I’d love some feedback on the latest update on my kickstarter. Would anyone be willing to make sure I haven’t misstated the math and/or concepts?
Hypercapitalism: Information Theory and Bayes
skilesare
Karma: 136
Yes...the kickstarter portion is a bit of a mess. I threw it together at the last minute when I was invited to pitch my ideas. I’m more to net rested in feed back on the math and concepts behind it. I think I understand bayes and am trying to apply it. This seemed like the right place to seek guidence.
This was a really helpful and encouraging read. Thanks for taking the time to post it. I’m doing some research in the are of economics and have had trouble finding people willing to discuss my ideas. I realize I’m going to have to get much more familiar with the math. This makes it seem possible.
My wife tells me there are four love languages and I’m not very good at speaking her’s.
This so needs to be quad lingual.
I’d love to talk more with you about Bitcoin and overcoming the negatives. I recently jumped in head first when I discobeed that btc had some established tech that I could use for hypercapital.info. Send me a message if you’d like to talk more.
Yes , this is the issue. I’m deficienct in the math area. I’m working on an out line to lay out the math, but it has been a long time since I took diffeq and it is coming back slowly.
Thanks for the feed back.
We have an economic system with N actors. Each actor has its own utility function that it uses to attempt to spend/invest money in areas that will grow. The system as a whole doesn’t know these functions and the nodes can’t see them internally. They just make a judgment and spend/invest. If they spend in an area that grows, more money comes to them via an agent in the system that redistributes cash as it flows to the originators of cash in a node.
For example, If N1 pays a dollar to N2 for a bottle of wine, N1 gets a share in N2. As cash flows through N2, little bits get funneled back to N1. So if N2 becomes the next big wine maker, many bits will flow to N1 and it will be rewarded for sending money to N1 early in time.
Does it follow from bayes theorem that if I keep passing cash through this system, that over time, the success rate will ocellate around the actual success rate of each nodes utility function? In this scenario, if you fail you get your cash back slowly over time,if you succeed you get it back more quickly.
I’m anticipating that a set of actors in this situation would end up in an economy where the level of wealth for each node converges on their true ability to create value.
If I’m totally misinterpreting, I’d love some pointers to good info to read.
That is interesting....what do you mean ‘on its own’. Are there some other things that affect the application of Bayes to a system?
Let me think about reforming the question now that I’m not on an iphone.
Lets think about it as information theory. If there are 5 different types of ‘information’ that different people are listening for, then we need to reduce the entropy in 5 different channels to get the right information signals. This is much more complicated than trying to just reduce the entropy around 1 thing.
Maybe language isn’t the best term, but all 5 of the information sets are important in different amounts to different human beings. So your utility function is going to get messy.
Question updated. Any more clear?
That person is me. Check out hypercapital.info. I’d like to think it is more than just randomness though...in this system you knowledge and group wisdom leads to who wins the lotteries.
Have you considered that one of your base assumptions that you can ‘store value’ is false?
All value us future value. You can’t store it. You can make a bet on a piece of capital. Gold has been a decent bet, but far from stable. Bitcoin is just another bet.
Also, btc has a significant long term risk as well. The system is terrible for an interstellar economy. You can’t have a blockchain when you have to wait light years for payment confirmations. Maybe this isn’t a big deal in the short run, but if you’re looking really long term, it is an issue.
A couple of other things to consider in your favor, bitcoin is going to have significant uses in the short term that don’t involve holding it. It is a transfer medium and a protocol. Having a stake in the protocol will be very valuable.
I guess the stupid question is does it follow from Bayes that if you keep measuring the same probability over and over that you will converge on the ‘actual’ probability.
This was really helpful and gives me some great stuff to look at.
Thank you.
My theory is that actors in an economy spend cash on things and some of those things produce lasting value in the economy and some don’t. Each actors probability of making a valuable choice that leads to overall growth is unknown. If we reward those that make a valuable voice with fresh cash, they then have the opportunity to succeed or fail again. If we do this over and over the ‘right’ probabilities will emerge and we will see who the ‘best spenders’ are by who has the biggest rewards flowing back.
We optimize for value creation and in the long run have a system with better and better information.
Numenta’s stuff made a lot of sense. They kept things simple by removing the recursion of HTMs...and I think that is probably the key to the whole thing working.
All that being said, their latest product Grok seems to have some success in the network monitoring space. http://numenta.com/grok/
On Intelligence was my first intro to the idea of Bayesian thinking.
Great Question. The ‘bits’ in the system I’m proposing are based on a system wide demurrage or ‘decay rate’ of currency. Simply switching to a different node doesn’t change the decay on cash you hold. There isn’t an incentive to create a new node. On the positive side, existing customers have a loyalty factor. N1 will be more likely to buy the same commodity from N2 than from a random Nx. This behavior has a limited life though because diminishing returns eventually catch up and suddenly the benefit from being one of the first contributor to Nx is greater than the loyalty to N2.
This gives a lifespan to legal entities and increases the turnover thus increasing the likelihood of more fit entities emerging(if you assume that entities can cross generationally share information).
You basically get the the attractiveness of youth, the steadiness of adulthood, and the slow decline to oblivion. (and with this an increased incentive to figure out immortality by creating enough value to outrun the diminishing returns)
I did ask it in the stupid questions thread. :)
I think that both can be true and yet still have real results. Take humans, reproduction, and marriage. Typically a man is fertile for more years than a woman. We see in marriage a tension between men staying loyal to the wife of their youth and moving on to a more fertile partner. I don’t have statistics in front of me, but over history the tendency is to stay loyal. Patrimonialism has a profound evolutionary basis and my theory is that you can use that built in bias to form a sustainable system where legal entities have life spans instead of immortality. If the life span is too short, than it is useless.
As far as the fees go, Bitcoin’s fees are non-zero but very close to zero and many alternate payment schemes can be constructed. Typical CC transactions are 3%...much higher than the about .05 needed for a BTC transaction. There are also ways to convince miners to mine your transactions even though no BTC Fees are provided.
I’ve tried to set up a system where tax avoidance is reduced or eliminated. Because the transaction system will reject transactions that don’t pay the fee when they use their cash, they are stuck with the decision to participate in the system or not. Once the cash is in the system, they must pay the tax or the tax will be taken from them(using btc multi-sig where the decay charging authority is held accountable to only charge the fee on delinquent accounts.
N2 can certainly set up Nx and move all cash over there. Lets use a real example.
N1 spends $100 with N2. N2 wants to avoid the decay(but the system always charges at least one day of decay during a transaction) So they move the cash to Nx. The transaction occurs and $0.003 cents goes back to N1. Now the cash is in Nx. What are they going to do with it here. If they let it sit for 30 days they will be auto charged a decay fee of about $0.10. This flows to N2.
Even if N2 is proactive and sends it back to N3 immediately, $0.0000032 will flow back to N1. A small amount to be sure, but overtime these small amounts add up.
And if Nx uses the cash to develop something that brings in far more cash than went in, the amounts get much bigger.
That is besides the point because we want to avoid the situation entirely where N2 tries to devalue N1s benefits by passing to a shell corporation Nx
Nothing can keep someone from just passing cash and on and on and on to cash it owns except rule of law and accountability. Accountability can be observed in the blockchain and bad actors identified. Rule of law comes later. (I try to cover this in STH. Statutory Theft—https://github.com/skilesare/art_and_democratic_hypercapitalism/blob/master/the_pattern_language/sth_statutory_theft.md )
Re: Fees—I don’t have a great solution to this other than offering miners a share of future pref payments for any mined items that they charge no fee for. This involves them taking risk, but also provides substantial long term rewards.
All of this goes much deeper than the original question which I think now is best framed as ‘does having a backflow of cash based on amount spent enhance the information we can get out of an economic system over the standard capitalistic model of today.’ If we add too many things in we end up in a conjecture bias situation.
Once I answer the first question in the affirmative, then I can move on to whether the implementations of the system are rational or not. If achieving the prior is a priority, there likely exists an implementation that can achieve it. At least I think.
I’ve been working through the sequences and have been trying to apply things to my current project that has to do with remaking money. I’d appreciate any feedback. Did I miss anything major? Am I on the right track to fix my thinking?
Maybe check out Christopher Alexandrt’s Nature of Order.
It sounds like his wholeness function is what you are poking around.