Complexity_science

Complexity science deals with the phenomenon of how novel dynamics and time-stable phenonema arise from chaotic systems and how the right feedback systems can create emergent order. In our case with [Pangaia] and [Hackerspaces], it is to create an ideal environment that encourages and amplifies interesting dynamics using the social context as an novel, non-arbitrary environment along with a voting system to keep everything at that ever-vibrant boundary between order and disorder.

Dynamic systems occur when there is an interaction between energy gradients1. I extend this to also include knowledge and ignorance grandients. Believe it or not, we've got a grand unified theory that will blow your blooming mind (1Various conversations between Stephen Guerin and Mark Janssen at the Sante Fe Complex, via email exchange at Santa Fe, New Mexico.

[Xer0Dynamite]

Conversation between Stephen Guerin and Mark Janssen.1 where we were nailing down terms used in Complexity Studies. (I've added to my side of the conversation, so is slightly edited.) You can ignore this.

DEFINITIONS

[Stephen Guerin:]
> Agent/Entity - agent in agent-based model. (eg, person in epidemic model, mass-based particle in physics, or photon, neuron in cognitive model, etc)

[me:]
Agent: An entity capable of doing work.
Entity: An agent with one or more degrees of freedom removed or constrained.

> Degree of Freedom - possible change in an agent's state. The behavioral repertoire of an agent. Measurable in bits. (eg. rotation, vibration or translation in particle dynamics, left or right party in a voting model, moving to eight neighbor patches in a cellular model)

DoF:  (slight change in wording) the possibility of change (i.e. not necessarily the change itself) in an Agent's state.

I'm not sure DoF can be measured in bits, especially if Constraints are similarly measured with the same unit; I feel conflation occurring, like confusing a CPU process instruction with memory.  If the term is going to be used in both places, then at what place?

> Action - a unit change in unit time of in an agent's degree of freedom. Assumed to be discrete. In physics energy x time. Measured in units of joules-second.

That is from the observer's p.o.v.  From the Agent's point of view, an action is a unit change along a step to a desired goal (so yes discrete).  From this view a gradient is implied.

> Constraint - a decrease in degrees of freedom of an agent. Measurable in bits.

Yes, with reservations mentioned for DoF.

> Interaction - transfer of action between two agents.

That's fine, though because there is directionality, this may need refined. Without this noted directionality, this is just an entropic event.

> Work - constrained transfer of energy. Energy is transferred into degrees of freedom with unequal probability

Sounds good.  The constraint is related (and made) by the intention of the agent (from the social point of view).

> Power - Amount of Work performed during a period of time

Yes, although this is the view from the outside (the view of physics).  From the agent doing the work, power is only gained with success toward the goal -- this power accumulates "reputation" and the ability to do more work.  I.o.w, success towards a goal creates a special cognitive gradient that can be applied elsewhere (assuming the goal was worthy).

> Heat - unconstrained transfer of Energy. Energy is transferred into behavioral degrees of freedom with equal probability

Good, from the physic's point of view, but since you used the word "behavioral" I'll add that from an agent's point of view, work done without a goal is the equivalent of "heat" within the social ecosystem -- [[Why this project is needed. |much like our current economic system]].

> Information - a measure of constraint on an agent as it interacts with > another agent or environment. Measurable in bits.

Sure (from the outside).  From the inside this is simply called "the Constraint".  Information cannot be applied/gleaned without "cheating" or defeating the goal (i.e. collapsing the gradient).

> Energy - action / time

Okay, large E.

> Equilibrium

• - when total constraint in system = 0
• - when energy is equally distributed over degrees of freedoms of agents.

Okay.  From consciousness's point of view this is equal to Death. Thanatos.

> Equipartitioning Theorem - maximum entropy

But why call it a partition?

> Far From Equilibrium - total constrained energy in a system >> 0

No argument. ...Although, with conscious agents, if the energy got "constrained" because it accomplished a goal, then it is in equilibrium and holding onto its earned accomplishment/power. It can only produce new action with a new "higher" goal.

> Entropy  - a measure of evenness of a distribution of action/time over degrees of freedoms of agents. measurable in bits

Sounds good, though there should be an addtional measure of granularity', signifying the amount of energy/(each agent).

> Gradient - A vector field which points in the direction of the greatest rate of increase in action, and whose magnitude is the greatest rate of change. a non-uniform distribution of energy over the degrees of freedom of the agents.

Good. Although, if a field should be plural. In which case, I would call this the invisible landscape that the agents are "exploring".

>Self-organization is the spontaneous process of constraint construction to dissipate a gradient at faster rates than unorganized flows.

Okay, but the controversy I'd add is that a conscious intention (that cannot be observed without effecting the system) is always at work and creates the ongoing dynamic and, in theory, can transcend the 2nd Law of Thermodynamics, by discovering new order not anticipated by the programmer-scientist. That was the power of Mandelbrot's work with the graphical set named after him and, I'd argue, [[meta|the force]] which created life itself.