Post
Topic
Board Development & Technical Discussion
Re: Thoughts on M of N systems (competing and heirarchically nested)
by
godislove
on 21/06/2014, 11:50:56 UTC
In summary, I have felt some regret that M of N methods may be the only way to tie the syntax of Ethereum, etc to the semantics of the outside world, and in particular to our main interest, human society.  However, bitcoin is already based on a competitive M of N method. M of N forms a logic gate that is Turing complete when M of N "contracts" are "wired" together in a minimally sufficient way, which at least requires feedback and memory states before being provably capable of being a complete "computer".  But by not being as simple as a 2-input universal NAND gate, having more inputs, it is also fault tolerant and used extensively in recognizing patterns as in the brain, economics, evolution, and artificial intelligence.  (note for A.I. readers: Bayesian networks, Numenta's CLA, and Neural Networks are logically reducible to M of N and can be made Turing complete when feedback wiring is included. )  When tied to a currency that represents the total available resources falling under the jurisdiction of the legal system that the system of M of N contracts represent, the network of M of N contracts can be designed (via politics affecting the legal system) to evolve on its own to make more efficient use of the system's resources to make more agents behind the M of N contracts .  At the risk of off-putting the reader with an overly-grand suggestion, this might be the "meaning of life" or the "definition of intelligence".  The M of N contracts need to be enabled to competitively evolve, fed by the legal power over the real resources (represented by the currency) in order to make imperfect copies of themselves. I'll say more on this negative-entropy generation idea later.

In at least some of these intelligent networks of M of N of operations, the currency is not the only distribution of power.  There are also usually mechanisms to prevent the currency from being concentrated into the hands of a few M of N agents, which could act like viruses or cancers.  In society we use functioning democracies (arguably this does not include the U.S.) or advanced vote-like communism (arguably like China, whose government is economically empowering the people out of a fear of a USSR-like breakdown and real fear of the people themselves) to prevent kings and oligarchies. The purpose of 1 vote per person is to prevent legal power from being accumulated in the hands of a few white Christian wealthy men, as occurred in the past. By giving more power to women, blacks, non-Christians, and the middle-class (via progressive tax of the 1950's), democracy is able grow faster and overthrow other systems of governing via economic power.  In evolving artificial intelligence methods this is seen as a redistribution of computing resources if a few agents are acquiring too much computer power and thereby settling on a specific solution that may not be the optimum solution. In biological evolution species-diversification is regarded as healthy in finding future solutions to the survival puzzle.  The puzzle is really about acquiring as much energy as possible to make as many DNA-base pairs as possible.

Getting off the track of a summary in order to address something I've left out....

In order to tie the votes and thereby economic health to people in these M of N contracts, the individual agents behind the contracts (people) to which I assume all readers of this post ultimate care or should care about, it is necessary to prove the existence of specific people who are behind the contracts.  This is not only needed for voting, but for enforcement of the contracts.  This does not mean identities should be public, but only that the contract system itself should be aware of individual "IP addresses", unless a physical force needs to be applied (incarceration).  Some will want at least a portion of their cyber-identity (not necessarily their complete cyber or any of their physical identity) to be public in order to build trust as in name brands, selling stuff on ebay, getting a job, and dating.  In order to do all this, a physical device(s) is needed. It should be capable identifying a physical individual based on multiple bio-signatures before they can interact with the system of M of N contracts, including the currency which is in a sense a top-level M of N contract that runs parallel with intelligent and hierarchical voting M of N contract (but I do not advocate democracy as ideal) as a check and balance if not optimization process on the distribution of the currency, to not only see if the system is achieving its goal of maximum happiness per median person, but to optimize the process.  Competing, non-compatible systems of M of N contracts with different currencies would struggle for more citizens to leave other systems to join their own. Instead of fighting to keep citizens out or locked in, these "socio-economic-politico" governments would fight to provide the most benefits to the best and brightest.  There would be local competing governments like this that still fall under a more regional government, leading up to world government.  Without world government limiting the competition, we should be prepared for real war and more brutal economic competition for happiness and food. The world government would not let us be too lazy or local "viruses" and "cancer" will develop. Utopia is now technologically possible with world-wide communication to prevent war and reach agreement, even optimum agreement for all.

Getting back to a summary...

The ultimate goal of life is to extract energy to make copies. Although this increases entropy system-wide, the copies are a local increase in real physical and informational order (negative entropy) that is the foundation of what we perceive as "good" about life (a la Schrodinger's paper "What is life?).  Energy is converted to copies instead of being locked up in potentials or converted to waste heat. This is a rapidly growing re-interpretation of evolution.

The system needs an agreed-upon goal that the core programmers need to be thinking about if they want to avoid enabling just as many bombs as benefits.  This is arguable since nuclear weapons did not destroy us and the blindness of the internet protocol has served us well. But by being intrinsically capable of bombs much more than the internet, the acceptance and legitimate use of Ethereum or Zerocash -like methods may get backlash.

These are the highest-level ideas concerning where this technology is going, or should go, that I do not see written elsewhere, including Nick Sabo and Wei Dai.