Jump to content

Archived

This topic is now archived and is closed to further replies.

Christopher

Bitcoins - types of money, how might it develop

Recommended Posts

I just read up on Bitcoins. An interesting idea they have there. I think it might be worth discussing here:

It is an interesting try to get rid of fiat money.

It might be worth noting how it could fail down the road. If nothing else just to see if my predictions are right.

For hero it might in particular be used as a Supervillain plot in Superheroic.
And to model currency models in Future Scenarios (especially to check if something like Bitcoins will realy replace Fiat Money, or only become a new kind in the long run).

 

Question 1: Is it currency?

The debate is still on, but we can look at the 3 types of money humanity has developed over the centuries:

Commodiy Money, Representative Money, Fiat Money

 

The oldest type is "Commodity Money":

https://en.wikipedia.org/wiki/Commodity_money

The value of a gold (silver, other; but I will just be using gold here) coin is based on the gold in said coin.

In a way it can not really be forged or printed - you actually have to dig up/produce the commodity to make more.
"Forging" here is usually the process to create a coin with less parts of the comodity - like gold coated lead coins or a different alloy with less gold but same amount weight and volume.

It is somewhat save against inflation, because you can not print money without getting more gold out of the ground first. But it only scaled up so far as soceities and countries formed.

Example: Any of the old Coins. Any direct exchange of goods or services without money as intermediary. Latinum in Star Trek (it can not be replicated, only mined).

 

The 2nd type is representative money:
https://en.wikipedia.org/wiki/Representative_money

"We got x amount of gold stored at several places over the country. Every single Dollar can be exchanged for 1/[total amounts of dollars issued] part of that gold at any give time."

That makes it very similar to a stock share - each dollar is effectively a stock.

The advantage is that the money is much less subject to local influences. A dollar will have the same worth close to a goldmine as a thousand miles away, because only the state buying it makes it part of the calculation.
The bigger advantage however is that we can partition it how small we want/need, without having to melt down existing coins. If we got 10 million dollar and 30 bars of gold and then print 20 million dollar more, the value for every existing dollar would effectively go down to 1/3. Without having to touch any of those dollar bills/coins already around.

The value of a currency is based on the trust we have that it will continue to exist and how much/little is being printed.

Example: Every older Paper Money. The Dollar, when Fort Knoxx still matered. A Stocks book value. Most of the money kinds in Fallout (except Prewar Money)

 

The 3rd type is Fiat money:

https://en.wikipedia.org/wiki/Fiat_money

This is the kind we use now and oddly the wierdst type, because there is no value in the money pieces (a dollar note is as little worth as it was with Representative) but also there is no more need to store valuable material somewhere. We kinda just agree on the value.

Another way to look at it is that we own stock shares on the total worth of an economy. With the economy being defiend as "everyone that uses the currency and what they own".

Example: Just about any currency in use right now. A Stocks trade value. Fallout Prewar Money.

 

Bitcoins are odd compared to that, as it has elements of 1 and 3, while being neither:

It has the limited supply/addition per year of Commodity Money. The mathematic formula behind it will only generate about 21 million bitcoins total (reached by approxmiately 2040) at a rate of 25 every 10 minutes. It is a commodity money where we can actually predict the exact amount of mining taking place for the next 24 Years.

At the same time there is no physical value behind it. There is no gold/valuable good stored anywhere to back it. So it is closest comparable to Fiat money, with the worth being based on the trust we put in it.

 

Other interesting cases in Finction:

The currencies of the Fallout series.

Prewar money is clear Fiat money. But the other kinds tend to be Representative Money - backed by Gold, Water - or Fiat Money again - backed by the power of a local Government or the Power of the Brotherhood of steel.

 

Question 2a): Who keeps track of who owns what?

Answer: Literally everyone running the full bitcoin client.

Basically the ledger is stored on a P2P Network, with every full client holding a full copy of the entire Transaction log since inplementation (around 65 GiB by now), plus a set of transaction not yet part of the Blockchain.

 

Question 2b): And what are thier incentives for doing so?

The most recent transactions are not part of the Blockchain (yet).

Every 10 minutes (give or take) a new part (Block) of the Block chain is created from those unblocked Transactions.
It is a complicated mathematical process that requires picking the right random number and is intentionally padded/cut to take around 10 minutes. Whoever is the first to get said number get's 25 new coins plus a part of the transaction fees incurred over that timeframe.

Those +25 are the only way new bitcoins enter the ledger and is also part of the ledger. Once Bitcoins reach thier absolute cap only the transaction fees will still be gained.

 

Question 2c): How save is it agaisnt manipulations/faults?

One of the funny parts about peer-to-peer exchange of data is, that they suffer from extreme peer pressure.

When one Client thinks A is the answer but it is told from a majority B is the answer (and no or little that A is it) it will just asume it made a mistake and correct it's answer to B. While bad in a human social context, it works pretty well in a Computing context. It is how all Distributed systems guard against problems - just run the same operation 10 times in paralell, the result comming out most often is the right one (and the peer sproducing the faulty ones might need to be checked for hardware issues down the line).
That is how P2P networks keep data from degrading/being manipulated. Any attempt to alter the file/block chain on one client will simply be overriden when it get's the info it is wrong from other clients/his faulty data is not accepted onto the other peers.

One would have to alter the data on a distinct majority of the clients at the same time to get the other clients to accept your forged version as "right".

Bitcoin actually had two such cases:
A client exploit that allowed generation of unexpected Bitcoin amounts and a case where due to a bug two version of the ledger were existing at the same time (both accepted by half the network as "right").
The first case simply never made it past the P2P synchronisation. The exploit only affected one client (and no two clients in the exact same way), so the others just overruled it.

The second case was quickly realised. And solved itself when the mayority switched to a patched/previous version. With the clear mayority again arriving at the same conclusion and overriding all still buggy ones.

As long as a mayority agrees on one answer, localised mistakes can happen.

 

Point 3 Transaction Rules:

You can not transfer more money then you have avalible in your wallet. Any transaction that violates that rule will be denied at some point in the synching process and the real result will apear accordingly with next the ledger update. (Of course ideally your local client will always now it's balance)

You can send extra money - called the transaction fee - to give your transaction a higher priority for procssing. That transaction fee is then distributed among the mining people.

 

Question 4 How Save against Inflation:

In theory there is no individual or small group that can decide to just "increase the maximum of bitcoins", same way it works with local currencies and inflation. It would take a majority to agree, and any partial transition will potentially break the network for some time (as did the time of two ledgers because no clear majority found itself for one or the other).

That means the Bitcoin Generation rules can change at any given time, but will only stick when the mayority of clients accept that new Formula (and the resulting change in the ledger during each Block creation pass beyond the base rules).
That still sounds fine in theory when a lot of private people are holding full clients. But in practice the hosting of Bitcoin clients for the purpose of mining money is already being Industrialised.

Give it 10 years and I bet you the Firms holding the big Mining pools (and thus the bulk of the valid Ledger generation power) will decide amongst one another to just change the rules (shorter times between block generation, more fixed income per verfieid first generation of a block). At that point it pretty much becomes another plain fiat currency with a hardcoded mayority vote.

 

Question 5, Bitcoins and existing money:

Bitcoins are not accepted that widly.so far (around 100k shops, but a bunch of websites). However there are "Bitcoin exchanges", where local representative or fiat money can be exchanged for Bitcoins and vice versa. This is also the main way we got any exchange rates.

Share this post


Link to post
Share on other sites

Edit: Just found out about ASIC rigs for bitcoin mining. Last I looked ppl were using video cards. The ASIC rigs look cheap enough to play around with. Hmm...

Technically any GPU is a ASIC or a near-ASIC too. It is the perfect design for image processing (according to the current needs of Graphics engineers) wich makes it a good off-ASIC for Bitcoin Mining and similar tasks too.

They just went something better specialised for Bitcoin mining, rather then repurposing the next best common thing. An ASIC is basically just a step back to the old days

 

There are 3 levels of Paralellisation:

Inherently Serial - the Fibbonaccy sequence and any Encryption/hashing worth anything fall into this (for the later it is even a desing purpose)

Paralelliseable - you can Multithread it, but it requires work and will not scale up indefinitely. There is some bottleneck somewhere in the formula, where results have to be exchanged.

Emberrasingly parallel - You can just divide it up indefinitely, with next to no work and run as many in paralell as you have the hardware for

 

By it's nature Image production is emberrasingly parallel. Any pixel of the output image can be calculated independant of the other pixels in the output image.

GPU's are purpose built for embarassingly paralell problems. As such any embarrasingly paralell problem can just be run on a GPU.

 

The idea behind the bitcoin networks distribution was simple:

Every part of the Blockchains needs a nonce. With the [transactions in said block + nonce + previous block hash] when hashed having a specific result. The nonce would be generated randomly, so the one who guessed right first would be the one getting the bitcoins+transaction fees for that chain element.

 

What the designer failed to realise:

Any problem that revolves around finding a specific number in a big set can be paralellised. All you have to do is have one paralell task per possible nonce. And a modern GPU can run around 1366 x 768 (1.049.088) nonces in paralell.

That the network increases the difficulty target (and thus the number of possible nonces) only means more specialised hardware (GPU clusters, ASIC farms) are more likely to find the nonce, moving not as specialised hardware users out of the game of mining bitcoins.

 

Specialsation/Industrialsiation leads to a few people holding the bulk of the nodes on the mid to long run, likely corporations, governments or similar constructs. Whoever controls the bulk of nodes generating valid new Blockchains defines how many Bitcoins/Blockchain and where it cut's off (and thus the ability to print money) and how many transaction fees.

Having that power with a small group was exactly the whole damn thing the designer wanted to prevent when inventing Bitcoins.

 

As a old proverb says: "Humany never developed as system it could not figure out how to break."

China already banned Bitcoins, so they propably figure out that weakpoint.

I figured it out after reading the Wikipedia Article once and cross referencing it with my programming knowledge.

Share this post


Link to post
Share on other sites

I was mining one of the cryptocurrencies several years ago, dogecoin pretty sure, the machine I was mining with died, never reinstalled the client.

 

Somewhere I have an agent with a few hundred of the currency hanging out doing nothing...

 

The biggest weakness I actually see with cryptocurrency is I've never seen any plan in place to reclaim lost or abandoned currency; with a finite system that's a problem.

 

But the concept of cryptocurrencies is interesting and fascinating. Putting the net value into the hands of all carriers subverts things like the national control, interest rates are no longer centralized and regulated... not sure how that'd play out in a large scale environment.

Share this post


Link to post
Share on other sites
The biggest weakness I actually see with cryptocurrency is I've never seen any plan in place to reclaim lost or abandoned currency; with a finite system that's a problem.

 

But the concept of cryptocurrencies is interesting and fascinating. Putting the net value into the hands of all carriers subverts things like the national control, interest rates are no longer centralized and regulated... not sure how that'd play out in a large scale environment.

There can be no reclaiming of credentials without a central, encrypted storage. Everythng in the ledger is public, so they can not just store your security question.

For Bitcoin they choose the Private/Public key pair. Mostly because you can publish your public key without endangering the more important private key you use for encryption.

 

Declaring the currency abandoned:

Again this can not happen without a central oversight agency. Even if you do that, who are those coins given too?

There is no central bank that can just take them and give them out as mining reward later.

Since they are stored in a digital distributed system with full logs going back to implementation of it, there is also no wear&tear like on physical money.

 

Lack of central oversight:

I am very uncertain about that actually holding true all the way to 2040.

As I said: Mining is being industrialised. From a CPU, to the GPU to ASICs is a lot differnce in specialsiation. And as the mining reward goes down, this will only accelerate (after all mining costs power/up front investment).

The moment one coorporation (or a conglomerate of coorporations) holds enough nodes to decide how the next block should look like with some certainty, they effectively own Bitcoins.

Share this post


Link to post
Share on other sites

Technically any GPU is a ASIC or a near-ASIC too.

 

I was mainly thinking of the horrendous price and energy inefficiency of the old GPU-based systems.

 

After researching it a bit, it seems like it's not really feasible on a small scale, other than just doing it for fun.

Share this post


Link to post
Share on other sites

Mining has been industrialized since the start. RaspberryPi units are extremely efficient and have a much higher ROI than tasking even a game console to mining. A lot of groups have attempted to capitalize on owning a whole block on their own, the very nature of the cryptocurrency mining goes a long way to preventing this. Specialized CPUs have been created and built specifically to reduce power consumption (so the cost of running a unit is less than the number of currency it helps generate).

 

I never got enough into cryptocurrencies to really delve into the economics of it (actually, I got bored with the idea and stopped reading...); but you seem really late to the game with questions answered already. I'd recommend finding some forums dedicated to these things and inquiring there to further your thoughts, they aren't bad, I just think they're already answered. Bitcoin is not new. Especially not in internet time. The whole point of cryptocurrenices is decentrialized currency and most (if not all) have measures in place to prevent centralization.

 

Even if someone knows what the next block looks like (and that's nearly impossible, given the nature of the block mining) starting in a few months the next block is only worth 12.5 coins. And even if you have one block, the next one after that is going to be finished by a different node, that's the nature of the mining process (the next block relies on the previous block, once you have one, you can start down the child-chain); The nature of the system is almost 100% self correcting, the faster you mine the harder it is to mine the next block. For bitcoin that's retained an average of 10 minutes per block since inception (7 years now).

Share this post


Link to post
Share on other sites

Even if someone knows what the next block looks like (and that's nearly impossible, given the nature of the block mining) starting in a few months the next block is only worth 12.5 coins. And even if you have one block, the next one after that is going to be finished by a different node, that's the nature of the mining process (the next block relies on the previous block, once you have one, you can start down the child-chain); The nature of the system is almost 100% self correcting, the faster you mine the harder it is to mine the next block. For bitcoin that's retained an average of 10 minutes per block since inception (7 years now).

You don't have to fake the next block, if by pure chance you happen to get the next block. And if you happen to own 75%* of all Block generating computation power, that chance is pretty darn good.

*Or whatever percentage you need to convince all other nodes they are wrong.

 

The economics of reduced income for mining cuts both ways - it hit's the small miners as well as the big ones. Except the big ones might have the ability to cope with it, coordinate with other big ones*, or just invest the money for the longterm gain of owning Bitcoins.

By all rights slowdown of supply should worsen the industrialstion.

 

*In this phase thier biggest danger is to try to "out calculation power" one another. Ending up with the equivalent of a ASIC armsrace, wich drives everyone out of business and might even negatively impact the value of the Currency.

Cryptocurrencies basically go back to the times of the "gold rushes", where a lot of smalltime surface miners could make a living of it. But in the end that idea died off, replaced by dedicated Mining Coorporations

 

You are right that I propably should ask that on a dedicated Forum. It seems almost to easy to subvert. Yet at the same time it seems mathematically impossible to not have it end up with a few big miners rather then lots of small ones.

Share this post


Link to post
Share on other sites

You don't have to fake the next block, if by pure chance you happen to get the next block. And if you happen to own 75%* of all Block generating computation power, that chance is pretty darn good.

*Or whatever percentage you need to convince all other nodes they are wrong.

 

There's your problem - you would have to own 100% of all nodes in existence to spoof even one block.

 

That's the point of decentralized encrypted currency. You can't spoof it. Or you can't spoof it for very long.

 

Faking a block is nearly impossible to do (I would say impossible, but nothing is); and even then faking it wouldn't be a matter of mining capacity, it would be a factor of altering the data in the block itself, and every block it's ever interacted with, and every part of the chain that contains information on the blocks existence. They aren't stand alone, their part of a decentralized chain.

Share this post


Link to post
Share on other sites

There's your problem - you would have to own 100% of all nodes in existence to spoof even one block.

 

That's the point of decentralized encrypted currency. You can't spoof it. Or you can't spoof it for very long.

 

Faking a block is nearly impossible to do (I would say impossible, but nothing is); and even then faking it wouldn't be a matter of mining capacity, it would be a factor of altering the data in the block itself, and every block it's ever interacted with, and every part of the chain that contains information on the blocks existence. They aren't stand alone, their part of a decentralized chain.

Except any Distributed storage node must be able to accept that "apparently it made a mistake and that data from all the other nodes is right".

 

Let's asume you own the Majority of nodes, whereever that figure is.

Let's asume that the first time you deviate from the original rules is Block A. All your nodes come to the same conclusion: Node Y generated that Block first, and Node Y get's 100 Bitcoins for it.

However the rest of the nodes have a different result. They get Block A2, where Node X generated the Blockchain part first. And get's the predetermiend amount of Bitcoins for it.

Except every node that Generated Block A2 will be confronted by a mayority with Block A and decide "if you all think A is the right one, I guess I was wrong with A2". And take Block A as part of the ledger.

 

Now we get to Block B. This one will take the hash of Block A as part of it's definition. So we end up with these version of Block B

A+Y/100 = B

A2+X/predetermined = B2.

A+X/predetermined = B3.

And since it is your nodes again creating Block B it has again the majority and again everyone will have to accept "Oh, I guess I was wrong".

 

You do not need to control all nodes. Just enough nodes that the others will cave in under peer pressure and just asume they themself are wrong.

Share this post


Link to post
Share on other sites

There is no "cave in under pressure". Modifying a block requires you to modify both all it's parents and all its children; the error checking just goes back up the chain and determines the error. You literally can't get more bitcoins from a block than is set to be distributed. Children blocks are dependent on parent blocks, the nonce required to even mine them is determined by what came before, you can't fake it, or you can't fake it for long. The whole chain contains the information to verify where a bitcoin came from, what transactions it's involved in, and who has it now. A section of chain will never assume its wrong, it has it's entire history distributed along with it.

 

This is also why mining gets you very few bitcoins per hour, you will never mine a block on your own. The next block is mined by the entire collective as a whole. That's why the iterations needed to determine the next nonce is in the hundreds of trillions at the moment.

Share this post


Link to post
Share on other sites

There is no "cave in under pressure". Modifying a block requires you to modify both all it's parents and all its children; the error checking just goes back up the chain and determines the error. You literally can't get more bitcoins from a block than is set to be distributed. Children blocks are dependent on parent blocks, the nonce required to even mine them is determined by what came before, you can't fake it, or you can't fake it for long. The whole chain contains the information to verify where a bitcoin came from, what transactions it's involved in, and who has it now. A section of chain will never assume its wrong, it has it's entire history distributed along with it.

 

This is also why mining gets you very few bitcoins per hour, you will never mine a block on your own. The next block is mined by the entire collective as a whole. That's why the iterations needed to determine the next nonce is in the hundreds of trillions at the moment.

P2P Networks are inherently vulnerable to data corruption, intentional or accidental. Any one of the Nodes might distribute it's corrupted data to any other node. That fact is as unavoidable as UDP not having recieve notifications/reliable transmissions. Like UDP however, P2P applications programmers learned to live with it by adding proper checks around it.

Faking a part of the Block Chain is no different from changing a chunk on any other P2P network. And the usual way to solve that is to ask other nodes for thier opinion.

 

Actually my appraoch is known and was at least feasible at one point. It is called the "51% attack":

http://blogs.wsj.com/moneybeat/2014/06/16/bitbeat-a-51-attack-what-is-it-and-could-it-happen/

It kinda resolved itself because a lot of Mining pools jumped ship, pushing that groups hashing power all the way down to 27%.

Short term solutions were discussed, but not implemented because they would only have delayed/moved the problem down the road.

 

A 51% attack was always possible for P2P, but never a realistic outcome. Because never before was there a financial incentive to own 51% of the Network. With Cryptocurrencies there is.

 

https://en.wikipedia.org/wiki/Bitcoin#History_modification

People fear those attacks will be used to crash the network/currency. That of course would be economical bullshit. You would have to get and hold 51% of the Nodes for a long period of time to do that. And it would be damn obvious who you are. And you would make no mone from this.

 

However what would the short term effect be if they "just" increased the Bitcoins gotten from mining beyond the original formula with thier 51%-60% mayority? I guess it would be the same as any other Centralised Inflation.

 

What if thier 51% holds and they keep that up for 10-20 years? It is not a severe change. In 10-20 years there might not even be a outcry against it. Or a few big corporations hold the mining pool majority anyway, so it would be in thier best interest (short to midterm) to just agree on that change.

Private Miners that are still using the Original Formula nodes would be incentivised to either adop the alternate Formula (so they could mine again). Or just cut thier losses (decreasing the cost to hold the 51%).

You can increase the percentage needed to say 75%. But that would still not prevent it, only put it further down the road.

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.

×