[From Rick Marken (920513 09:00)]
Dag Forssell (920513-2)] says:
That there is no one "reasonable"
setting for anything at any level may be quite valid. Is that a reason
to never discuss any suggested settings at the principle level?
The change in the height of a column of water depends on the volume, not
the mass, of an object that is placed in the water. Is that a reason
never to discuss ways to bring the water level to a particular height by
suggesting settings for the mass of the object to be added? I think the
answer to your question is another question; What do you consider to
be a waste of time?
Ed Ford (920513.12:12) says:
We all of us have a variety of perceptions of how things ought to
be. This is found at our systems concepts level.
There are called references; they define what we ought to be perceiving.
These exist at ALL levels in the model -- not just the system concept
level. So we have references for how much pressure to feel on our fingers
and how much like a fist our hand configuration should be in and how
rapidly our hand configuration should be changing (ESPECIALLY WHEN CAPTURING
SATELLITES __ WHAT A GREAT PIECE OF CONTROL!).
It seems to me we are setting standards for ourselves and in
cooperation with others (such as in a home, a community, or within an
organization) all the time.
This is the crux, I think. We care about 'standards' because they often
determine lower level actions that might influence the variables controlled
by other people. I think Greg picked up on this in his last post:
Greg Williams (920514) suggests
ethical CONTEXTUALISM (rather
than relativism or absolutism) might be the best model.
An ethical contextualism offers a middle road, recognizing that within a
particular context (sometimes quite broadly defined -- i.e., we're all human),
there are certain standards which DO "work," but also recognizing that if the
context is different, those standards might cease to "work."
Yes; and the important context is other control systems. My spreadsheet
model has to be expanded to two (or more) simultaneous hierarchical
systems working in the same envronment of numbers (degrees of freedom).
I think you would find that these models would quickly run
into conflict if their HIGHER LEVEL (level three) systems were controlling
for the same variables relative to different reference levels. There
would always be less conflict at the lower levels because the references
for those levels can be changed by the higher level systems that see that
there are lower order errors.
Actually, I think I will do this modeling effort; but my intuition is that
the only way to solve the problem of multiple interacting control systems,
operating in the same environment, is to align the references for the
highest order systems that are controlling the same perceptual variables.
I wonder if the solution would be found automatically (through reorganization)
or whether there needs to be a system that actually perceives that there
is conflict and looks for a cooperative solution. I think the former might
work.
So I think it's possible that alignment of higher order references may be
a natural consequence of being reorganizatble, hierarchical control systems.
Of course, the values at which these systems get aligned are not necessarily
determined -- just as long as they are aligned. I think this is why we
see such remarkable differences in cultures (as Greg noted -- we ought
to get some anthropologists in on this). But there are remarkable differences
between cultures in terms of system concepts like marriage (polyandry,
monogamy, polygamy, etc) -- and they all work; apparently because everyone
buys into that reference. Of course, once pressures lead individuals to
shift references (our society seems tacitly moving from monogamy to serial
monogamy -- largely as a result of an unpredictable disturbance; people
are living longer) conflicts between control systems increase -- as would
be expected until the group is able to "realign".
By the way -- all you "standards" fans; fear not. It's highly unlikely that
any society will align on a system concept that demands really "bad"
standards like murder. There are standards that are self correcting
(the people who aligned on the system concept that demands "murder"
would be quickly eliminated from the pool of control systems). Note,
by the way, that most societies have aligned on system concepts that make
it perfectly ok to murder the members of other societies. But that's
getting into more substance than I think is appropriate -- back to models.
It's all contextual.
Ok, I'll buy it.
How about another:
It's all interacting control systems
Best regards
Rick
···
**************************************************************
Richard S. Marken USMail: 10459 Holman Ave
The Aerospace Corporation Los Angeles, CA 90024
E-mail: marken@aero.org
(310) 336-6214 (day)
(310) 474-0313 (evening)