a Duck by any other name is not a goose

CSGnet Archive CSG1996
1

[From Bill Powers (960201.1415 MST_]

Hans Blom, 960201 --

     So you are serious about the name change after all? This is really
     confusing. I thought -- "knew" -- that you were joking. I certainly
     was. Do you really think that a simple interchange of words can
     solve any of our problems? Would a rose by any other name be
     something else?

I think this (temporary, I hope) interchange of words is going to have
some interesting results. Rick Marken said it well: the problem is not
that we are all just using different words for the same thing. To use my
own takeoff on Rick's example, the problem is that I am talking about a
goose and others are talking about a duck, but the others want to use
the word "goose" for anything with feathers -- ducks, eagles, and even
pillows.

To repeat:

Let Y be some variable that is a determinate function of x1, x2, ... xn.

Y = f(x1 .. xn)

The basic indicator that phenomenon X (my next choice if "spontefaction"
is rejected) is taking place is that a change in one or more variables
x2 .. xn results in a change in x1 such that the value of Y is
maintained close to a specific value Y0. Of course you could interchange
the x variables; what matters is that one of them changes to cancel the
effects on Y of changes in any or all of the others.

There are some unusual circumstances where this effect might occur
without any X-ing system producing it. For example, x1 might just by
chance vary so that it happens to cancel the combined effect of all the
other x's on Y. There might be a link between each of the x's and x1 so
that each one directly causes the appropriate change in x1. Or something
else might be acting on Y to keep it close to the value Y0 (for example,
Y might be physically nailed in place so it can't change). In the latter
case, of course, the expression for the function f is wrong; there is
another variable on which Y depends that isn't listed.

I assume that sufficient additional conditions are specified to rule out
each alterative, leaving only the one explanation for the X phenomenon:
that there is a system sensing the state of Y, comparing it with some
standard value Y0, and producing an action based on Y - Y0 that causes
x1 to vary in the required way.

That is the phenomenon I used to call control, meaning only that and
nothing else. But as Bruce has demonstrated, the EAB use of "control"
allows us to designate any variable among the x's as having "control"
over Y, without introducing any second relationship. You have spoken of
"open loop control" and other usages of the word that differ from my
definition. As long as such usages persist, I can't say "control" and
rely on its being heard to mean what I intend it to mean. In fact, I can
pretty much rely on being misunderstood, and being told that I have no
right to dictate how a word is to be used.

     We will have to live with our misunderstandings, I suppose, regret-
     table as it may be.

I refuse to do that. I would rather not communicate than continue to be
misunderstood. You can use any word you like to refer to this phenomenon
that I am talking about, as long as we agree that this word will be used
to denote that phenomenon and NOTHING ELSE. If you want to talk about
something else, find a different word.

···

-----------------------------------------------------------------------
Bruce Abbott (960201.1020 EST) --

     Has the spark gone out of your romance?

Not tonight, Bruce, I have a headache.

How about a comment on your use of control, influence, and determine as
if they had equivalent meanings?
-----------------------------------------------------------------------
Best to all,

Bill P.

2

[Hans Blom, 960202d]

(Bill Powers (960201.1415 MST_)

To use my own takeoff on Rick's example, the problem is that I am
talking about a goose and others are talking about a duck, but the
others want to use the word "goose" for anything with feathers --
ducks, eagles, and even pillows.

To non-experts, ducks and geese are sometimes hard to discriminate.
To experts, too, if they're looking from a great distance, in a fog,
or if they are in a hurry. Great example! This is much better than
Rick's roses and thorns which are too unsimilar. To use your example:
what's wrong with "control" to describe the class of birds, "feedback
control" to describe the geese, and, if I may be so presumptuous,
"adaptive control" to describe the ducks?

Y = f(x1 .. xn)
The basic indicator that phenomenon X (my next choice if
"spontefaction" is rejected) is taking place is that a change in one
or more variables x2 .. xn results in a change in x1 such that the
value of Y is maintained close to a specific value Y0.

Not good enough: "close to" is unspecified and subjective. We need a
hard criterium. Maybe good enough would be: there is a significant
negative correlation between one or more of the variables x2 .. xn
and x1. There are pretty good tests for significance. But I don't
think you mean that.

What if the significance is marginal? Is that still "control"? It is
more and more my conviction that it is _quality_ of realizing an
objective that is the important central concept that we need, what-
ever the mechanism that enables us to realize the objective.

Remember that in many cases a control system is used where we would
actually prefer a "stiff" linkage, such as in the ailleron control
system of an airplane, if such were only possible. Complex mechanical
linkages, however, have too much friction and would require more
power than a human can provide. Yet, the stiff (but frictionless)
linkage remains the ideal which the control system has to model as
accurately as possible. A car's power steering and power braking come
to mind as additional examples. In all these cases, a control system
is employed where we would actually prefer a "stimulus-response" or
"input-output" system, if such were only possible. There is nothing
advantageous to a feedback control system per se.

Or something else might be acting on Y to keep it close to the value
Y0 (for example, Y might be physically nailed in place so it can't
change). In the latter case, of course, the expression for the
function f is wrong; there is another variable on which Y depends
that isn't listed.

How can you be certain that f is correct? Introduction of a dummy
variable ("disturbance") works, but is hardly satisfying. The "world"
is as it is and knows not of disturbances; using the term "disturb-
ances" just shows that we cannot, don't have the time to, or are just
too lazy to be complete in our description of the process. As my old
measurement theory professor kept saying: "One man's noise is another
man's signal".

That is the phenomenon I used to call control, meaning only that and
nothing else. But as Bruce has demonstrated, the EAB use of
"control" allows us to designate any variable among the x's as
having "control" over Y, without introducing any second
relationship.

As long as PCT works with unexplainable dummies ("disturbances"),
control in the PCT sense will be a gradual notion, too. As long as we
talk about "disturbances" that "influence" perceptions, we must
recognize that actions cannot fully "determine" perceptions. The
"close to" doesn't help to make things clearer.

We will have to live with our misunderstandings, I suppose, regret-
table as it may be.

I refuse to do that. I would rather not communicate than continue to
be misunderstood.

Then communicate differently, in such a way that misunderstandings
cannot arise. Words ("interpretations") introduce misunderstandings,
formulas don't, or to a much lesser extent. Learn a lesson from
quantum mechanics: all quantum physicists agree on the same set of
formulas, although they differ greatly in their interpretation of
what those formulas _mean_. The same with control, or whatever you
want to call it: formulas are clear (although they may be profound),
words are not (and the profoundest of words are the least clear).

You can use any word you like to refer to this phenomenon that I am
talking about, as long as we agree that this word will be used to
denote that phenomenon and NOTHING ELSE. If you want to talk about
something else, find a different word.

Therefore I don't think that this solution works. Find a language
that is clearer than words.

Greetings,

Hans