Proof of Proof - an alternative to proof of ___ systems

[Meni Rosenfeld]

Standard disclaimer first:  I am often wrong.

But I've got a nagging feeling that all of the pure Proof-Of-X (where X != Work) systems would set up a dynamic of "the rich and powerful get more rich and more powerful."

The more coins you have, the more you get, as far as I can see in all of the proposed schemes (another disclaimer: I only vaguely pay attention to all of the Proof-of-X schemes, so feel free to tell me how I'm wrong). Seems to me that would end up being a destructive feedback loop, where your decentralized currency naturally gets more and more centralized over time.

I should note that:

1. I don't think it's much different from proof of work - those with resources can buy hardware to mine more coins. Since what you can get is linear in what you put in, I think it will maintain the status quo rather than magnify any gaps.

2. The changes in some proposals relate (as far as monetization goes) just to the transaction fees, not the coinbase. And, since the premise is to make the total cost of securing the network cheap, there shouldn't be huge profits to be made here.

[Explodicle]

One minor note: if/when the block reward is too low and we risk a tragedy of the commons, we can pay for work using assurance contracts.

[detailed example]
So let's say I run a business that saves money by using Bitcoin, or at least hold some Bitcoins. I want a higher difficulty. I can buy "insurance" that the difficulty will be over X, so I get paid if difficulty is low. Miners would be on the other side of these contracts, getting paid if difficulty is high. And bear in mind that we could use PPCoin, Litecoin, or whatever for these contracts in case you're worried that a 51%-induced Bitcoin collapse would make your payout worthless.
[/detailed example]
I suspect (but can't back up) that this:
* Could recover from a 51% attack and re-establish a chain
* Would be cheaper for users than present-day coinbase inflation

We've got some time to develop these secondary markets as the subsidy decreases. IMHO the Coasian provision of public goods is the next big thing for cryptocurrency; it doesn't stop at currency-related problems.

[justusranvier]

One minor note: if/when the block reward is too low and we risk a tragedy of the commons, we can pay for work using assurance contracts.

I'd be very cautious of assuming that we can know who will be mining and for what reasons 10+ years from now.

I can easily imagine a future in which Walmart running a mining pool just to make sure the transactions of its customers and employees get processed quickly. A pool like that might or might not try to recover its costs via transaction fees.

[MoonShadow]

I can easily imagine a future in which Walmart running a mining pool just to make sure the transactions of its customers and employees get processed quickly. A pool like that might or might not try to recover its costs via transaction fees.

Indeed, this 'Walmart' scenario was one of those that I brought up in many previous threads.  If Wal-Mart wanted to provide their customers with a fee-less payment option, while still granting themselves relatively quick access to those funds, they are going to have to sponsor a mining operation that favors transactions into and out of their own address set.  Any major retailer that competes with Walmart would have the exact same motivations, while also desiring to exclude the fee-less transactions of their competitors.  While this is a form of competition for those retailers, each of those mining operations still contribute to the security of the blockchain as a whole, and their support for their chosen mining operations would not (primarily) be connected to the amount of fees that could be collected by the act of mining itself.  This is not a tragedy-of-the-commons scenario.  However, time will tell whether or not this kind of external competition is a significant portion of mining or not.  I imagine that bank-like institutions would also form mining cartels in order to offer fee-less transaction processing to their members; as well as international trade institutions would sponsor mining contracts to protect the value of their 'letters-of-credit' operations using bitcoins, much like a bank invests & maintains a physical bank vault at a loss.

The long point is, PoW is working and we can see that it does work.  There is no evidence, at the present, that it would not continue to work well.  Whereas there is a lack of evidence that PoX methods offered can actually live up to their promises.  Don't fix what ain't broke.

[Meni Rosenfeld]

The long point is, PoW is working and we can see that it does work.  There is no evidence, at the present, that it would not continue to work well.  Whereas there is a lack of evidence that PoX methods offered can actually live up to their promises.  Don't fix what ain't broke.

We can see that if we sponsor PoW with 25% annual monetary inflation it works, yes. Hypothetical scenarios for how it can work in the future are hypothetical.

Anyway, we'll do just what you suggested - research new solutions and try them out in alts, so if it ever is broken we'll be ready with a fix.

Edit: Actually quite a few things are wrong with this Walmart thing, but the main issue is - even if it does make strategic sense for them to mine, this will incur a cost, and to maintain the same level of profitability they will have to increase the prices. So the customers still pay indirect transaction fees as a result of the need to sponsor mining, TANSTAAFL. How much exactly is a subject for a different debate.

[d'aniel]

The long point is, PoW is working and we can see that it does work.  There is no evidence, at the present, that it would not continue to work well.  Whereas there is a lack of evidence that PoX methods offered can actually live up to their promises.  Don't fix what ain't broke.

We can see that if we sponsor PoW with 25% annual monetary inflation it works, yes. Hypothetical scenarios for how it can work in the future are hypothetical.

Anyway, we'll do just what you suggested - research new solutions and try them out in alts, so if it ever is broken we'll be ready with a fix.

The motivation for finding an alternative makes perfect sense; with PoW miners are expending value to the aether, so if this value leak in the bitcoin economy can be plugged by making mining a zero sum game (non-miners included as players), then great.

Edit: not actually zero sum, since the non-miners win by having a cool currency, but you get my drift.

[ShadowOfHarbringer]

Standard disclaimer first:  I am often wrong.

But I've got a nagging feeling that all of the pure Proof-Of-X (where X != Work) systems would set up a dynamic of "the rich and powerful get more rich and more powerful."

The more coins you have, the more you get, as far as I can see in all of the proposed schemes (another disclaimer: I only vaguely pay attention to all of the Proof-of-X schemes, so feel free to tell me how I'm wrong). Seems to me that would end up being a destructive feedback loop, where your decentralized currency naturally gets more and more centralized over time.

You are correct, however what other solutions (besides some kind of cementing) can you produce that can mitigate the risk of 51% attack ?

Also, i have sent you a PM.

[MoonShadow]

Anyway, we'll do just what you suggested - research new solutions and try them out in alts, so if it ever is broken we'll be ready with a fix.

Preparing in the event that an alternative to PoW were to be required, is both wise and rational.  If your methods work out in the alt-chains, and none of the concerns that I have materialize, I'll be one of the first to advocate altering the mainline protocol.

[iddo]

Standard disclaimer first:  I am often wrong.

But I've got a nagging feeling that all of the pure Proof-Of-X (where X != Work) systems would set up a dynamic of "the rich and powerful get more rich and more powerful."

The more coins you have, the more you get, as far as I can see in all of the proposed schemes (another disclaimer: I only vaguely pay attention to all of the Proof-of-X schemes, so feel free to tell me how I'm wrong). Seems to me that would end up being a destructive feedback loop, where your decentralized currency naturally gets more and more centralized over time.

Another way to look at it: the service that the stakeholders provide is even more useful to the health of the network than the service that the miners provide, because the stakeholders have to maintain a full node and therefore they help in preventing network attacks, and they decentralize the power that generates the blocks (the power that synchronizes the txns). Because it costs less to provide this service than it costs to provide the PoW mining service, it might be reasonable to expect that the stakeholders will demand lower fees than the miners (due to the competition among stakeholders), while providing an even more valuable service.

It's true that a stakeholder who never spends his coins will continuously accumulate more coins, but coins are worthless if you never use them. That's also true with fiat currencies that aren't too inflationary, you can have the money sit in the bank and continuously earn interest, without ever using it.

We also have proposals where stakeholders are penalized when they don't help to secure the network, rather than rewarded when they do help.

[misterbigg]

I'm glad that my mocking has stimulated an actual discussion. If anything, we now have some new terminology:

PoW "Proof of Work"
PoX" "Proof of {something other than work}"

If this is the wrong thread to ask this please point me in the right place but how on earth can "Proof of Stake" work in an automated way? Or does it require that the stakeholder (a human) go through all the recent blocks and make sure there's no double spending attacks?

[Meni Rosenfeld]

If this is the wrong thread to ask this please point me in the right place but how on earth can "Proof of Stake" work in an automated way? Or does it require that the stakeholder (a human) go through all the recent blocks and make sure there's no double spending attacks?

You don't need proof of anything to tell if there were double-spends - if there are two transactions spending the same output there's a double spend. (And a computer would be much better than a human in finding that out.)

Proof of work/stake/X exists to signal "this set of transactions is valid and any transaction that is a double-spend on any of them is invalid" in a way that guarantees that everyone agrees on the same set of transactions. As to how this is achieved you'll have to read the specific proposals (e.g. at https://en.bitcoin.it/wiki/Proof_of_Stake), the point is that the stakeholder's computer signs messages that affirm his support of a specific block, and the collection of such signatures is the signal.

[Sergio_Demian_Lerner]

I'd like to point out that in the thread https://bitcointalk.org/index.php?topic=131230.0 (Proof of Bets) we are analyzing a system where Miners have an incentive to cooperate to achieve greater profits, rather than to compete. I don't know if it can be done, but we're trying...

[misterbigg]

You don't need proof of anything to tell if there were double-spends - if there are two transactions spending the same output there's a double spend. (And a computer would be much better than a human in finding that out.)

Yeah, I guess the part that I don't quite understand is what is the incentive for a proof of stake signer to verify the correctness of a series of blocks that he didn't participate in?

[Meni Rosenfeld]

You don't need proof of anything to tell if there were double-spends - if there are two transactions spending the same output there's a double spend. (And a computer would be much better than a human in finding that out.)

Yeah, I guess the part that I don't quite understand is what is the incentive for a proof of stake signer to verify the correctness of a series of blocks that he didn't participate in?

Signature fees. Like the transaction fees we know, but payable to stakeholders providing signatures. Also, signing should be fairly cheap.