of the observed relationship between
input variables and output variables.
This statement, together with the several examples you have
presented, suggests that "flow chart" can only be applied to the PCT
type of "control system." Otherwise, we need independent definitions
of "relationship between," "variables," "input" and "output."
As to the TOTE unit, Bill Powers 940727.0715 MDT) Subject:
Definitions; PCT and IT; the argument, you suggest:
I hope that you will not be confused when I say, for example, that
the TOTE unit is a flow chart, not a diagram of a control system.
At this time, B:CP p 17 and p 167 and Bill Powers (940507.1430 MDT)
[--> (940705.1430)] are my only sources of information about TOTE
units. With my definitions, the TOTE unit is certainly not a
"control system." As you point out in the same post, when the unit
"EXITS," it ceases to operate. And it certainly is a "flow chart" in
my terms. A sequence (temporal) of events is stated, so it is a
"flow chart" by my definition.
And you add:
The TOTE unit describes behavior as a flow-chart.
But your definition of "flow-chart_ doesn't seem to apply to the TOTE
unit as you describe it. There is no "input variable" pointed out,
nor is there an "output variable" pointed out. How do you apply your
definition in this case?
As to system diagrams versus flow charts, I see that you are not
willing to change your position, so I guess we will simply go on
using those terms in different ways.
First, the only reason you have offered for me to "change your
position" is your assertion that your definition is, somehow "right."
I think I am entitled to more than your assertion.
Second, I do not think I am confused (at least not yet), but I think
others may benefit from a more complete exposition of your view of
"flow chart" versus the more ordinary view.
Regards, Bob Clark