What incentives the users/businesses (those who ultimately give value to the coin) have for using a system being centralised by a single miner?
Who spoke of "a single miner"? Understand that the miners have to some extent the ability to maintain a certain illusion of decentralization. I don't propose that bigger blocks will immediately translate to consolidation of miners in hashing power (what the user is really worried about) but it certainly leads to risky behaviours that reduces the security of the network. The cost externalization to nodes is not something most regular users would think twice about yet it severely undermines the nature of Bitcoin. |
|
|
|
Will mining centralize around one single miner? No one knows. This is why Bitcoin is still a risk.
But if this doesn't happen--if there remains more than a single miner--then the fee market does exist.
I think the point of contention is really about this conclusion. My understanding is that it may also happen with several miners if they have an incentive to synchronize their mempools (with a mechanism like IBLT).
I ask, just because I want it to be clear to readers that this "point of contention" is about a hypothetical future scenario. And that we already know that there are hypothetical future scenarios, such as mining centralized around a single super pool, that would be bad for Bitcoin. It absolutely doesnt. Miners already optimize for profit by centralizing in various ways. The recent SPV mining near-catastrophe is an excellent example. Your conclusions hold under no existing and future scenario. It makes assumptions that are untenable and require absolute altruism from miners. The crux of the issue is that miners incentives are not aligned with the network's users. The max block size cap mitigates the fact that they are expected (for security reasons) to prioritize financial profit over network decentralization. Yes, there might be no other way to align these incentives than by forcing it through consensus code. hey! i disagree!!  in my view miners are nothing if it was not for the users they validate the transactions for: users do the speculation, users gives bitcoin its value, which then makes it profitable to mine (and not any other altcoin) . remember what happened btc price wise when ghash.io almost had 50% of the network? massive sell out..  hence, miners do have an incentive to keep on mining the longest VALID chain, which, by definition means the one that has the more value potential. so they better not screwing around with some power grabbing fork, else they loose everything... this is bitcoin's consensus. I agree with most of what you said but I don't believe it really addresses my point. The GHash situation is very different as it was very public. The SPV mining fiasco is a better example of a situation users were not readily aware of yet causes considerable risk to the ecosystem. The miners do have incentives to maintain the users trust but they have a certain flexibility and ability to optimize for profit in a "bend but don't break" manner. Anyway, Peter Todd always makes a good point that we should not build this system based on "best-behaviour" assumptions but rather expect any potential attack. The centralization I refer to relates to geographical and technical optimization that give larger miners a slight advantage over smaller ones that slowly but surely dries out the smaller miners revenue. It is not something that is immediately apparent to most users and the extent to which such centralization can occur is difficult to examine. That is not to suggest they are expect to fork off or create immediate damage but it is a slippery & dark road that can slowly lead to a level of centralization that enables considerable throughput hence the centralization pressure externalized to the network nodes. |
|
|
|
Will mining centralize around one single miner? No one knows. This is why Bitcoin is still a risk.
But if this doesn't happen--if there remains more than a single miner--then the fee market does exist.
I think the point of contention is really about this conclusion. My understanding is that it may also happen with several miners if they have an incentive to synchronize their mempools (with a mechanism like IBLT).
I ask, just because I want it to be clear to readers that this "point of contention" is about a hypothetical future scenario. And that we already know that there are hypothetical future scenarios, such as mining centralized around a single super pool, that would be bad for Bitcoin. It absolutely doesnt. Miners already optimize for profit by centralizing in various ways. The recent SPV mining near-catastrophe was an excellent example. Your conclusions hold under no existing and future scenario. It makes assumptions that are untenable and require absolute altruism from miners. The crux of the issue is that miners incentives are not aligned with the network's users. The max block size cap mitigates the fact that they are expected (for security reasons) to prioritize financial profit over network decentralization. Yes, there might be no other way to align these incentives than by forcing it through consensus code. What are the incentives for miners to centralised more than they currently are? The Ghash episode demonstrated that the market have no incentive for such a thing and have reacted accordingly. Ghash is now a thing of the past. Miners' centralization is not only reflected by a combination of their hashpower. The incentive is clear: mitigate costs derived from creating bigger blocks to capture more transactions fees I am asking the incentives from a market perspective.  Reduce propagation costs of creating bigger blocks. They apparently are already doing it in some form under the existing block size. What exactly do you mean by "from a market perspective"? |
|
|
|
Will mining centralize around one single miner? No one knows. This is why Bitcoin is still a risk.
But if this doesn't happen--if there remains more than a single miner--then the fee market does exist.
I think the point of contention is really about this conclusion. My understanding is that it may also happen with several miners if they have an incentive to synchronize their mempools (with a mechanism like IBLT).
I ask, just because I want it to be clear to readers that this "point of contention" is about a hypothetical future scenario. And that we already know that there are hypothetical future scenarios, such as mining centralized around a single super pool, that would be bad for Bitcoin. It absolutely doesnt. Miners already optimize for profit by centralizing in various ways. The recent SPV mining near-catastrophe was an excellent example. Your conclusions hold under no existing and future scenario. It makes assumptions that are untenable and require absolute altruism from miners. The crux of the issue is that miners incentives are not aligned with the network's users. The max block size cap mitigates the fact that they are expected (for security reasons) to prioritize financial profit over network decentralization. Yes, there might be no other way to align these incentives than by forcing it through consensus code. What are the incentives for miners to centralised more than they currently are? The Ghash episode demonstrated that the market have no incentive for such a thing and have reacted accordingly. Ghash is now a thing of the past. Miners' centralization is not only reflected by a combination of their hashpower. The incentive is clear: mitigate costs derived from creating bigger blocks to capture more transactions fees If, for example, the majority of miners are in China (they are), and there is really poor connectivity in and out of China (there is) and a miner naively optimizes for profit, they will create blocks which are large and take a while to relay out of China. By simple trial-and-error an individual large miner might notice that when they create larger blocks which fork off miners in other parts of the world, they get more income. http://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-May/008364.htmlMoreover, miners now know better than to repeat GHash's mistake. They can easily distribute their hashing power to different pools to maintain a "decentralization theater" |
|
|
|
Using the estimate of Seven Transactions per Second, the network is capable of handling only ~ 604800 transactions a day. That's not even enough for one million people to use on a daily basis. Bitcoin will never be able to replace fiat even if it scales up 100X. or 1000X.
Bitcoin was designed as an electronic peer to peer cash system. Lack of scalability is clearly a severe limit to adoption in that regard. It may be useful for other purposes, but each time it is used for something else, it's capacity as an electronic peer to peer cash system is reduced.
When I first read the story of the Coinwallet stress test, I thought immediately that they were planning to do something harmful, but then I thought about it. Bitcoin is antifragile. It needs to be stressed so that it can grow stronger. Is there even a working process for increasing scalability? We may find out soon. If there isn't one, then knowledge of that deficiency will be the first step towards solving it.
You are not qualified to speak about these issues. Keep to financial speculation please. |
|
|
|
Will mining centralize around one single miner? No one knows. This is why Bitcoin is still a risk.
But if this doesn't happen--if there remains more than a single miner--then the fee market does exist.
I think the point of contention is really about this conclusion. My understanding is that it may also happen with several miners if they have an incentive to synchronize their mempools (with a mechanism like IBLT).
I ask, just because I want it to be clear to readers that this "point of contention" is about a hypothetical future scenario. And that we already know that there are hypothetical future scenarios, such as mining centralized around a single super pool, that would be bad for Bitcoin. It absolutely doesnt. Miners already optimize for profit by centralizing in various ways. The recent SPV mining near-catastrophe is an excellent example. Your conclusions hold under no existing and future scenario. It makes assumptions that are untenable and require absolute altruism from miners. The crux of the issue is that miners incentives are not aligned with the network's users. The max block size cap mitigates the fact that they are expected (for security reasons) to prioritize financial profit over network decentralization. Yes, there might be no other way to align these incentives than by forcing it through consensus code. |
|
|
|
Here is a comment from Matt Corrallo that explains why Peter R's research should not be considered to make our decisions. Again, it might add up in a vacuum under precise conditions but is not applicable to reality. If, for example, the majority of miners are in China (they are), and there is really poor connectivity in and out of China (there is) and a miner naively optimizes for profit, they will create blocks which are large and take a while to relay out of China. By simple trial-and-error an individual large miner might notice that when they create larger blocks which fork off miners in other parts of the world, they get more income. http://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-May/008364.html |
|
|
|
People were always talking about this, dont know why it was never proposed earlier as a BIP. But its very difficult now for any BIP proposal to get accepted regardless of if majority want it because miners will never vote for anything other than BIP 100.
Unless you take a different approach and fork based on 90% or so of node majority rather than using hash majority, but you wouldn't want to do that without first confirming there was already a very strong consensus within the economy for the BIP.
I certainly hope it's not too late, but fear you might be right. I feel out of all the proposals, BIP100 is by far my least preferable, unless it's altered to add an algorithmic element. Give me a simple piece of code over a central entity dictating monetary policy any day. There's still a chance, but we need to get some momentum built up behind this proposal if there's any possibility of BIP106 being chosen. Hopefully it'll be discussed at the scaling conference. Have you considered the block extension proposal? https://www.reddit.com/r/Bitcoin/comments/39kqzs/how_about_a_softfork_optin_blocksize_increase/?sort=confidenceIt might not be the most simple or easy to implement but certain from my point of view the one that satisfies most both sides, in effect creating a solution that respects consensus that there is no consensus.  |
|
|
|
A big problem with Gavin's stuff about his home-grade internet is that it is rather superior than most of the developing world (about 10x at least).
So if we only want Bitcoin to be for the "developed world" then I guess we should all just follow Gavin and Mike.
If we want Bitcoin to be for everyone then we actually need to think a bit harder.
Well, it would stand to reason that less developed places with slower internet would be using more SPV clients. If you reason like Mike Hearn yes. If you respect Bitcoin's ethos, no. |
|
|
|
My paper implies that the most "efficient" network configuration (at least from a superficial perspective) is a single miner in a large data center. No one is disputing this fact. No one is also disputing the fact that a monopoly miner like this has the ability to censor transactions and to double-spend (and earn a greater profit this way too).
Will mining centralize around one single miner? No one knows. This is why Bitcoin is still a risk.
But if this doesn't happen--if there remains more than a single miner--then the fee market does exist.
Great, but that's completely irrelevant to the decisions we have to make. In fact it strongly suggest against removing entirely the block size cap so you should really stop pulling up this material as it absolutely does not support your position. What you essentially propose is to see "how far" the miners will be willing to centralize toward "the most efficient network configuration"My paper says nothing about this. It just shows that a transaction fee market exists without a block size limit assuming the network isn't already centered around a single miner. Now on to a topic outside of the scope of that paper: You are under the impression that increasing the block size limit will for some reason lead to more centralization (which I don't understand--there's "centralization pressure" at any block size limit). I, on the other hand, see this as allowing Bitcoin to grow and gain new users, new nodes, new miners, and in general protection in numbers. Furthermore, you're scheme of implementing a production quota on block size requires centralized control to enforce the quota. You have said it yourself that we need a centrally-planned limit to keep Bitcoin decentralized. Do you not find that logic absurd?
To be exact: My paper says nothing about this. It just shows that a transaction fee market exists without a block size limit assuming the network is perfectly decentralized At which point every incremental increase of the block size encourages equivalent centralization by miner to mitigate costs brought about by propagation time. I am not "under the impression" of anything. Increasing block size limit has a strong centralization pressures on the nodes who necessarily see their cost of maintenance increased considerably. This has the effect of increasing the barrier-to-entry for network governance. There exists no such pressure at current block limit as far as nodes are concerned. While you are right that centralization of mining already exists it is a whole different set of problems in itself and it is easy to understand why these would be precipitated by raising the cap irresponsibly. There is nothing to support your theory that new nodes are brought about by increase use of the network and the last two years show this is absolutely not an assumption you can make. Moreover, you are tragically misguided as to how Bitcoin should grow and WHY it will grow. We have been throught this discussion of centrally-planned limit over and again and no, the logic isn't absurd. To consider: "Are you saying we need centrally planned limit (21,000,000) to keep Bitcoin from turning into fiat"
"Are you saying we need centrally planned block interval to keep Bitcoin secure"
The quota is necessary and effective precisely because the negative externality you make a footnote of in your material is very real and needs to be a focal point of any decision made going forward. |
|
|
|
Strike One: Playing Doubting Thomas to blockchain tech, despite compsci/fintech consensus that (per Horowitz) "it may be the most important computer science breakthrough since packet switching."
And it will end poverty and corruption. And maybe cure cancer too. Strike Two: Declaring XT would win, after getting suckered in by Gavin and Hearn's false sense of urgency and other social engineering attacks.
I don't recall saying that BitcoinXT would win; if I did, I apologize profusely. No matter how much I try, I always overestimate the intelligence of bitcoiners. (I just saw a bunch of them exchange hate comments about Gavin on reddit, as if he had always been the arch-enemy of bitcoin. Amazing what you can do with a smear campaign, if you have 21 M$ of capital to defend and a couple thousand bitcoiner minds to play with.) I still think that the limit will be raised, and the Blockstream guys will either cede or be left sucking their thumbs. But I may be overestimating again... Bonus Strike: Being embarrassingly jealous of Dr. Back's far more lucrative and history-making, world-changingly influential CS career.  After having admired Dr. Back's words, and even having had the honor of debating directly with him on reddit, I must say that, for his technical and ethical qualities, he is one of the most outstanding bitcoin personalities -- right there besides Roger Ver. I do hope that his career develops the way he deserves. Who pays you though? Sometimes I'm curious if you, Lambtroll and a couple others sit at the office cafeteria table and try to think of the most stupid brain dead shit to say |
|
|
|
I'm just gonna drop this here https://botbot.me/freenode/bitcoin-wizards/2015-08-30/?msg=48477664&page=1It's infuriating when someone makes an intellectually weak argument, but slateers it in so much (very nicely constructed) pretext and trappings that people who aren't interested or lack the context to evaluate it on its merits are too busy being mesmorized with the leather binding.  Which paper of Peter's is Greg referring to? I thought Peter is now arguing we don't need any block limit for a "healthy fee market". Is this a different paper? Greg claimed the paper is fundamentally flawed and asked me several times to publicly retract it. Instead, I have examined his objections, and in addressed them I have been able to remove assumptions and strengthen the claims of the paper. The fee market paper is interdisciplinary and really outside the realm of Greg's expertise: it is a mixture of physics and economics. My opinion of Greg is that he is an professional cryptographer, an amateur physicist, and a lousy economist and game theorist. Remember, he already "proved" that decentralized consensus was impossible... And it is, Bitcoin uses carefully aligned incentives to work around the problem. Your paper is fundamentally flawed because it addresses nothing resembling the current dynamics at stake in Bitcoin. More precisely it ignores the incentives for miners to centralize (as they have shown to have) to mitigate propagation times. In effect your paper clearly demonstrates it is more profitable to do so under free-floating blocks and you essentially rely on their altruism to maintain the validity of your model to make decisions going forward. In short, your work might be sound from a technical standpoint, in a vacuum, but can not be used to construct security models that depend on worst-behaviours assumptions. I cannot accept that you could comment on Greg's game theory intelligence when you continue to wave away the clear negative externality present in your models. You chose to ignore the obvious tragedy of commons at stake and that is why your opinion can never be considered as long as you don't recognize the cost externalized to node and the overall centralization pressure suggested by YOUR alignment of the incentives. My paper implies that the most "efficient" network configuration (at least from a superficial perspective) is a single miner in a large data center. No one is disputing this fact. No one is also disputing the fact that a monopoly miner like this has the ability to censor transactions and to double-spend (and earn a greater profit this way too). Will mining centralize around one single miner? No one knows. This is why Bitcoin is still a risk. But if this doesn't happen--if there remains more than a single miner--then the fee market does exist. Great, but that's completely irrelevant to the decisions we have to make. In fact it strongly suggest against removing entirely the block size cap so you should really stop pulling up this material as it absolutely does not support your position. What you essentially propose is to see "how far" the miners will be willing to centralize toward "the most efficient network configuration" The fees SHOULD be irrelevant right now, but some people (in this very thread) are pointing to arguments that we have to have small blocks because of fees, so I think Peter's paper actually is relevant if it dispells those fears. You absolutely don't understand the implications of what Peter's paper suggest . I couldn't careless if fees are irrelevant right now we need a model that takes them into account for the future. |
|
|
|
Practically every sentence Greg types is cryptic and technical.
He's actually excellent at making intelligent analogies and "dumbing-down" material for casual users. |
|
|
|
I'm just gonna drop this here https://botbot.me/freenode/bitcoin-wizards/2015-08-30/?msg=48477664&page=1It's infuriating when someone makes an intellectually weak argument, but slateers it in so much (very nicely constructed) pretext and trappings that people who aren't interested or lack the context to evaluate it on its merits are too busy being mesmorized with the leather binding. Which paper of Peter's is Greg referring to? I thought Peter is now arguing we don't need any block limit for a "healthy fee market". Is this a different paper? Greg claimed the paper is fundamentally flawed and asked me several times to publicly retract it. Instead, I have examined his objections, and in addressed them I have been able to remove assumptions and strengthen the claims of the paper. The fee market paper is interdisciplinary and really outside the realm of Greg's expertise: it is a mixture of physics and economics. My opinion of Greg is that he is an professional cryptographer, an amateur physicist, and a lousy economist and game theorist. Remember, he already "proved" that decentralized consensus was impossible... And it is, Bitcoin uses carefully aligned incentives to work around the problem. Your paper is fundamentally flawed because it addresses nothing resembling the current dynamics at stake in Bitcoin. More precisely it ignores the incentives for miners to centralize (as they have shown to have) to mitigate propagation times. In effect your paper clearly demonstrates it is more profitable to do so under free-floating blocks and you essentially rely on their altruism to maintain the validity of your model to make decisions going forward. In short, your work might be sound from a technical standpoint, in a vacuum, but can not be used to construct security models that depend on worst-behaviours assumptions. I cannot accept that you could comment on Greg's game theory intelligence when you continue to wave away the clear negative externality present in your models. You chose to ignore the obvious tragedy of commons at stake and that is why your opinion can never be considered as long as you don't recognize the cost externalized to node and the overall centralization pressure suggested by YOUR alignment of the incentives. My paper implies that the most "efficient" network configuration (at least from a superficial perspective) is a single miner in a large data center. No one is disputing this fact. No one is also disputing the fact that a monopoly miner like this has the ability to censor transactions and to double-spend (and earn a greater profit this way too). Will mining centralize around one single miner? No one knows. This is why Bitcoin is still a risk. But if this doesn't happen--if there remains more than a single miner--then the fee market does exist. Great, but that's completely irrelevant to the decisions we have to make. In fact it strongly suggest against removing entirely the block size cap so you should really stop pulling up this material as it absolutely does not support your position. What you essentially propose is to see "how far" the miners will be willing to centralize toward "the most efficient network configuration" |
|
|
|
Like ArticMine pointed out, there is already a cost to produce a large spam block, regardless of whether the costs is paid by a user in the form of fees, or by a miner in the form of increased orphan risk.
-
Smooth is also correct, however, that just because the spam block has a finite cost and that a fee market exists, it doesn't necessarily mean that the fees will be the "right" fees to keep people who want to run Bitcoin nodes on low-cost hardware with slow internet connections happy, for example. As the chart above shows, both transaction fees and spam costs (measured in bitcoins) decrease as the propagation impedance of the network improves.
This relates to the question of externalities.
(a) Does retaining the block size limit as an anti-spam measure (i.e., a production quota above the free market production Q*) result in a negative externality? If so, is it worth it to implement a production quota below Q* and suffer a loss of economic activity equal to the region shaded in brown?
(b) If the answer to both questions in (a) is "yes," then the question is whether it is even possible to enforce a production quota without a strong organization or government willing to use force if necessary.
-
Your paper is fundamentally flawed because it addresses nothing resembling the current dynamics at stake in Bitcoin. More precisely it ignores the incentives for miners to centralize (as they have shown to have) to mitigate propagation times. In effect your paper clearly demonstrates it is more profitable to do so under free-floating blocks and you essentially rely on their altruism to maintain the validity of your model to make decisions going forward. In short, your work might be sound from a technical standpoint but can not be used to construct security models that depend on worst-behaviours assumptions.
I cannot accept that you could comment on Greg's game theory intelligence when you continue to wave away the clear negative externality present in your models. You chose to ignore the obvious tragedy of commons at stake and that is why your opinion can never be considered as long as you don't recognize the cost externalized to node and the overall centralization pressure suggested by YOUR alignment of the incentives.
Relevant material: http://pastebin.com/jFgkk8M3https://botbot.me/freenode/bitcoin-wizards/2015-08-30/?msg=48477664&page=1http://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-August/010737.htmlEveryone is free to read and make their own judgment but considerable holes have been poken into Peter's work and he has often been urged to revise his conclusions. He is now parading his charts, illustration all over the forum in an attempt to obtuse the debate, confusing more impressionable users who do not have time, ability or care for validating his propositions. |
|
|
|
I have to wonder if Greg is just trying to discredit people like Peter who threaten his intellectual high ground.
Comments like "Wow. bitcoin-development has reached a new low." really strike me as unwarranted emotionalism.
If you don't chose to make a rational judgment between the arguments presented your opinion is of no use here. |
|
|
|
I'm just gonna drop this here https://botbot.me/freenode/bitcoin-wizards/2015-08-30/?msg=48477664&page=1It's infuriating when someone makes an intellectually weak argument, but slateers it in so much (very nicely constructed) pretext and trappings that people who aren't interested or lack the context to evaluate it on its merits are too busy being mesmorized with the leather binding. Which paper of Peter's is Greg referring to? I thought Peter is now arguing we don't need any block limit for a "healthy fee market". Is this a different paper? Greg claimed the paper is fundamentally flawed and asked me several times to publicly retract it. Instead, I have examined his objections, and in addressed them I have been able to remove assumptions and strengthen the claims of the paper. The fee market paper is interdisciplinary and really outside the realm of Greg's expertise: it is a mixture of physics and economics. My opinion of Greg is that he is an professional cryptographer, an amateur physicist, and a lousy economist and game theorist. Remember, he already "proved" that decentralized consensus was impossible... And it is, Bitcoin uses carefully aligned incentives to work around the problem. In fact, if you wanna accuse him of being wrong on this aspect, then you should realize you are making the same mistakes he previously did which is to examine the problem in a vaccum. Your paper is fundamentally flawed because it addresses nothing resembling the current dynamics at stake in Bitcoin. More precisely it ignores the incentives for miners to centralize (as they have shown to have) to mitigate propagation times. In effect your paper clearly demonstrates it is more profitable to do so under free-floating blocks and you essentially rely on their altruism to maintain the validity of your model to make decisions going forward. In short, your work might be sound from a technical standpoint but can not be used to construct security models that depend on worst-behaviours assumptions. I cannot accept that you could comment on Greg's game theory intelligence when you continue to wave away the clear negative externality present in your models. You chose to ignore the obvious tragedy of commons at stake and that is why your opinion can never be considered as long as you don't recognize the cost externalized to node and the overall centralization pressure suggested by YOUR alignment of the incentives. |
|
|
|
I'm just gonna drop this here https://botbot.me/freenode/bitcoin-wizards/2015-08-30/?msg=48477664&page=1It's infuriating when someone makes an intellectually weak argument, but slateers it in so much (very nicely constructed) pretext and trappings that people who aren't interested or lack the context to evaluate it on its merits are too busy being mesmorized with the leather binding. Which paper of Peter's is Greg referring to? I thought Peter is now arguing we don't need any block limit for a "healthy fee market". Is this a different paper? No, it is referring to the same paper. If you want to understand the assumptions that make any of what Peter has posted here wrong I suggest you read the above. Sometimes it is nice to have two sides of the story. |
|
|
|
This recent passage from the bitcoin-wizards channel is worthy of mention here: CodeShark the blockchain is there to protect you in the event of an uncooperative counterparty
gmaxwell A more useful mental model for the system is that the network is a trustworthy AI Judge that makes sure you complied with the contracts specified in your contracts. Now, it's possible to transact by taking every contract to the judge, but this is inefficient.
CodeShark right, the blockchain is like a court
CodeShark you don't go to court over every contract you enter
CodeShark only the ones where there's a breach |
|
|
|
I'm just gonna drop this here https://botbot.me/freenode/bitcoin-wizards/2015-08-30/?msg=48477664&page=1It's infuriating when someone makes an intellectually weak argument, but slateers it in so much (very nicely constructed) pretext and trappings that people who aren't interested or lack the context to evaluate it on its merits are too busy being mesmorized with the leather binding. |
|
|
|
|
Show Posts
