To follow: discussion on information, meaning, models, Dretske

I've become involved in a recent discussion on the email list CSG-L
(devoted to Bill Powers control theory, which has significant overlap
with PCP) about the role and nature of information and information
theory in control systems. The pure CSG position is that an active
feedback control system is essentially closed to information, i.e.
that the information flowing THROUGH the loop swamps out any
information coming INTO the loop to such an extent as to make
consideration of such input information essentially innacurate. They
state this as "there is no information about the disturbance in the

Naturally, some people argue with them, which has prompted a
fascinating discussion about the nature of information in feedback
control. As usual, there's lots of confusion about different senses of
the term "information", and partially to help this I posted some
information ( ;-> ) about Fred Dretske's semantic information theory,
which I've been reading about recently. This gets quickly into issues
of information vs. meaning and semiotics in modeling systems, which is
of keen interest to both PCT (Powers' Control Theory) and PCP
(Principia Cybernetica).

The next two messages posted to PRNCYB-L (but not to the other
individuals in the discussion, who've seen them already) will be the
current state of the discussion. The third message will be my latest


Cliff Joslyn, Cybernetician at Large, 327 Spring St #2 Portland ME 04102 USA
Systems Science, SUNY Binghamton NASA Goddard Space Flight Center

V All the world is biscuit shaped. . .