The Social/Cultural Information Layer

This is my first posting to the fis list, I think, so let me introduce myself.

I am working in the area of information infrastructure to support
collaboration of many people and organizations to conduct very complex
projects. We have a real problem in how we model the behavior in these
systems and the information which supports them. For reasons not of
interest here, probablistic and modal systems are inadequate, and we likely
cannot do as well here as we want unless there is a major breakthrough in
information science for this domain.

I have only been thinking about these issues in an fis way since Vienna, so
please forgive the incompleteness of these thoughts.

First, I understand "information" to mean two different things:

---information(1) is the way of representing the dynamics of systems so
that we can understand them. As such, the abstractions provide basis for
science, the purpose of which is to understand and predict.

---information(2) is the way of explaining the organizational pull on systems.

A main theme of fis as I understand it, is to examine info(2) to shed light
on info(1) rather than the traditional mode of the other way around.
Info(1) requires an observer whereas info(2) does not. The appeal of
info(2) is that one can also exclude the clockmaker from the equation as
well as the clockwatcher and look purely at how the clock organized itself.
In the Biosystems issue, I noted two ideas that follow from this fis
perspective that may be useful.

Idea 1: Vertical Information Flow

The world's information flows are layered. The dynamics of what organizes
systems at the atomic level are somehow fundamentally different from those
at, say, the cellular level, and again different from the social level.
(Forget for a moment how many layers there are and whether the division is
canonical.) In other words, the abstraction space of physics information is
different than that of biomechanics. The latter can, of course, be at least
partially "explained" by particle physics, but some new abstractions are
introduced at the "higher" level, and all abstractions from the lower level
must be remapped.

Yet, it is clear that from an info(2) perspective, some of the
organizational rules of partical physics are at play in how biological
systems organize. The information flow through this "vertical" layer (form
physics to biology) is interesting. We not only have an information dynamic
which organizes quarks into atoms and molecules and thence into galaxies
(so-called "horizontal" information flow), we also have a (presumably
information-motivated) organizational principle that (perhaps creates and)
spans these vertical layers.

Most fis-ers work purely within one layer and then look for analogies that
can be mapped from one horizontal to another. But I think this vertical
flow is much more basic. Conrad particularly impresses me because he is
empirically working at one of these boundaries.

This notion is important when it comes to the social layer. I believe that
we can only go so far in looking at the horizontal mechanics of biosystems
for insights and analogies to inform the (horizontal mechanics of the)
social/cultural layer. Instead, the jumping of the vertical abstraction
layers seems more fundamental, more rewarding to examine.

(Incidentally, I believe symmetry primitives may provide an abstraction
vocabulary which can carry information through the different conceptual
spaces of the various layers. But that is another message
)

Here is the payoff. At the boundary below the social/cultural layer,
info(1) and info(2) become more congruent, because the "users" are the
actors as well. In Vienna, I proposed that this boundary is the most
rewarding for fis attention because of both:

---the potential for a breakthough by using the ideas of linguistics and
messaging (distinctly info(1) concepts) since these only have meaning with
a reflective observer. Matsuno impresses me as having done some good
thinking here. As well, I personally have found liguistic-related ideas to
be the most leverageable (situation and category theories).

---the idea of what this is all about. We want (new insights for) a
_science_. That's an info(1) idea. It requires an observer and a platform
above the vertical boundary below the human/cultural) level.

Moreover, that layer is much more in obvious need of new science, than say
partical physics. This need, and the costs of not having a useful science,
means that the work may be not only more immediately useful and accepted,
but more readily funded.

Bottom line of this idea:

---vertical flow is more promising than the already visible fis preference
for horizontal dynamics;

---for a variety of reasons, the social/cultural layer (and relations to
lower levels) is the most promising area for focus;

---the distinction between info(1) and (2) is important, and probably
leverageable in bringing in mathematical principles of language and
information as tools for exploration

Idea 2: An Extropic Motivator

Within both horizontal and vertical flow, we presume that there are
information-based organizing principles at work. Conventional science
posits two concept spaces for this: the space of the "creator" from whence
the laws at work come (which are immutable) and the space of the phenomena
which (must) follow these laws. The differentiation of concept spaces for
clockworks from clockmaker is the same as between info(1) and (2). Focusing
on (2) allows us to be freed the clockmaker, and the vexing idea of entropy
that follows. (Entropy is a specifically (2) concept.)

An info(2) view allows us to posit an anti-entropic (or extropic)
mechanics. Here, "lost information" instead of being lost, adds to the form
and strength of the law at work. The clock creates itself and the laws
under which it operates by being and operating as a clock.

There's more on this idea. But for now, let's just look at implications for
the social/cultural level. I suppose that the situation is not a simple
layer cake with fundamental particles at the bottom and societies at the
top. Instead, this is a circular collection. The top/bottom layer is what
we've filled with the blunt concept of God.

The leverage here is that we now have a layer above the social/cultural
layer and an outline for an info(3) which are the extropic principles. It
is the analogy of the observer status except the observer is the universe.

Bottom line of this idea:

---information is not only the principle around which systems organize
within laws, but is also the principle around which the universe organizes
those laws;

---once again, examination of the social/cultural layer is the high payoff
area here, this time on the boundary above;

---also again, the nature of abstraction itself is the key idea which could
inform us. But it has to be a recursive concept, since the laws of
abstraction are part of the laws which self-create. The discussions on
cosciousness bear on this.

Comments (I've been periodically dipping into the discussion; please
forgive me for whatever misunderstandings this imperfect student has):

Artigiani: in his first sentences of the kickoff message, he defines the
info(1) and (2) problem. However, for my purposes, VEMs are not sufficient;
they are not abstract principles which can inform a new science, unless he
has something interesting under the thumbnail of "values." I see this as a
slightly better articulation of the higher level than "God." He does,
however, introduce persistence of information outside the layer, which I
think is a vertical process.

Marijuan: I do not understand infopoieis. But I do understand his
suggestion that mechanics of division are more fundamental than those of
unification, and that laws of death (autonecrosis) may be more rewarding to
examine than laws of creation (on which we seem to be stuck). The
partitional calculus seems intriguing and I'd like to hear more about the
language principles invoked.

Stonier: frustrates me because he seems to stick to one layer and be
concerned with horizontal dynamics. However, the idea of examining
granulatity seems rewarding, since that is likely related to verticality.
Interestingly, his term "infon" (at least for my use) is identical to the
same term invented by Barwise and Perry in 1981. But infons do not capture
a notion of state. That is why in Vienna, we introduced the notion of
"acton" which is at once more fundamental and includes a state delimiter.
But I've already noted these are not objective. In fact, they could serve
as the basis for non-realism.

Matsuno: as noted, seems to have done some careful thinking about language
primitives and "grammatology." His notions of tense mirror our info/acton
conversion, and in particular can be seen as a basis of causality.

Best, Ted

_____________
Ted Goranson
Sirius-Beta, Virginia Beach USA
757/426-6704, fax 757/721-0781
Received on Sat Sep 26 17:54:41 1998