Thanks that is helpful. Yes it appears we are more or less talking about the same thing. Information entropy or Shannon entropy is simply a way to emperically measure and quantify what you are calling emergence. Here are a couple of papers on this if you are interested in reading more.

Measuring Emergence, Self-organization, and Complexity Based on Shannon Entropy
http://journal-cdn.frontiersin.org/article/244727/files/pubmed-zip/versions/1/pdf



Information Entropy As a Basic Building Block of Complexity Theory
http://www.mdpi.com/1099-4300/15/9/3396/pdf



 

Emergent properties can be predicted, sometime they are not entropic

But entropy is only mesurable against expected behavior from a constructed system to measure how the data fit the theory behind the design of the system.

This connection between emergent property & entropy works for properties emerging from a designed/constructed system, not for measuring "natural" behavior out of the context of a fabricated system. It's only entropy if it's measured as a difference with predicted outcome.

Source: http://www.coolpage.com/

stopped reading there.

The market will adjust just like it always has, just google how many market panics there has been since centralized banking alone as well as the shifting of the job market and labor force in regards to employment and implementation of new technologies. What happened during the industrial revolution when a huge % of the labor force shifted from agrarian to industrial? Or when the industrial companies of the U.S all got outsourced to 3rd world countries for cheap later? The market and labor force adjusted again and we saw a huge rise in the service industry. Or when now IT and newer/lower-education level medical/care-taker jobs are being a larger % of the labor force? Robots will never totally replace humans and vice versa. It isn't a competition and never has been. Can anyone list any SINGLE precedent in which a fundamental new technology "devastated" ANY labor force much less what the biggest fluctuation was? People, the market, supply, demand, and the education/skill needed for the labor force at large is always adjusting within any dynamic system such as a labor market.

More technological advancement has ALWAYS meant higher standard of living in the longer run. The cotton gin invention didn't outlaw slaves but made more people realize it wasn't necessary and more of a burden and "way of life" than an actual long-term feasible commodity. Every labor advancement from the assembly line to the internet has made it easier for human beings and we've adjusted what our duties are and the duties that are no long necessary because of technology in every scenario.


This was written by some crank on some obscure website "projecting" something that will happen in 2033, 15 years down the line. Who could have predicted bitcoins increase to peak 3000 even 3 months ago?

You must understand that all of our Freedom is an illusion. We use crypto currency today and believe that because of it we are not available for decentralization and free in our choice. But this is not so. If you completely understand all the subtleties of Bitcoin structure and how it all was created, then I would not be surprised that in this question there were the comma structures that control everything and everywhere.

I am actually somewhat sympathetic to this position. Ideally it would be Anonymint here defending his definitions but he is boycotting the forum at the moment after getting his most recent incarnation banned so I will do my best to defend his thesis in his absence.


Let's dive into the definitions and see where that takes us. From the papers I linked above:.

Emergence can be understood as new global patterns which are not present in the system’s components.

Self-organization, in its most general form, can be seen as a reduction of entropy. Self-organization is the complement of Emergence and a metric of order and regularity.

Complexity comes from the Latin plexus, which means inter-woven. something complex is difficult to separate. Complexity represents a balance between change (Emergence) and regularity (Self-organization), which allows systems to adapt in a robust fashion. Regularity ensures that information survives, while change allows the exploration of new possibilities, essential for adaptability. In this sense, complexity can also be used to characterize living systems or artificial adaptive systems, especially when comparing their complexity with that of their environment. More precisely, complexity describes a system’s behavior in terms of the average uncertainty produced by emergent and regular global patterns as described by its probability distribution.

So where does entropy come in?

Information entropy is a deterministic complexity measure, since it quantifies the degree of randomness. If we can measure degree of randomness we also quantify emergence with some limitations

As you said entropy is only measurable against expected behavior in a constructed system. The complexity of different phenomena can be calculated using entropy-based measures. However, to obtain meaningful results, we must first determine the adequate function to be employed for the problem. In the case of "natural" behavior out of the context of a human fabricated system this becomes problematic as we don't know the underlying function.

I believe it was Gödel who said, the world is either a perfect order of God, or chaos. The difference is in the belief that infinity comes before entropy.

If we go with infinity then we can assume that some underlying function exists for all systems including "natural" ones. Lacking an understanding of the system we may not be able to measure the the entropy and the associated emergence but we can assume the relationship persists outside of our knowledge.

What you call the process of individuation is not just about maximizing emergence aka maximizing entropy. The "process of individuation" is the maximization of emergence while maintaining overall self-organization. It is the long term maximization of the complexity of the system.

Well the concept of entropy is mostly relevant in the context of engineering, where one build a system, and has to consider that what matter in term of right or wrong is that the machine work as planned wanted by the designer, and in this perspective, the entropy is always something unwanted, or defined negatively as a divergeance from the expectedwanted result.

In the overall i'm not big fan of newton theories lol I prefer liebniz =)

Newton is ok if really limited to experimental physics, but the second law of thermodynamics already it show where it want to get at, and meaning that what is to be considered as good in environment is things that works as planned by a designer, and 'chaos' is seen by definition as something to be avoided deconsidered, or as some kind of waste of energy that make the system not as efficient as it should be to full fill its purpose.

There are the social theories with social darwinism who say that even so before it was fitting to law of nature that was driving evolution, but then since maybe rome or so, it's more social selection, and how one will fit into a culture or another that will be driving force of evolution, or even now how he fits into economic agenda, which can give this sort of sense of 'ideal individual' in the context of social or economic development, and that society has to shape it's citizen to fit this ideal, and like if human being were just passively waiting to be programmed to exist as a person, and anything that popup from a person not being programmed is seen as entropy, as seeing individuals as only target for social engineering.

This is because we have until very recent focused our engineering efforts on creating predictable or "dumb" devices.

In the context of the discussion upthread the goal has been to accomplish a fixed task and then maximise the self-organisation of the system ideally driving emergence to zero aka minimising informational entropy. In this context entropy represents loss or misdirected efforts.

However, if we want an adaptive machine capable of responding to unanticipated environmental changes or improving over time then we need a component of emergence and thus Shannon entropy.

This can be seen when looking at one of the more famous new machines Google's Go playing machine AlphaGo. This machine must respond to unpredictable responses from opponents and still win in a game that is to complex to simply play via simple brute force.

Christopher Burger who has a Ph.D in machine learning wrote this interesting analysis of how AlphaGo works.

https://www.tastehit.com/blog/google-deepmind-alphago-how-it-works/

IadixDev I would actually agree with your description of the universe above but would also argue that it is incomplete as it focuses only on self-organisation and neglects the other aspects of complexity. This is a similar objection to the one you raised against the term entropy.  

I take the position that the entire universe is in a process of ever increasing complexity and there is not one particle or quanta in the whole thing that is not participating in this growing complexity.

Anonymint the author of the essays linked in the opening post is a self described anarchist and focuses on emergence, entropy, and freedom. You seem to view the world more as a process of self-organisation.

I believe both of these conceptions can be brought into harmony under the broader umbrella of complexity.

Complex systems exhibit four characteristics:
–  Self-organization
–  Non-linearity
–  Order/Chaos Dynamic
–  Emergence

Informational entropy provides a way to empirically measure emergence but emergence is only one aspect of complexity. Self-organization can be looked at as a process that actually reduces entropy yet it undeniably also increases complexity.

The chaos in this context is a observation of system dynamics. Systems exist on a spectrum ranging from equilibrium to chaos. A system in equilibrium does not have the internal dynamics to enable it to respond to its environment and will slowly (or quickly) die. A system in chaos ceases to function as a system. A system on the edge of chaos will exhibit maximum variety and creativity, leading to new possibilities. The field of complexity analysis is new and still in its infancy.


 

 

I will take example with turing machin and OO programing, maybe it will be clearer what i'm talking about =) As the concept of entropy is quasi inexistant with turing machine, and like this we know we are not talking about something mystical

And i think it can interest also shelby because he is into this sort of problematics with language design lol

The problem is this conceptions from metaphysics to organize the world based on fundemental 'objects' with properties, and 'entelechy' , which is abtracted with the OO semantic of having class of objects with properties and 'entelechy' through the alteration of its state by its methods.

So far good, but then the probelm is when you want to program interaction between all the different type of object that can be present in the world, with OO programing generally it become quickly a design problem.

Either you will add a member function in all class to program the interaction with each other class with specialized functions for each type, but then it mean either you have make the interaction in double in each class, or then one class doesn't know or contain the interaction it can have with the other class, which is bogus from metaphysical stand point.

Either you do a visitor class for each pair of objects, and then each time you add a new type of object, you need to add visitor class for all the combination that the new object can interact with, but it's still bogus from metaphysical point of view because it mean the interaction between the object are not contained in the object themselves, but applied from the exterior through a visitor class that visit the two object in questions.

Even to program physic simulation like bullet physic it's not a small problem, when need to compute mutual gravity from two object for example, and that's only a simple case. And even if there is no entropy in turin machine, it's easy to see if you run complex real time physic simulation two time, you will never have the same result at the end, but it's not really entropy, neither really chaotic function , it's not fractal or strange attractor, only plain linear newtonian physics. I think the 3 body problem of gauss run into this problem.

This whole design of hard typed object make emergent property very hard to program and conceptualize.

Past days i've been digging more into haskel, already through reading the discussion of shelby on the git, i'm starting to get where they want to get at, because in fact in haskel there is this concept of monad, which are generic base object that can be used in generic code 'type classes' and can be specialized into pretty much anything, and the language allow to do meta programing very easily based on monad interactions, which allow to write generic code that can apply to any type, and i think i saw somewhere they are doing stuff to be able to handle emergent property kind of things based on type class like this.

http://www.haskellforall.com/2012/08/the-category-design-pattern.html

But it's the same principle i wanted to get at with my framework, to have generic place holder for holding reference to meta typed object with monomorphized access function in sort that the code is independant from the type of the data it manipulate as long as the node can convert it's data to the type required in the code.

It allow to manipulate collection of heterogenous objects and apply generic function to them without having to do specialized function that apply to each specific combination of class.

I'm not sure if typeclass are considered as being turing complete language before they are compiled/monomorphized to concrete type, but if they are turing complete already it mean it could allow to program things based on runtime dynamic data and emergent property between them, and it would probably lead with unpredictible result in the end, even if it's not chaotic functions or entropy, but in the realm between the turing undecidability and mutual interaction problem in physic.


http://number-none.com/product/Predicate%20Logic/index.html , http://number-none.com/product/My%20Friend,%20the%20Covariance%20Body/



https://books.google.fr/books?id=HBZADQAAQBAJ&pg=PA103&lpg=PA103&dq=haskell+monad+emergent+property&source=bl&ots=Rk20Gl6GEy&sig=iJyXEi8DbY4LuTgFjmFNgeOxpgo&hl=en&sa=X&ved=0ahUKEwi_r9iIusrUAhWMKMAKHbThAEcQ6AEIKjAF#v=onepage&q&f=false