Steve,
Good, so if we’re “only to create a similarly
huge number of unconsidered regions” by acknowledging that some systems
are both highly systematic and completely out of control, let’s pick it
apart a little. It is indeed sad that we seem to need such dramatic
demonstrations as having our whole economic world vaporize to get our attention.
The Missouri mule problem I guess, a little something to spark our attention.
My take, for many years, is that standard procedure is the problem,
systematically transferring earned incomes to unearned incomes, by accumulative
%’s. In a physical world that’s the only cause needed.
So I think that’s what we’re seeing snap back this week, the elastic
links between limitless and limited things, as I and others have said would be
the direct consequence of continuing to use that procedure while waiting to see
what would happened. After a rubber band has snapped there’s
less one can do, of course, but still worth having strategies reflect an
understanding of the problem.
I think there are several kinds of “uncontrolled
systems”. Some appear uncontrolled but actually represent continuing
unaltered regular processes, like random walks, chaotic systems, bifurcation
and ABM systems where all the parts follow the same unchanging rules. That
sets a kind of odd partition, as these all seem to be “logical systems”.
The ones that don’t have rules to follow, and display systematic
behaviors that come and go are generally found to be “individual physical
systems”. Of course that’s sort of begging the
question, since it’s saying that all mathematical control is ‘imaginary’
and has to work through individual physical systems that are uncontrolled, but
that’s one of the weird walls (categorical differences) I think you naturally
run into.
One of the most interesting groups of uncontrolled phenomena
are the temporary divergent processes, maybe looking like a kind of ‘tunneling’
through potential barriers, that may be rather quick but you can often catch a
glimpse of. Energy transfer by fluid convection begins like
that, as when warm air is under cool air and needs to have a ‘hole’
in the barrier layers above develop to make the energy flow channel.
The little “run-aways” that start that process don’t seem to
have precedents or to be responding to any influence from what will disrupt
them later, but to propagate freely for a short period. That
separation between things that will collide but can’t respond to approaching
collision till it happens… is a key characteristic. Another more
personal example would be your own quick smile in waving to a friend that
carries no pretense of the grimace, when in your exuberance, your hand
accidentally clips the person you’re sitting with… I did that
this week!! Those kinds of systems that are part of
“happening” wouldn’t work at all the same way if there wasn’t
space and time gaps between them, their beginnings, ends and collisions,
etc. Whether predictable or not, individual events often start with
events that have no past and then as they develop, shuttle back and forth
between being divergent from everything around them and then achieving behavior
quite close to some equilibrium. That when they run into
things that have no past for them they are disrupted and change form in locally
emergent ways puts the same big “?” at the end of the chain too.
It’s not that one couldn’t perhaps develop a
‘God’s eye view’ of local circumstances and devise some
equation that could imitate recorded measures, and get a reasonable single circumstance
emulation. Modeling artificially controlled conditions has worked extremely
well for lots of things. The larger operational definition of ‘uncontrolled’
then is all the stuff which that method and interest in control hasn’t
worked for, and then throw in all interests other than control we might have
asked about. The trick for exploring the divergent
processes in my view is that the parts that are interacting don’t have
the option to know the unknown, nor a place to store such a formula, or any ways
to follow one. They have to act without knowing what
they’re going to run into. That makes the theory of endless compound
multiplying economies, guaranteed to never run into anything, seem more
ironic than any irony scale could possibly measure. Yet it “made
perfect sense” to so very many!
The most useful identifier I found was that the continuity
of events requires every systemic change in behavior to begin with a
“little bang”, a divergent eruption of behaviors having no direct
precedent. Where those “little bangs” are evident or implied
the uncontrolled systems they initiate I call “independent”,
in that their beginnings have no precedent causes that the later developed
system could have had any information about. Too bad about the
several hundred trillion we blew this week… but maybe it’ll turn
out to be well worth it. It’s a little like getting our
first peek over the edge of the teacup we thought was the whole universe, full
of strange delights.
I think that category of uncontrolled individual ‘happenings’
of conserved change is a big category, and my methods for investigating them is
probably of general use and needing work. There might well be other ways
to explore them and other kinds of phenomena that can be turned into windows
into nature and either fun or useful ways to explore them. Any
ideas for what part of the spectrum you’d find interesting to poke into?
Phil
From: [hidden email]
[mailto:[hidden email]] On Behalf Of Steve Smith
Sent: Thursday, October 09, 2008 3:54 PM
To: Aku; The Friday Morning Applied Complexity Coffee Group
Subject: [FRIAM] Self-awareness and blind spots Was: Self-awareness
Phil -
Spot on Phil. I'm CC:ing a friend (Aku) with whom I often discuss this
point (thus I'm leaving the cruft at the bottom).
Science's biggest failing (perhaps) is it's (natural) blind spots.
When I came to LANL in 1981, the Center for Non-Linear Studies was pretty new
and for the most part there was little, if any, study of non-linear systems
going on in the world. This was mainly because of a lack of tools to work
with nonlinear systems. Once (most) scientists got over their fear
of computation, many more complex systems could be studied than before.
Like the man looking for his lost keys under the streetlamp even though he
dropped them a block away "because the light is better here".
"... because as we know, there are known knowns; there are things we
know we know. We also know there are known unknowns; that is to say we know
there are some things we do not know. But there are also unknown unknowns --
the ones we don't know we don't know." - Rummy
Rummy is the new Rumi?
The emerging work in Science studying itself (primarily through studying
Citation Networks) offers some hope that we can begin to fill in some of the
blind spots. Our own Marko Rodriguez & friends, for example: http://www2007.org/poster860.php
That is not to say (as you seem to here) that we haven't just filled out a huge
amount of unexplored territory only to create a similarly huge number of
unconsidered regions within that territory.
I'm not completely up on your view on this topic but there also seems to be a
theme regarding "far from equilibrium systems"? Is that what
you mean by "uncontrolled systems"?
carry on!
- Steve
Steve,
Well, might you also say science is self-organized to be 'robustly' avoiding
the subject of uncontrolled systems too??