Bitcoin & Tragedy of the Commons

[cunicula]

Let me rephrase that.  I would like to see a scheme that regulates a 51% attacker to insure that it does not double-spend.

It seems extremely improbable to me that the monopolist could profit from double-spending. That is regulation enough in my view.

Somehow, I almost think you are serious. Nah.

I am serious. Completely.

https://en.bitcoin.it/w/index.php?title=Proof_of_Stake&action=view#The_Monopoly_Problem

Potentially, the monopolist could choose to do this in malicious ways, such as double spending or denying service.

Cbeast, I'm sorry but you are going back on my ignore list.

[cbeast]

Let me rephrase that.  I would like to see a scheme that regulates a 51% attacker to insure that it does not double-spend.

It seems extremely improbable to me that the monopolist could profit from double-spending. That is regulation enough in my view.

Somehow, I almost think you are serious. Nah.

I am serious. Completely.

https://en.bitcoin.it/w/index.php?title=Proof_of_Stake&action=view#The_Monopoly_Problem

Potentially, the monopolist could choose to do this in malicious ways, such as double spending or denying service.

Cbeast, I'm sorry but you are going back on my ignore list.

I'm just pointing out a minor contradiction in your logic. Denying a problem won't make it go away. Besides, I only took you off my ignore list because you seemed to have conviction regarding proof-of-stake. I will await a fully fleshed out hypothesis without contradictory statements. I'll keep an eye on the wiki. It will be especially interesting to see how you refute the arguments of Nobel winning economists based on the premise of Tragedy of the Commons.

[Meni Rosenfeld]

You're overlooking what "tragedy of the commons" means. As you said yourself, the payment processor will make 325 € per block whether he mines or not, so he has no incentive to mine himself - he would much rather have others mine instead.

Besides, if payment processors and banks are the only ones doing mining, it means Bitcoin has become fairly centralized.

You are missing the point: the payment porcessor will make 325€ ONLY if the network is functionnal and secure. So the value add fees are his/her incentive to mine (not the bitcoin mining fees).

I'm sorry, but you are. The contribution of the payment processor's own mining to network functionality is negligible, since he's only a small part of it. It's in everyone's best interest that everyone will mine/pay for mining. But then it's also in everyone's interest to personally cease mining, since the network will be just fine with everyone else mining. The contribution of a miner / mining sponsor is shared by everyone, but the cost is all his own. I can't explain this any clearer, I can only direct you to Wikipedia.

Miners are all different sizes, this gives some of them pricing power in terms of selling "get my txs into a block soon". If all miners are very small and all use the "accept all with fee rule" many may indeed drop out lowering overall power and difficulty. This will make it easier and more likely for someone to gain a pricing power ability. The existence of someone with pricing power ability raises the profitability of all miners causing more small players to enter the mining biz and reducing the possibility/ease with which one party can get >50% (which may not be disastrous anyway, but still makes me uncomfortable).

For example, suppose DeepBit changed policy to only include tx that had fee >= .01BTC. Some people will not care because they can with high likelihood get in a block within 10 blocks anyway. But the few people who need near certainty of getting in within an hour now attach a .01 fee. All miners profit from this in proportion to their hashing power except DeepBit. DeepBit profits (or maybe not if they pick the wrong price) by less depending on how many of the .0005 fees they no longer get.

This makes sense, but I think that quantitatively it will only be enough if power is concentrated with a few (where "few" could be 100) strong players.

[cunicula]

This makes sense, but I think that quantitatively it will only be enough if power is concentrated with a few (where "few" could be 100) strong players.

Empirical work indicates that market power for undifferentiated goods is extremely limited unless you have very few market participants (approximately less than 5).

[phelix]

[...]
For example, suppose DeepBit changed policy to only include tx that had fee >= .01BTC. Some people will not care because they can with high likelihood get in a block within 10 blocks anyway. But the few people who need near certainty of getting in within an hour now attach a .01 fee. All miners profit from this in proportion to their hashing power except DeepBit. DeepBit profits (or maybe not if they pick the wrong price) by less depending on how many of the .0005 fees they no longer get.
[...]

I thought this was the whole point of the tragedy of the commons: that each miner for himself has no benefit from dropping tx with very low fees. So all tx fees would go towards one satoshi.

There sure are solutions to this problem.

Like a network enforced minimum fee determined by a somehow balanced vote.

[cunicula]

[...]
For example, suppose DeepBit changed policy to only include tx that had fee >= .01BTC. Some people will not care because they can with high likelihood get in a block within 10 blocks anyway. But the few people who need near certainty of getting in within an hour now attach a .01 fee. All miners profit from this in proportion to their hashing power except DeepBit. DeepBit profits (or maybe not if they pick the wrong price) by less depending on how many of the .0005 fees they no longer get.
[...]

I thought this was the whole point of the tragedy of the commons: that each miner for himself has no benefit from dropping tx with very low fees. So all tx fees would go towards one satoshi.

There sure are solutions to this problem.

Like a network enforced minimum fee determined by a somehow balanced vote.

Increasing fees is not a good solution. Proof-of-stake makes large fees unnecessary.

[cbeast]

[...]
For example, suppose DeepBit changed policy to only include tx that had fee >= .01BTC. Some people will not care because they can with high likelihood get in a block within 10 blocks anyway. But the few people who need near certainty of getting in within an hour now attach a .01 fee. All miners profit from this in proportion to their hashing power except DeepBit. DeepBit profits (or maybe not if they pick the wrong price) by less depending on how many of the .0005 fees they no longer get.
[...]

I thought this was the whole point of the tragedy of the commons: that each miner for himself has no benefit from dropping tx with very low fees. So all tx fees would go towards one satoshi.

There sure are solutions to this problem.

Like a network enforced minimum fee determined by a somehow balanced vote.

Of course there are manageable solutions to the problem. Tragedy of the Commons implies that there will not be outside factors influencing the commons and they will be exploited to the point of failure. American manufacturing is an example. Exploitation of health insurance has caused insurance premiums to skyrocket. The high cost of health insurance has in turn caused goods to become overpriced because the labor pool demands higher compensation. When labor compensation is reduced, manufacturing slows due to a lower labor pool. Manufacturers in countries that provide well managed NHC like Japan, Korea, and now even China are able to produce goods with lower labor costs and thus be more competitive. (This is not an argument for or against NHC, just an example of Tragedy of the Commons and a solution.)

If we further this analogy and apply PoS theory to health care and manufacturing, then the solution would be to allow manufacturers with the large bank accounts to dictate healthcare policy for workers. This is where the US is currently. See how well it works?

[Meni Rosenfeld]

[...]
For example, suppose DeepBit changed policy to only include tx that had fee >= .01BTC. Some people will not care because they can with high likelihood get in a block within 10 blocks anyway. But the few people who need near certainty of getting in within an hour now attach a .01 fee. All miners profit from this in proportion to their hashing power except DeepBit. DeepBit profits (or maybe not if they pick the wrong price) by less depending on how many of the .0005 fees they no longer get.
[...]

I thought this was the whole point of the tragedy of the commons: that each miner for himself has no benefit from dropping tx with very low fees. So all tx fees would go towards one satoshi.

Yes, but FreeMoney's point is that if the miner is big enough, his own impact on the equilibrium by playing hard-ball is enough to justify his costs.

There sure are solutions to this problem.

Like a network enforced minimum fee determined by a somehow balanced vote.

That's certainly a possible solution, but deciding on the minimum fee is not trivial. Alternatively you could limit the block size, but again there's no clear way to decide on the limit.

[cunicula]

That's certainly a possible solution, but deciding on the minimum fee is not trivial. Alternatively you could limit the block size, but again there's no clear way to decide on the limit.

Enforcing a large, unnecessary fee is a solution,  but it is grossly inferior one. Why add significant fees in when they are not needed?

[phelix]

That's certainly a possible solution, but deciding on the minimum fee is not trivial. Alternatively you could limit the block size, but again there's no clear way to decide on the limit.

Enforcing a large, unnecessary fee is a solution,  but it is grossly inferior one. Why add significant fees in when they are not needed?

it may be much easier to implement and not require a fork. and fees would still be low.

[cunicula]

That's certainly a possible solution, but deciding on the minimum fee is not trivial. Alternatively you could limit the block size, but again there's no clear way to decide on the limit.

Enforcing a large, unnecessary fee is a solution,  but it is grossly inferior one. Why add significant fees in when they are not needed?

it may be much easier to implement and not require a fork. and fees would still be low.

How much would they be? I believe that the fees would need to be similar to or greater than those for credit cards or paypal in order for them to credibly discourage a monopolist. A monopolist might lower fees somewhat because he wouldn't need to use much electricity, but how much could we trust a proof-of-work monopolist?

Right now, the current difficulty level is supported by an implicit inflation tax on all bitcoin holdings of around 30% per annum. The amount sent per annum is not known, but let's say that (largely voluntary) donations currently occur at a rate of 0.01% per send. I believe I am already low-balling here, but even if you believe that 0.001% is the true donation rate, we still need significant fees. Gross Txn fee revenue is currently around 0.03 BTC per block. To match block generation revenue, the fee submission rate needs to increase by 50/0.03 = 1666-fold. That is we need to move fees on sends from 0.01% (my guess of donation rates) to 16%. Volume of sends will decline in response, so even 16% won't actually generate this much revenue. In fact, I expect it is impossible to support current block rewards using txn fees.

What is all this fear of a fork? Better to fork sooner rather than wait for the walls to crumble around you, no? I thought we were revolutionizing currency. Obvious flaws need to be fixed by bitcoin or they will get fixed by some other coin. A tradeoff between no security to speak of and onerous fees is not a quibble. It is a core issue.

[Rassah]

I've read a lot about monopoly here, as if they are guaranteed, but no real mention of why. Monopolies generally happen only because of economies of scale (http://en.wikipedia.org/wiki/Economies_of_scale, http://en.wikipedia.org/wiki/Natural_monopoly) In short, where the average cost of an extra unit of production falls as you add more units. I'm not convinced that applies to Bitcoin.

At the risk of possibly making this sound stupid, I'm going to keep this simple:
Bitcoin mining is direct Electricity Cost -> BTC profit. You can't make your electricity cost per miner go down by adding more miners.
Sunk costs (infrastructure) play a part, but I don't think it's a large part.
Barrier to entry is low. Buy a GPU or wherever new tech comes out, and you'll get a fraction of overall profits. Just join a pool or P2Pool to reduce variance.
So, I'm not convinced that monopoly is inevitable, and I think the most likely outcome will resemble that of agriculture, where anyone can grow crops, and buy from large corporate farms or locally grown, except even less monopolistic than agriculture due to lack of government incentives (subsidies, import tariffs). In Bitcoin terms, that will mean we will have a combination of large pools and individual miners, just as we have now, but will only rely on increased value of bitcoins in transaction fees for security.
I also wouldn't be surprised if those with large personal Bitcoin holdings will buy their own mining hardware as well. It's a switch from paying $200 a month to your bankers to keep your money/gold/whatever safe, to paying $200 a month in electricity to run your personal rig to help keep yourself safe.

[cunicula]

I've read a lot about monopoly here, as if they are guaranteed, but no real mention of why. Monopolies generally happen only because of economies of scale (http://en.wikipedia.org/wiki/Economies_of_scale, http://en.wikipedia.org/wiki/Natural_monopoly) In short, where the average cost of an extra unit of production falls as you add more units. I'm not convinced that applies to Bitcoin.

At the risk of possibly making this sound stupid, I'm going to keep this simple:
Bitcoin mining is direct Electricity Cost -> BTC profit. You can't make your electricity cost per miner go down by adding more miners.
Sunk costs (infrastructure) play a part, but I don't think it's a large part.
Barrier to entry is low. Buy a GPU or wherever new tech comes out, and you'll get a fraction of overall profits. Just join a pool or P2Pool to reduce variance.
So, I'm not convinced that monopoly is inevitable, and I think the most likely outcome will resemble that of agriculture, where anyone can grow crops, and buy from large corporate farms or locally grown, except even less monopolistic than agriculture due to lack of government incentives (subsidies, import tariffs). In Bitcoin terms, that will mean we will have a combination of large pools and individual miners, just as we have now, but will only rely on increased value of bitcoins in transaction fees for security.
I also wouldn't be surprised if those with large personal Bitcoin holdings will buy their own mining hardware as well. It's a switch from paying $200 a month to your bankers to keep your money/gold/whatever safe, to paying $200 a month in electricity to run your personal rig to help keep yourself safe.

Added to ignore list.

[Meni Rosenfeld]

I've read a lot about monopoly here, as if they are guaranteed, but no real mention of why. Monopolies generally happen only because of economies of scale (http://en.wikipedia.org/wiki/Economies_of_scale, http://en.wikipedia.org/wiki/Natural_monopoly) In short, where the average cost of an extra unit of production falls as you add more units. I'm not convinced that applies to Bitcoin.

Monopolies form because when you're the only one selling a product, you can demand whatever price you want. In Bitcoin mining, if you're a monopoly you can require high transaction fees, and you can keep difficulty artificially low thus lowering your cost.

[Rassah]

Added to ignore list.

Thank you! Saves one of us the hassle.

I've read a lot about monopoly here, as if they are guaranteed, but no real mention of why. Monopolies generally happen only because of economies of scale (http://en.wikipedia.org/wiki/Economies_of_scale, http://en.wikipedia.org/wiki/Natural_monopoly) In short, where the average cost of an extra unit of production falls as you add more units. I'm not convinced that applies to Bitcoin.

Monopolies form because when you're the only one selling a product, you can demand whatever price you want. In Bitcoin mining, if you're a monopoly you can require high transaction fees, and you can keep difficulty artificially low thus lowering your cost.

If you manage to get 51% and force a protocol change that makes only your blocks valid, yeah. If you capture 75% of the mining market, there is still nothing to prevent a small mining business from entering and mining all fees, including ones the monopoly doesn't take. As those unaccepted small fee transactions add up, mining for them becomes profitable as well. If there is something that can give a miner monopoly power and prevent new entrants, I'm not aware of it. Also, how can mining difficulty be kept artificially low?

[Boussac]

You're overlooking what "tragedy of the commons" means. As you said yourself, the payment processor will make 325 € per block whether he mines or not, so he has no incentive to mine himself - he would much rather have others mine instead.

Besides, if payment processors and banks are the only ones doing mining, it means Bitcoin has become fairly centralized.

You are missing the point: the payment porcessor will make 325€ ONLY if the network is functionnal and secure. So the value add fees are his/her incentive to mine (not the bitcoin mining fees).

I'm sorry, but you are. The contribution of the payment processor's own mining to network functionality is negligible, since he's only a small part of it. It's in everyone's best interest that everyone will mine/pay for mining.

The assumption that bitcoin payment processors would rely on volatile, thrid party, volunteer mining and take the risk of losing their business is just plain wrong.
They will mine the hell out of the network because they will want to keep it secure.
At equilibrium, processing 1% of the bitcoin tx will entail x% of the hashing power. With more bitcoin tx (in 40 years ?), the value of processing 1% of the bitcoin tx will exceed the cost of x% of the hashing power.

[Rassah]

On a similar note, does anyone know if MtGox owns any mining hardware or contracts?

[flower1024]

On a similar note, does anyone know if MtGox owns any mining hardware or contracts?

they pay for eligius' server so eligius includes their feeless transactions

[Meni Rosenfeld]

volunteer mining

That's my point, mining should be incentivized rather than volunteer work.

The assumption that bitcoin payment processors would rely on volatile, thrid party, volunteer mining and take the risk of losing their business is just plain wrong.
They will mine the hell out of the network because they will want to keep it secure.
At equilibrium, processing 1% of the bitcoin tx will entail x% of the hashing power. With more bitcoin tx (in 40 years ?), the value of processing 1% of the bitcoin tx will exceed the cost of x% of the hashing power.

I give up.

I've read a lot about monopoly here, as if they are guaranteed, but no real mention of why. Monopolies generally happen only because of economies of scale (http://en.wikipedia.org/wiki/Economies_of_scale, http://en.wikipedia.org/wiki/Natural_monopoly) In short, where the average cost of an extra unit of production falls as you add more units. I'm not convinced that applies to Bitcoin.

Monopolies form because when you're the only one selling a product, you can demand whatever price you want. In Bitcoin mining, if you're a monopoly you can require high transaction fees, and you can keep difficulty artificially low thus lowering your cost.

If you manage to get 51% and force a protocol change that makes only your blocks valid, yeah. If you capture 75% of the mining market, there is still nothing to prevent a small mining business from entering and mining all fees, including ones the monopoly doesn't take. As those unaccepted small fee transactions add up, mining for them becomes profitable as well. If there is something that can give a miner monopoly power and prevent new entrants, I'm not aware of it. Also, how can mining difficulty be kept artificially low?

It depends on how well we can identify the monopoly's blocks and whether we choose to penalize it if it misbehaves. But in principle, as long as the monopoly is in power, it can simply reject any blocks by competitors. Therefore, nobody will try to mine. Which also means the total network hashrate (and hence the difficulty) is its own hashrate, which is as low as it pleases - as long as it's not so low a single entity can easily overthrow it.

[Rassah]

It depends on how well we can identify the monopoly's blocks and whether we choose to penalize it if it misbehaves. But in principle, as long as the monopoly is in power, it can simply reject any blocks by competitors. Therefore, nobody will try to mine. Which also means the total network hashrate (and hence the difficulty) is its own hashrate, which is as low as it pleases - as long as it's not so low a single entity can easily overthrow it.

But... with a 51% attack..... Fuuuuuuu!!!... 
Is this really how this works? 51%, reject any blocks but your own, and the whole network is screwed??