[From Rick Marken (981129.1210)]
Me:
It wasn't clear what everything after the "if" meant. But I don't see
how even a generous interpretation of its meaning can save it. It is
simply not true that "all [outputs] will _eventually_ work if you let
the direction of error change determine the next output".
Bruce Gregory 9981129.1250 EDT)--
Would you be any happier (a lot to ask I admit) if I told you that
the "if" means you can start out any way you like as long as you
let the direction of error change determine the next output?
Not really. I would be happier if you told me that some control
system outputs are "right" (they allow control to occur in the
environment in which the system exists) and the rest are not.
Bruce G.
What is the nature of a "disturbance" as far as E. coli is concerned?
Me:
Changes in perceived concentration that are _not_ caused by E.
coli's own actions.
Bruce G.
Well, I have learned something after all. The world is a
disturbance as far as E. coli is concerned. This puts a new
slant on the meaning of disturbance that I had not appreciated
before.
I'm glad you now appreciate this; it's a fundamental fact
about the nature of control systems.
The world is a disturbance for any intentional system.
More accurately, it's a disturbance to (an independent effect on)
a controlled variable (cv). In the simplest case cv = o + d; the
value of a controlled variable (a cv like chemical gradient) is
determined by system outputs (direction of movement, o) and
environmental disturbances (shape of the gradient, d) that
are independent of these outputs.
Interesting way to think of the world, sort of Buddhist is flavor.
All suffering comes from allowing the world to disturb you.
Yes. All suffering (error) comes from failure to prevent disturbances
from moving a controlled aspect of the world (cv) away from it's
reference state.
If you simply match your reference levels to your perceptions,
you will never be disturbed. Nice point.
If you match your reference levels to variations in disturbances
to the perceptual variables that are controlled relative to these
reference levels then you will _always_ be disturbed in the sense
that your perceptions will always vary right along with the
disturbances to these perceptions; there is no control at all.
So there is really no reason to be a control system (with reference
specifications for input) if this is how you deal with the
disturbances to those inputs. It's cheaper and easier to just be
dead.
Me:
only the outputs that oppose the net effects of disturbances
to a controlled variable are "right"; all other outputs are
"wrong".
Bruce G.
With my new found understanding of "disturbance," I can hardly
disagree.
Well, that's great. I sure hope you're not being facetious.
Best
Rick
ยทยทยท
--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken/