We talk about reorganization as consisting of "random" trials of altered
reference signals. But it doesn't seem to me that the reference values to
be tried can be picked (?) randomly in the sense the word has in
probability theory (every alternative equally likely). How about
arbitrary, opportunistic, fitful, volatile, protean, transilient, erratic,
unsystematic, ....?
Hi, Phil--
We talk about reorganization as consisting of "random" trials of altered
reference signals. But it doesn't seem to me that the reference values to
be tried can be picked (?) randomly in the sense the word has in
probability theory (every alternative equally likely). How about
arbitrary, opportunistic, fitful, volatile, protean, transilient, erratic,
unsystematic, ....?
I would pick "unsystematic." When I think of random reorganization, I'm not
thinking so much of equiprobable alternatives, but as choices that are
without rational foundation. If you have a rational foundation for choosing
a new reference condition, you're just using a control system that already
exists at the logic level or the principle level or some such. But if
you're using an already-existing organization, you're not reorganizing,
you're just controlling. Reorganization is what happens when no
already-existing system can solve your problem. Therefore it must be random
in the sense of not using any learned system, method, principle, algorithm,
and so on. If a learned system could solve the problem it would have done
so already.
Bill
[From Chris Cherpas (990511.1020 PT)]
Philip Runkel wrote:
We talk about reorganization as consisting of "random" trials of altered
reference signals. But it doesn't seem to me that the reference values to
be tried can be picked (?) randomly in the sense the word has in
probability theory (every alternative equally likely).
Alternatives don't have to be equally likely for an
outcome to be random. Would you prefer reorganziation
be described as a "stochastic process?"
Best regards,
cc
No, "unsystematic" or "arbitrary" is good enough for me. I just wanted to
be clear that the necessary idea was
going-beyond-available-control-systems, not randomness.
Thanks for comment.
···
On Tue, 11 May 1999, Chris Cherpas wrote:
[From Chris Cherpas (990511.1020 PT)]
Would you prefer reorganziation
be described as a "stochastic process?"
[From Chris Cherpas (990516.0945 PT)]
Philip Runkel wrote:
... "unsystematic" or "arbitrary" is good enough for me.
I just wanted to be clear that the necessary idea was
going-beyond-available-control-systems, not randomness.
Sounds good to me. I think of randomness _as_
unsystematic or arbitrary, myself, rather than
meaning a uniform distribution of elementary
events in a sample space, per se, or any other
particular distribution.
In addition to going-beyond-available-control-systems,
the notion of randomness includes, for me, that only
certain possibilities exist for modifying or forming
new control systems, not that "anything goes" -- e.g.,
divine intervention.
Best regards,
cc