( [From: Bruce Nevin (Wed 930915 14:09:07 EDT) ) --
How would you model:
Volunteering
Taking responsibility
Appointing
Disappointing
···
-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-
It strikes me that a left-to-right orientation of the canonical PCT
diagrams is easier to draw:
:
+------>:----+
e> : |
r +-+-+ : |
+------>| | : |
> +---+ : | d
+-+-+ ^ : |<----
---->| | p| : |
+---+ | : |
^ | : |
p> > : |
+---------+-------:<---+
:
CONTROL SYSTEM : ENVIRONMENT
:
Doing so might be sensible at least when talking with engineering control
theorists, for whom superficial differences of notation may present a
distraction from more essential points.
-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-
Tom Bourbon [930914.1302]
And you think you're slow! I think I just got it. (For one "it,"
anyway.) Indistinguishable are: two persons, two hands of one person,
two degrees of freedom for one hand of one person, the combination of a
software model with a person. So two control systems are two control
systems are two control systems, no matter how they are implemented or
"housed." Having them within the same person does not make them less
autonomous; or is it that having one within a person and the other not
within that same person fails to make them more autonomous with respect
to one another? Social control fans assume the latter, but all we can in
truth say is that they are, as you say, "functionally equivalent."
every time I make the diagrams, I end
up with representtaions of whole persons who adopt reference perceptions
that include "acting like this function in a diagram," and that is not the
same as *being* that function in a system.
Maybe that's all they are doing. Maybe that's how you model
"volunteering."
-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-
Now model volunteering while believing that you have no choice.
I suspect social control is only this. Gene Sharp, at Harvard, comes to
something like this view in a long series of books studying what has
worked and what has not in non-violent political actions.
-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-
:
+------>:----+
e> : |
r +-+-+ : |
+------>| 2 | : |
> +---+ : | d
r +-+-+ ^ : |<----
---->| 1 | p| : |
+---+ | : |
^ | : |
p> > : |
+---------+-------:<---+
:
CONTROL SYSTEM : ENVIRONMENT
:
CS 2 receives its reference signal r from the error output of CS 1.
:
+------>:----+
e> : |
r +-+-+ : |
+------>| 2 | : |
> +---+ : | d
+-+-+ ^ : |<----
---->| 1 | p| : |
+---+ | : |
^ | : |
> > : |
+---------+-------:<---+
:
CONTROL SYSTEM A : ENVIRONMENT
:
:
+----+..:....+
e> > : |
r +-+-+ | : |
+------>| 2 | | : |
> +---+ | : | d
r +-+-+ ^ | : |<----
---->| 1 | p| | : |
+---+ | | : |
^ | | : |
p> > > : |
+---------+----+..:<---+
:
CONTROL SYSTEM B : ENVIRONMENT
:
CS A and CS B are both capable of controlling p, but B trusts A to
control it while B attends to other controlled perceptions. B "controls"
p in the imagination loop. B may be alerted to A's failure to control p
as B would have done by attending to p periodically, or by the relation
of p to some other perception that B is controlling. A controls p with
respect to r as determined internally to A. But while B is teaching A or
supervising A (they are the same thing, if managers would but realize
it), B provides communicative "feedback" (socalled) on how A is doing
with respect to r as determined internally to B. A "catches on" or not.
Communication per Martin's layered protocol scheme.
:
+------>:----+
e> : |
r +-+-+ : |
+------>| 2 | : |
> +---+ : | d
+-+-+ ^ : |<----
+---->| 1 | p| : |
> +---+ | : |
> ^ | : |
> > > : |
> +---------+-------:<---+
+-+-+ :
----->| 0 | +----------->:-------+
+-+-+ | : |
^ +-+-+ : |
> --->| LP| : | +-----+
> +---+ : | | |
> ^ : |<--+CS B |
> > : | | |
> > : | +-----+
+------------+------------:<------+
:
:
:
CONTROL SYSTEM A : ENVIRONMENT
:
Here, LP in a box abbreviates the control hierarchy for layered-protocol
communication using various means such as language. The r input to the
LP function has to do (in this case) with a perception that B is
satisfied with A's control of p. Note that LP and 1 and 2 (or the system
of 1 over 2) are mutually autonomous with respect to explicit signals in
the model. What links them is the affect associated with error whatever
its origin, perhaps. (There are those who would claim that this is what
links otherwise autonomous control systems such as humans--how do you
model empathy?--but I will assume this is a nonverbal aspect of the LP
box.) And of course they are linked through the environment, in
particular the CEV (not shown) represented by p in A and by a presumably
similar p in B.
-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-
every time I make the diagrams, I end
up with representations of whole persons who adopt reference perceptions
that include "acting like this function in a diagram," and that is not the
same as *being* that function in a system.
The function that controls p in A is in fact the function that controls p
in the team of A and B. When B gives A some correction or guidance, B is
not in fact controlling p. Perhaps we should be able to model a way for
B to "shadow" A's control of p, with no effector output into the
environment (low gain?), in a way that is distinguishable from the
imagination loop.
Because the corresponding control loop within B is not controlling p,
unless (by that fact) B overrides A and usurps control of p. In that
case, the social role or responsibility shifts from one person to
another, just as it does when A's shift ends and another person comes on
the job. Then the function that controls p in the team is in fact the
function that controls p within the new team member, the person replacing
B on the team. The organization is (arguably) a control system whose
constituent control systems may change location. A person suffers a mild
stroke, and the replacement ECSs are a bit green at first, but after a
while they catch on. Reorganization? Learning? Compensation?
Recruitment? Conscription? Volunteering? On the job training? Are these
so easily distinguishable?
Bruce
bn@bbn.com (still)