Topic: Blockchain = Powerful Tool for Keynesian Monetary Policy - page 3. (Read 11479 times)

If such a patch is coded, how would we know it is applied benevolently and affirm that it is genuine? If a 51% attack is overt, this might work, but it's more likely that such an attack will be covert and not immediately noticed. It would be better to simply guard against it by adding more hashrate. That is not a matter of how, but who will do it?

Rassah,
If you made some effort to understand cunicula's posts, you'd have realised that the mining business with PoW is very similar to saving in PoS: you invest money into it and are rewarded with newly-minted coins for that.

The differences are strongly in favour of PoS though:
* With PoW minting wastes power, with PoS it doesn't.
* With PoS mining returns are predictable and fair (proportional to the capital invested), with PoW it's more like a gamble (some people make a lot, others take a loss).

I don't know why you must resort to insults instead of making your point clear.

cunicula, I salute your patience!

Rudd-O, myrkul, bad, bad trolls.

I concur. That is why I added him to my shitlist and he gets no more replies from me. My time is far too precious to squander on the likes of that fool.

I find that to be entirely likely. The level of statism in his posts exceeds even the most brainwashed posters I've talked to, and approaches Colbert Report-style parody.

Nope. My argument is based on you being a rude asshole who thinks your specific flavor of Keynesian economic theory is the only correct one, and your belief that anyone who questions it must be an idiot not even worth talking to. That's pretty much been your entire argument on the whole forum. So my entire counter-argument to you will always be, "No, you're the idiot *sticks out tongue*."
Other than that, I honestly can't imagine getting anything substantial or informative out of you. Debate is give and take, point and counterpoint, where each side considers the merits of the other side before refuting them. It's a skill you sadly lack. Either that, or you're trolling.

The linked post is about the central bank and does not mention PoS at all. Sigh, so you weren't asking questions about proof-of-stake? Do you even know what you are asking?

Your argument about deflation is a red herring. If the rewards are arbitrarily small, any type of deflationary effect associated with the reward system is also arbitrarily small. Regardless of whether you think deflation is bad or good, the effect can be negligible. If you want to argue that something is worthy of consideration, you need to make a case that it has a non-negligible effect. You don't seem to understand this.

Also, I am a bit confused. Is your argument based on a dislike for my avatar? I understand that you are a furry. As is referenced in your avatar. I don't have a problem with that at all. More power to you. Do you think it is reasonable for me to base my argument on the picture of the cat in your avatar?

When you said:


I was going to reply with, "Are you serious?!" and quote this


But now that I know that you apparently are only interested in abstract concepts, instead of real concrete ones, that explains everything about your weird posts. It even (almost) explains you promoting the 1% savings reward, and saying things like even though this reward will make a $100 TV cost $99 a year later, that deflation isn't a bad thing, which is really weird coming from Mr. "Krugman is tired of trying to reason blahblahblah" Keynesian like you.

I repeat.
'Does it contain any abstract reasoning concerning quantity or number? No. ... Commit it then to the flames: for it can contain nothing but sophistry and illusion.'

Yes, rewarding saving can encourage hording. But numbers are important here. Let's think about the nature of a 1% per annum reward.

You could purchase a television now for $100, or you could purchase it one year from now for $99. You think that you will live in privation for the next year for the sake of a $1 discount?

The $1 discount is an upper bound, it would be much lower than this if other people also save. If so, there will be inflation at up to 1% a year. At the limit, if everyone saves, I get no benefit from saving at all. There is some kind of stable equilibrium in between. I'm not sure where.

As far as a deflationary spiral goes, it is really no different from bitcoin. The additional 1% per annum max is pretty trivial. We could go about analyzing it, but then you need a Keynesian model. I'm not going to do that here because it would be a charade. No one here is interested in Keynesian models (including me).

[Rassah also mentions a 3% fee, but I have no idea where this comes from so I ignore it.]

Dafuc?? Saving in PoW requires a conscious decision to not spend your savings. That's it!

Yes, PoS incentivizes saving, or hoarding, or whatever, but doesn't that also mean it will icentivise a deflationary spiral? Or will the 1% you make by saving be counteracted by 3% you lose on spending as in your OP? And wouldn't that 3% fee plus 1% saving reward mean that no one will want to spend their money any more ever?

Rassah. Let's consider your argument seriously. It has some valid points.

You are arguing that PoS transfers additional wealth to savers. It is actually a sound argument. Saving in PoW requires a conscious decision to reinvest mining proceeds. Saving in PoS occurs by default. A wide body of research suggests that setting up default choices for people influences their behavior. See the pop-economics book "Nudge."

Your argument also has some serious weaknesses. Look at the quotation in my signature line, "Does it contain any abstract reasoning concerning quantity or number? No. .... Commit it then to the flames: for it can contain nothing but sophistry and illusion." What your argument is missing is some 'reasoning concerning quantity or number'. In particular, we don't know if the effect you are describing could be quantitatively important. If fees are near-zero, the money given to stakeholders will also be near zero. Thus, the quantitative significance of this phenomenon could be negligible. If so, we can safely ignore the argument.

To make things more concrete let's discuss the phenomenon in terms of PPCoin, the only operational PoS system. PPCoin awards 1% interest per annum to anyone who saves. This is the PoS award. Since fees are destroyed there is no other way for stakeholders to earn money.  How long would it take for a determined saver to double their holdings with this system? 70 years. That is also the minimum possible time for the money supply to double.

I own about 0.2% of PPCoin right now. Let's ignore PoW reward and assume issuance stops today. Let's also pretend that no one else tries to save. I'm the only one. How long will it take me to accumulate 51% from my 0.2% holding? 560 years of waiting. Suppose that everyone else also tries to save. Well, then I never acquire 51% and am stuck at 0.2% for eternity. So the amount of time necessary for me to wait is between 560 years at minimum and infinity at maximum.

I'm not going to live that long and I don't think cryptocurrency will either. Your concern seems to be irrelevant in practice. It is quite silly to worry about something that takes 560 years minimum to become a potential threat. Even the proposed 560 years is silly. It is based on the premises that: a) 20 generations of my descendants save these holdings and pass them on to the next generation. And simultaneously that b) no one outside my immediate family saves at all for the next 560 years.

What I was describing was essentially a Proof-of-Stake system, where the blocks you mine and the fees you collect are themselves an increase in your stake, and thus as if you are mining for more mining equipment. It really is a piece of shit system.

Two points, first is that the point of mining is to reveal bitcoins. Where do you propose these mined bitcoins go? Secondly, Bitcoin is an open protocol, there would be nothing to prevent the govt supplying as many of their own ASICs as they wanted.

You're falling for the ECB leading with a "Think of the children" ploy?

Identification is made harder as the attacker could relay blocks and appear to not have 51%.

However: IF the "attacker" is behaving nicely with reasonable fees, no reversals and even allows the blocks of competitors you are not really under a 51% attack.

Identification would be easy once you can't transact, fees are crazy, reversals happen or such.

Excellent point.

While I have many times said it would be "easy" to block a 51% attacker (such as government) it is NOT without cost. By locking out part of the network you slow things down and create forks between clients trusting different nodes.

Unless a full 51% attack is actually being executed, it makes no sense to the individual Bitcoiner to lock out anyone.

I have an AWESOME idea for how to improve a Proof-of-Work system:

Instead of getting a 25BTC block reward, have everyone who mines a block automatically receive a small mining ASIC instead. Also, make it so that no one can receive ASICs other than through block mining. Want to increase your hashing power and get more mining hardware? Mining a block is your only way. The best feature is that if you manage to get 51% control and process a majority of blocks, you can improve your hashing rate, and control more and more of the blockchain, just by continuing to mine, since you will keep getting more and more mining hardware just by mining blocks. And the great thing is that no one can take control from you, since the only way to get ASICs is by mining blocks, which is something you can do at a way higher rate than anyone else; a rate that only keeps increasing. If you wait long enough, you can even get to control 99% of the blockchain, pretty much by default, and if you want some money, just sell some of your ASICs.

What do you think? In a system like that, it would be fairly easy for someone to get control and then hold on to it and become rich.

I guess that depends on where you believe that the power of government comes from. If Bitcoin is the will of the people, then it will happen. If drugs are the will of the people, then they will happen. Look at the prohibition of alcohol. Maybe some faction of a government will declare bitcoin illegal, but that doesn't mean they have the moral imperative. Bitcoin is here, and folks are gonna like it whether it's good or not.

I don't think there is a complete defense. However, you can make attacks more difficult and less damaging. You can protect against a 99.999% work attack, but if you do this, you have to allow for 51% stake attacks. There is no way of assuring that only good guys get to vote.

I'd recommend the following approaches:

1) Implement proof-of-stake [drastically increases attack costs; lowers future fee burden on user base; insures users against attack risk] (implementation details to complex to discuss here)
2) Decouple fees and block size [allocating block space based on bidding will not work; Fees are primarily there for security reasons, not scarcity of space. It makes no sense to arbitrarily link security to block size.]
          a) Allow maximum block size to grow with time (say 20% per annum)
          b) Introduce formulaic mandatory fees; enforce fee rules as a block validity criterion (implementation details too complex to discuss here)

The advantage of formulaic fees is that they would reduce the bargaining power of miners to negotiate higher fees.  Miners could still demand bribes to include txns in blocks, but these bribes would have to be separate txns. I think users would find bribe payments more abhorrent than simple fee increases, particularly if there was no user-set fee to begin with.

Formulaic fees also make the mining reward predictable and a predictable reward leads to predictable security. This holds for both PoS and PoW-based systems. Thus even if bitcoin persists with insecure/inefficient/risky PoW, formulaic fees would still be an improvement. The PoW fees just need to be 10 or 20 times higher than PoS fees in order to provide adequate security.

The current limit on block size leads to a very unpredictable security outcome (e.g. onerous fees and high security, no fees and no security, onerous fees and inadequate security, etc.). Of course, block size limits also arbitrarily restrict bitcoin's potential applications.

I can see that, in a future with thousands of transactions per second, today's TOR network couldn't supply enough bandwidth. I've run a full node over tor for a long time, and I've never had any problems.

I'm trying to differentiate between *identifying* a 51% attack and *defending* against it. Everybody on this forum will happily wax lyrical about how hard it would be for govt to eliminate bitcoin. And yet, nobody seems to notice the contradiction in declaring how easy it would be to exclude govt from bitcoin.

Good post. I would add that some defence against a 51% attack should be implemented ASAP. If I were a central bank, I'd be buying up as many ASICs as possible to gain > 51% hashing power, or better still, > 90%. For the moment, then, I'd leave them hashing according to the standard rules and not make any hostile moves until I were certain how best to proceed.

Nice idea   However:


I bet you'll find that most miner's motivations are not exactly that. 

Some other purposes of printing money:  greed, warfare, space program, control of population, to pay off debt (i.e. bailouts), etc.. 

Cunicula does have insightful comments, and to a certain extent, I agree with many of his arguments. Unfortunately, he suffers from the Keynesian illusions of grandeur and ascribes omniscience and omnipotence to regulators.

Now, to a certain extent, his argument that the government can theoretically "hijack" the blockchain is plausible (I'll leave the economics of it by side). In order for it to work, however, two requirements are necessary which were not addressed sufficiently:

• The control over hashing power is insufficient, control over large parts of auxiliary infrastructure is necessary
• It must be uneconomical for market actors to differentiate between "govcoin" and "bitcoin"

First of all, if someone in charge of "govcoin" decides to force their own blockchain with transaction fees above the market price, the result would be that people won't be able to make payments, full stop. The reason is that the devices generating the transaction need to know about this restriction. At the moment, the transaction fees are partially hardcoded and partially left up to the end-user. The "govcoin" operators need to replace a significant proportion of this infrastructure with devices they can control. This can't be done stealthily.

Second of all, if the only action is to simply set a lower bound on fees, e.g. 3%, the operators of mining pools can simply decide to ignore all transactions and blocks that contain a 3% or higher fee. This would create a fork in the blockchain, and the resulting "govcoin" and "bitcoin" would have a floating exchange rate against each other. Even if the "govcoin" operators can maintain 51% strenght, they can't force anything they want upon others.

There are still open questions, like the exact nature of the transaction fee market price distribution mechanism (which does not exist at the moment), but in general, there are practical logistical hurdles that "govcoin" would need to handle, in addition to 51% of hashing power.

In summary, Cunicula's argument has some merit. It is hypothetically possible to do what he suggests. Therefore, my recommendation is to invest into the creation of open source block analysis tools, open source tools for analysing the transaction fee equilibrium and integrating them into mining pool bitcoind backends. These will allow to detect 51% attacks and automatically mitigate against them.