Control definition

CSGnet Archive CSG1995
1

[From Rick Marken (950208.1015)]

While surfing the Internet waves
I found a most wondrous Home Page
With a "control" definition
Worthy of exhibition
So I'm posting it here for the sage.

CONTROL: Choosing the inputs to a system so as to make the state or outputs
change in (or close to) some desired way. (Arbib).

This is the definition of "control" from Principia Cybernetica Project's
WWW home page.

If Cliff Joslyn is still on the net, perhaps he could tell us 1) why this
definition of "control" (rather than one provided by someone in CSG) was used
and 2) why the Principia Cybernetica Project's home page has no reference
to CSG or CSG-L (or does it)?

Best

Rick

2

Rick, thanks for the plug, and the poem!

To be complete, let me first advertise our URL:

        http://pespmc1.vub.ac.be

All references to "nodes" in this note should be understood to be
URLs relative to that address, e.g. node X.html is really

        http://pespmc1.vub.ac.be/X.html

Now, the definition you found:

CONTROL: Choosing the inputs to a system so as to make the state or outputs
change in (or close to) some desired way. (Arbib).

is at the node ASC/CONTROL.html. This is part of the glossary compiled
by the American Society for Cybernetics, and Stuart Umpleby in
particular. This is THEIR idea of control.

Of course, control is an absolutely central concept for PCP (Principia
Cybernetica Project). If you use the searchable index feature at
Index.ind to search for "control", some fifteen hits come back. One of
these, at the node CONTROL.html, is a small essay by Val Turchin on
the subject. I'd be interested in how coincident you think it is to
PCT.

Another hit is at AUG1992nodes/CONTROL.html, which is part of the node
structure that Turchin, Heylighen and I developed a while back. That
whole sub-web is a bit terse (deliberately), and there we say:

Synopsis: Mutual constraint with asymetric feedback indirectedly
filtered through representations and directly through actions.

That's a bit tough to unpack. But suffice it to say that, yes, I in
particular have brought a large awareness of PCT to PCP. This has
resulted in some significant collaboration between our groups,
actually. In particular, an issue of World Futures which should be
going to press any day now has a wonderful paper by CSG's Pater
Familias, and another by myself wherein I actually use the definition
of control offered by one Richard Marken.

Finally, please see the node RELATED.html#CyberControl, where PCT and
CSG receive a rather prominent mention.

O----------------------------------------------------------------------------->

Cliff Joslyn, NRC Research Associate, Cybernetician at Large, (301) 286-7816
Code 522.3, NASA Goddard Space Flight Center, Greenbelt MD 20771 USA
joslyn@kong.gsfc.nasa.gov http://groucho.gsfc.nasa.gov/joslyn/joslyn.html

V All the world is biscuit shaped. . .

3

[From Oded Maler (950209)]

[Rick Marken (950208.1015)]

While surfing the Internet waves
I found a most wondrous Home Page
With a "control" definition
Worthy of exhibition
So I'm posting it here for the sage.

CONTROL: Choosing the inputs to a system so as to make the state or outputs
change in (or close to) some desired way. (Arbib).

What's wrong here? The "system" in question here is the external world,
so the only complaint you could raise is for not making the implicit
explicit, that is, replacing "state" by "perceived state".

This is the definition of "control" from Principia Cybernetica Project's
WWW home page.

If Cliff Joslyn is still on the net, perhaps he could tell us 1) why this
definition of "control" (rather than one provided by someone in CSG) was used
and

This is the commonly used meaning of the world Control. This is what
an overwhelming majority (i.e. 99.99%) of the people who call
themselves Control theorists do. You might claim that this is not a
fruitful definition for the purpose of studying human behavior, for
the purpose of focusing on the controller's point of view, but you
cannot monopolize the use of the word. It's like protesting against
dictionaries of other languages..

Morale: don't surf outside the CSG-L neighborhood :slight_smile:

(Joel Judd same date:)

Well, the first recorded murder in Judeo-Christian literature occurred
with the jawbone of an as

If "recorded" means written in the bible, Samson's violence against the
Philistines has been preceded by numerous murders of all sorts.

--Oded

4

<[Bill Leach 950210.23:53 EST(EDT)]

[Oded Maler (950209)]

CONTROL: Choosing the inputs to a system so as to make the state or
outputs change in (or close to) some desired way. (Arbib).

What's wrong here? The "system" in question here is the external world,
so the only complaint you could raise is for not making the implicit
explicit, that is, replacing "state" by "perceived state".

What is wrong is that this is not a matter of Rick's or even Bill's
definition of the term "control". The term has had precise technical
meaning for over 60 years. Bill chose to accept that technical meaning
for use in PCT.

That there are people and groups of people that misuse terms like
"control" and "feedback" does not make them "right". "They" may "win" in
the long run but I sure hope not. Language and communications are bad
enough as it is without destroying the meanings for technical terms too.

-bill