[from Mary Powers 9312.15]
Osmo:
Your reply to me, in which you rejected practically everything I
said, is exactly the sort of response PCTers have had from
journals, and why publication has been difficult. Reviewers have
typically rejected PCT papers because the papers do not conform
to their image of how living systems are organized (i.e. control
of perception vs, control of output). They never even get to the
point of evaluating the quality of the experiments presented,
which yield results that are orders of magnitude more precise
than anything currently accepted. What gets rejected is the
concept, not the content, which in our view is not a legitimate
function of peer review.
The main problem with your answers is that they are rather flip
and superficial. You seem to be more interested in tossing off
the snappy remark than you are in thinking about what I said. You
seem to believe that you know all you need to know about control
theory, and if I said something that contradicts what you know,
then of course I must be wrong.
For instance: I said "a command from outside is not a reference
signal, it is a perception." This has nothing to do with home
thermostats. You and your neighbors set your thermostats from
outside, but a LIVING control system's reference signal cannot be
got at the same way. If you are chilly at 20 Celsius and your
neighbor is not, this is because your internal reference signal,
governing your metabolism, is different from his, and no one
outside of you can alter that. You can alter your own reference
signal, by exercise, or by running a fever, but no one can make
it happen except yourself.
You say: "I can put the reference damn well into the heads of the
30 recruits in a platoon and they can accomplish my reference."
This is how it seems to you. We are suggesting that you try
looking at it a different way. PCTers look at it a different way,
and find that doing so is productive and exciting. But you won't
discover that unless you try it.
What you are doing, you know, is first warping what people have
said about PCT to fit what you already believe. Then, having done
that, you say "there's nothing new about it." Well, of course
there isn't, after you've turned it into your version of it.
I said: The feedback is from the current state of the desired
perception, not action. Actions vary in order to produce desired
outcomes."
You say:"To me this is just the same thing being put the other
way around. It does not make any difference on the general
level."
What? Perceptions vary to produce actions? Perceptions are
maintained in or brought to a desired state by means of actions.
But the actions used are not controlled - it is the perception of
their effects that is controlled.
You say: "If you admit, let's say that the perceptions are the
outputs of a control system and we are happy all again. In a
closed loop, at a general level, it does not matter how you call
the branches."
No - we can call the branches Mary and Osmo or anything we want.
But we do NOT call perceptions the outputs of a system, we call
them inputs. It is inputs that are controlled. They are
controlled with reference to the desired perception, that is, the
reference signal. Outputs are what we use to control or change
perceptions so that they more nearly match the reference signal.
Since, so far, you do not understand this, then you do not
understand what is new about PCT.
Control engineers are used to designing systems that do something
in particular. What they do not notice is that what they really
design the system to do is to have an effect that the engineer
wants to see. That effect is a perception of the engineer's. His
goal is to perceive the system accomplishing what he wants it to
do. And he can build it any number of ways, and it can work any
number of ways, so long as the perceived result is what he wants.
As living systems, we grow our own control systems. We are both
system and engineer. We have goals, internal to ourselves, and we
move our bodies around in order to make what we perceive of the
world more like what we want. If we can't do something one way,
we try another. The outputs don't matter to us. Perceptions do.
Perceptions are what we control.
Think about it! 
Mary P.