From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from smtp4.osuosl.org (smtp4.osuosl.org [140.211.166.137]) by lists.linuxfoundation.org (Postfix) with ESMTP id AA418C0001 for ; Fri, 28 May 2021 14:40:55 +0000 (UTC) Received: from localhost (localhost [127.0.0.1]) by smtp4.osuosl.org (Postfix) with ESMTP id A5E3C4040E for ; Fri, 28 May 2021 14:40:55 +0000 (UTC) X-Virus-Scanned: amavisd-new at osuosl.org X-Spam-Flag: NO X-Spam-Score: -2.099 X-Spam-Level: X-Spam-Status: No, score=-2.099 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_NONE=-0.0001, RCVD_IN_MSPIKE_H3=0.001, RCVD_IN_MSPIKE_WL=0.001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Authentication-Results: smtp4.osuosl.org (amavisd-new); dkim=pass (1024-bit key) header.d=protonmail.com Received: from smtp4.osuosl.org ([127.0.0.1]) by localhost (smtp4.osuosl.org [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id McryF3y2GA4M for ; Fri, 28 May 2021 14:40:53 +0000 (UTC) X-Greylist: domain auto-whitelisted by SQLgrey-1.8.0 Received: from mail-40141.protonmail.ch (mail-40141.protonmail.ch [185.70.40.141]) by smtp4.osuosl.org (Postfix) with ESMTPS id 8957E406AE for ; Fri, 28 May 2021 14:40:52 +0000 (UTC) Date: Fri, 28 May 2021 14:40:45 +0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=protonmail.com; s=protonmail; t=1622212847; bh=yCs4q3Ii1UmVKGiwG9rT6pJg+pQou3fGeDWdDJBjqz0=; h=Date:To:From:Cc:Reply-To:Subject:In-Reply-To:References:From; b=TUmN4aBV6H/8Ung+Yris5LxGiPmK8pTpkE8IAGgI6wAeud69FQeBik/hBZjnhfmvr oFP/tfqlwHVJToIMjR8Yq4xMDia+d+1e/W72jJLfUu+lDItzZjfmYMGYwhJZrALOf2 6ViXU1FVLHlrTQ9wgGa8naMNUGv/4XDCkQ/l3UqI= To: Erik Aronesty From: befreeandopen Reply-To: befreeandopen Message-ID: In-Reply-To: References: <6do5xN2g5LPnFeM55iJ-4C4MyXOu_KeXxy68Xt4dJQMhi3LJ8ZrLICmEUlh8JGfDmsDG12m1JDAh0e0huwK_MlyKpdfn22ru3zsm7lYLfBo=@protonmail.com> <3TVoontwJmoMv0tp1S5MU_U8icxcQZfajtbNEXqOjuvO7GpfUQdh9pEGSIbLEYJndrDa_dJQqa0sSwY-BmuCmyHMRWqa9lEaUjZJSP5Vbyw=@protonmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-Mailman-Approved-At: Fri, 28 May 2021 15:12:31 +0000 Cc: Bitcoin Protocol Discussion , SatoshiSingh , Billy Tetrud Subject: Re: [bitcoin-dev] Opinion on proof of stake in future X-BeenThere: bitcoin-dev@lists.linuxfoundation.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: Bitcoin Protocol Discussion List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 May 2021 14:40:55 -0000 Erik, I am sorry, I have little knowledge about proof-of-burn, I never foun= d it interesting up until now. Some of your recent claims seem quite strong= to me and I'd like to read more. Forgive me if this has been mentioned recently, but is there a full specifi= cation of the concept you are referring to? I don't mean just the basic ide= a description (that much is clear to me), I mean a fully detailed proposal = or technical documentation that would give me a precise information about w= hat exactly it is that you are talking about. Sent with ProtonMail Secure Email. =E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90 Original Me= ssage =E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90=E2=80=90 On Wednesday, May 26, 2021 11:07 PM, Erik Aronesty wrote: > note: the "nothing at stake" problem you propose is not broken for > proof-of-burn, because the attacker > > a) has no idea which past transactions are burns > b) has no way to use his mining power, even 5%, to maliciously improve > his odds of being selected > > On Wed, May 26, 2021 at 9:12 AM befreeandopen > befreeandopen@protonmail.com wrote: > > > @befreeandopen I guess I misunderstood your selfish minting attack. Let= me make sure I understand it. You're saying it would go as follows?: > > > > 1. The malicious actor comes across an opportunity to mint the next 3 = blocks. But they hold off and don't release their blocks just yet. > > 2. They receive a new block minted by someone else. > > 3. The malicious actor then chooses to release their other 2 blocks on= on the second from the top block if it gives them more blocks in the futur= e than minting on the top block. And instead lets the top block proceed if = it gives them more blocks in the future (also figuring in the 3 blocks they= 're missing out on minting). > > 4. Profit! > > > > The problem with this attack is that any self respecting PoS system wou= ldn't have the information available for minters to know how blocks will af= fect their future prospects of minting. Otherwise this would introduce the = problem of stake grinding. This can be done using collaborative randomness = (where numbers from many parties are combined to create a random number tha= t no individual party could predict). In fact, that's what the Casper proto= col does to decide quorums. In a non quorum case, you can do something like= record a hash of a number in the block header, and then have a second step= to release that number later. Rewards can be given can be used to ensure m= inters act honestly here by minting messages that release these numbers and= not releasing their secret numbers too early. > > Yes, you misunderstood it. First, let me say that the above thoughts of= yours are incorrect, at least for non-quorum case. Since the transition in= the blockchain system from S1 to S2 is only by adding new block, and since= stakers always need to be able to decide whether or not they can add the n= ext block, it follows that if a staker creates a new block locally, she can= decide whether the new state allows her to add another block on top. As yo= u mentioned, this COULD introduce problem of staking, that you are incorrec= t in that it is a necessity. Usual prevention of the grinding problem in th= is case is that an "old enough" source of randomness applies for the curren= t block production process. Of course this, as it is typical for PoS, intro= duces other problems, but let's discard those. > > I will try to explain in detail what you misunderstood before. You star= t with a chain ending with blocks A-B-C, C being the top, the common featur= e of PoS system (non-quorum), roughly speaking, is that if N is the total a= mount of coins that participate in the staking process to create a new bloc= k on top of C (let's call that D), then a participant having K*N amount of = stake has chance K to be the one who will create the next stake. In other w= ords, the power of stakers is supposed to be linear in the system - you own= 10 coins gives you 10x the chance of finding block over someone who has 1 = coin. > > What i was claiming is that using the technique I have described, this = linearity is violated. Why? Well, it works for honest stakers among the com= petition of honest stakers - they really do have the chance of K to find th= e next block. However, the attacker, using nothing at stake, checks her abi= lity to build block D (at some timestamp). If she is successful, she does n= ot propagate D immediately, but instead she also checks whether she can bui= ld on top of B and on top of A. Since with every new timestamp, usually, th= ere is a new chance to build the block, it is not uncommon that she finds s= he is indeed able to build such block C' on top of B. Here it is likely t(C= ') > t(C) as the attacker has relatively low stake. Note that in order to p= roduce such C', she not only could have tried the current timestamp t(D), b= ut also all previous timestamps up to t(B) (usually that's the consensus ru= le, but it may depend on a specific consensus). So her chance to produce su= ch C' is greater than her previous chance of producing C (which chance was = limited by other stakers in the system and the discovery of block C by one = of them). Now suppose that she found such C' and now she continues by tryin= g to prolong this chain by finding D'. And again here, it is quite likely t= hat her chance to find such D' is greater than was her chance of finding D = because again there are likely multiple timestamps she could try. This all = was possible just because nothing at stake allows you to just try if you ca= n produce a block in certain state of block chain or not. Now if she actual= ly was able to find D', she discards D and only publishes chain A-B-C'-D', = which can not be punished despite the fact that she indeed produced two dif= ferent forks. She can not be punished because this production was local and= only the final result of A-B-C'-D' was published, in which case she gained= an extra block over the honest strategy which would only give her block D. > > Fun fact tho: there is an attack called the "selfish mining attack" for= proof of work, and it reduces the security of PoW by at least 1/3rd. > > How is that relevant to our discussion? This is known research that has= nothing to do with PoS except that it is often worse on PoS. > > > > > the problem is not as hard as you think > > > > I don't claim to know just how hard finding the IP address associated w= ith a bitcoin address is. However, the DOS risk can be solved more complete= ly by only allowing the owner of coins themselves to know whether they can = mint a block. Eg by determining whether someone can mint a block based on t= heir public key hidden behind hashes (as normal in addresses). Only when so= meone does in fact mint a block do they reveal their hidden public key in o= rder to prove they are allowed to mint the block. > > This is true, but you are mixing quorum and non-quorum systems. My obje= ction here was towards such system where I specifically said that the list = of producers for next epoch is known up front and you confirmed that this i= s what you meant with "quorum" system. So in such system, I claimed, the kn= own producer is the only target at any given point of time. This of course = does not apply to any other type of system where future producers are not k= nown. No need to dispute, again, something that was not claimed. > > > > > I agree that introduction of punishment itself does not imply introdu= cing a problem elsewhere (which I did not claim if you reread my previous m= essage) > > > > I'm glad we agree there. Perhaps I misunderstood what you meant by "you= should not omit to mention that by doing so, typically, you have introduce= d another problem elsewhere." > > Perhaps you should quote the full sentence and not just a part of it: > > "Of course you can always change the rules in a way that a certain spec= ific attack is not doable, but you should not omit to mention that by doing= so, typically, you have introduced another problem elsewhere, or you have = not solved it completely." > > You can parse this as: (CREATE PROBLEM ELSEWHERE) OR (NOT SOLVE IT COMP= LETELY) > > In case of the punishment it was meant to be the not solve it completel= y part. > > Also "typically" does not imply always. > > But this parsing of English sentences for you seems very off topic here= . My point is, in context of Bitcoin, reject such unsupported claims that P= oS is a reasonable alternative to PoW, let's stick to that. > > > > > As long as the staker makes sure (which is not that hard) that she do= es not miss a chance to create a block, her significance in the system will= always increase in time. It will increase relative to all normal users who= do not stake > > > > Well, if you're in the closed system of the cryptocurrency, sure. But w= e don't live in that closed system. Minters will earn some ROI from minting= just like any other financial activity. Others may find more success spend= ing their time doing things other than figuring out how to mint coins. In t= hat case, they'll be able to earn more coin that they could later decide to= use to mint blocks if they decide to. > > This only supports the point I was making. Since the optimal scenario w= ith all existing coins participating is just theoretical, the attacker's po= sition will ever so improve. It seems we are in agreement here, great. > > > > > Just because of the above we must reject PoS as being critically inse= cure > > > > I think the only thing we can conclude from this is that you have come = up with an insecure proof of stake protocol. I don't see how anything you'v= e brought up amounts to substantial evidence that all possible PoS protocol= s are insecure. > > I have not come up with anything. I'm afraid you've not realized the bu= rden of proof is on your side if you vouch for a design that is not believe= d and trusted to be secure. It is up to you to show that you know how to so= lve every problem that people throw at you. So far we have just demonstrate= d that your claim that nothing at stake is solved was unjustified. You have= not described a system that would solve it (and not introduce critical DDO= S attack vector as it is in quorum based systems - per the prior definition= of such systems). > > Of course the list of problems of PoS systems do not end with just noth= ing at stake, but it is good enough example that by itself prevents its ado= ption in decentralized consensus. No need to go to other hard problems with= out solving nothing at stake. > > On Tue, May 25, 2021 at 11:10 AM befreeandopen befreeandopen@protonmail= .com wrote: > > > > > @befreeandopen " An attacker can calculate whether or not she can pro= long this chain or not and if so with what timestamp." > > > The scenario you describe would only be likely to happen at all if th= e malicious actor has a very large fraction of the stake - probably quite c= lose to 50%. At that point, you're talking about a 51% attack, not the noth= ing at stake problem. The nothing at stake problem is the problem where any= one will mint on any chain. Its clear that if there's a substantial punishm= ent for minting on chains other than the one that eventually wins, every mi= nter without a significant fraction of the stake will be honest and not att= empt to mint on old blocks or support someone else's attempt to mint on old= blocks (until and if it becomes the heaviest chain). Because the attacker = would need probably >45% of the active stake (take a look at the reasoning = here for a deeper analysis of that statement), I don't agree that punishmen= t is not a sufficient mitigation of the nothing at stake problem. To exploi= t the nothing at stake problem, you basically need to 51% attack, at which = point you've exceeded the operating conditions of the system, so of course = its gonna have problems, just like a 51% attack would cause with PoW. > > > This is not at all the case. The attacker benefits using the describe= d technique at any size of the stake and significantly so with just 5% of t= he stake. By significantly, I do not mean that the attacker is able to comp= letely take control the network (in short term), but rather that the attack= er has significant advantage in the number of blocks she creates compared t= o what she "should be able to create". This means the attacker's stake incr= eases significantly faster than of the honest nodes, which in long term is = very serious in PoS system. If you believe close to 50% is needed for that,= you need to redo your math. So no, you are wrong stating that "to exploit = nothing at stake problem you basically need to 51% attack". It is rather th= e opposite - eventually, nothing at stake attack leads to ability to perfor= m 51% attack. > > > > > > > I am not sure if this is what you call quorum-based PoS > > > > > > Yes, pre-selected minters is exactly what I mean by that. > > > > > > > it allows the attacker to know who to attack at which point with po= werful DDOS in order to hurt liveness of such system > > > > > > Just like in bitcoin, associating keys with IP addresses isn't genera= lly an easy thing to do on the fly like that. If you know someone's IP addr= ess, you can target them. But if you only know their address or public key,= the reverse isn't as easy. With a quorum-based PoS system, you can see the= ir public key and address, but finding out their IP to DOS would be a huge = challenge I think. > > > I do not dispute that the problem is not trivial, but the problem is = not as hard as you think. The network graph analysis is a known technique a= nd it is not trivial, but not very hard either. Introducing a large number = of nodes to the system to achieve very good success rate of analysis of are= a of origin of blocks is doable and has been done in past. So again, I very= much disagree with your conclusion that this is somehow secure. It is abso= lutely insecure. > > > Note, tho, that quorum-based PoS generally also have punishments as p= art of the protocol. The introduction of punishments do indeed handily solv= e the nothing at stake problem. And you didn't mention a single problem tha= t the punishments introduce that weren't already there before punishments. = There are tradeoffs with introducing punishments (eg in some cases you migh= t punish honest actors), but they are minor in comparison to solving the no= thing at stake problem. > > > While I agree that introduction of punishment itself does not imply i= ntroducing a problem elsewhere (which I did not claim if you reread my prev= ious message), it does introduce additional complexity which may introduce = problem, but more importantly, while it slightly improves resistance agains= t the nothing at stake attack, it solves absolutely nothing. Your claim is = based on wrong claim of needed close to 50% stake, but that could not be fa= rther from the truth. It is not true even in optimal conditions when all pa= rticipants of the network stake or delegate their stake. These optimal cond= itions rarely, if ever, occur. And that's another thing that we have not me= ntion in our debate, so please allow me to introduce another problem to PoS= . > > > Consider what is needed for such optimal conditions to occur - all co= ins are always part of the stake, which means that they need to somehow aut= omatically part of the staking process even when they are moved. But in man= y PoS systems you usually require some age (in terms of confirmations) of t= he coin before you allow it to be used for participation in staking process= and that is for a good reason - to prevent various grinding attacks. In so= me systems the coin must be specifically registered before it can be staked= , in others, simply waiting for enough confirmations enables you to stake w= ith the coin. I am not sure if there is a system which does not have this c= ooling period for a coin that has been moved. Maybe it is possible though, = but AFAIK it is not common and not battle tested feature. > > > Then if we admit that achieving the optimal condition is rather theor= etical. Then if we do not have the optimal condition, it means that a stake= r with K% of the total available supply increases it's percentage over time= to some amounts >K%. As long as the staker makes sure (which is not that h= ard) that she does not miss a chance to create a block, her significance in= the system will always increase in time. It will increase relative to all = normal users who do not stake (if there are any) and relative to all other = stakers who make mistakes or who are not wealthy enough to afford not selli= ng any position ever. But powerful attacker is exactly in such position and= thus she will gain significance in such a system. The technique I have des= cribed, and that you mistakenly think is viable only with huge amounts of s= take, only puts the attacker to even greater advantage. But even without th= e described attack (which exploits nothing at stake), the PoS system conver= ges to a system more and more controlled by powerful entity, which we can a= ssume is the attacker. > > > So I don't think it is at all misleading to claim that "nothing at st= ake" is a solved problem. I do in fact mean that the solutions to that prob= lem don't introduce any other problems with anywhere near the same level of= significance. > > > It still stands as truly misleading claim. I disagree that introducin= g DDOS opportunity with medium level of difficulty for the attacker to impl= ement it, in case of "quorum-based PoS" is not a problem anywhere near the = same level of significance. Such an attack vector allows you to turn off th= e network if you spend some time and money. That is hardly acceptable. > > > Just because of the above we must reject PoS as being critically inse= cure until someone invents and demonstrates an actual way of solving these = issues. > > > On Tue, May 25, 2021 at 3:00 AM Erik Aronesty erik@q32.com wrote: > > > > > > > > > you burn them to be used at a future particular block height > > > > > > > > > This sounds exploitable. It seems like an attacker could simply f= ocus all their burns on a particular set of 6 blocks to double spend, minim= izing their cost of attack. > > > > > > > > could be right. the original idea was to have burns decay over time= , > > > > like ASIC's. > > > > anyway the point was not that "i had a magic formula" > > > > the point was that proof of burn is almost always better than proof= of > > > > stake - simply because the "proof" is on-chain, not sitting on a no= de > > > > somewhere waiting to be stolen. > > > > On Mon, May 24, 2021 at 9:53 PM Billy Tetrud billy.tetrud@gmail.com= wrote: > > > > > > > > > Is this the kind of proof of burn you're talking about? > > > > > > > > > > > if i have a choice between two chains, one longer and one short= er, i can only choose one... deterministically > > > > > > > > > > What prevents you from attempting to mine block 553 on both chain= s? > > > > > > > > > > > miners have a very strong, long-term, investment in the stabili= ty of the chain. > > > > > > > > > > Yes, but the same can be said of any coin, even ones that do have= the nothing at stake problem. This isn't sufficient tho because the chain = is a common good, and the tragedy of the commons holds for it. > > > > > > > > > > > you burn them to be used at a future particular block height > > > > > > > > > > This sounds exploitable. It seems like an attacker could simply f= ocus all their burns on a particular set of 6 blocks to double spend, minim= izing their cost of attack. > > > > > > > > > > > i can imagine scenarios where large stakeholders can collude to= punish smaller stakeholders simply to drive them out of business, for exam= ple > > > > > > > > > > Are you talking about a 51% attack? This is possible in any decen= tralized cryptocurrency. > > > > > On Mon, May 24, 2021 at 11:49 AM Erik Aronesty erik@q32.com wrote= : > > > > > > > > > > > > > your burn investment is always "at stake", any redaction ca= n result in a loss-of-burn, because burns can be tied, precisely, to block-= heights > > > > > > > > I'm fuzzy on how proof of burn works. > > > > > > > > > > > > when you burn coins, you burn them to be used at a future parti= cular > > > > > > block height: so if i'm burning for block 553, i can only use t= hem to > > > > > > mine block 553. if i have a choice between two chains, one long= er > > > > > > and one shorter, i can only choose one... deterministically, fo= r that > > > > > > burn: the chain with the height 553. if we fix the "lead time" = for > > > > > > burned coins to be weeks or even months in advance, miners have= a very > > > > > > strong, long-term, investment in the stability of the chain. > > > > > > therefore there is no "nothing at stake" problem. it's > > > > > > deterministic, so miners have no choice. they can only choose t= he > > > > > > transactions that go into the block. they cannot choose which c= hain > > > > > > to mine, and it's time-locked, so rollbacks and instability alw= ays > > > > > > hurt miners the most. > > > > > > the "punishment" systems of PoS are "weird at best", certainly > > > > > > unproven. i can imagine scenarios where large stakeholders can > > > > > > collude to punish smaller stakeholders simply to drive them out= of > > > > > > business, for example. and then you have to put checks in place= to > > > > > > prevent that, and more checks for those prevention system... > > > > > > in PoB, there is no complexity. simpler systems like this are > > > > > > typically more secure. > > > > > > PoB also solves problems caused by "energy dependence", which c= ould > > > > > > lead to state monopolies on mining (like the new Bitcoin Mining > > > > > > Council). these consortiums, if state sanctioned, could become = a > > > > > > source of censorship, for example. Since PoB doesn't require yo= u to > > > > > > have a live, well-connected node, it's harder to censor & harde= r to > > > > > > trace. > > > > > > Eliminating this weakness seems to be in the best interests of > > > > > > existing stakeholders > > > > > > On Mon, May 24, 2021 at 4:44 PM Billy Tetrud billy.tetrud@gmail= .com wrote: > > > > > > > > > > > > > > proof of burn clearly solves this, since nothing is held on= line > > > > > > > > > > > > > > Well.. the coins to be burned need to be online when they're = burned. But yes, only a small fraction of the total coins need to be online= . > > > > > > > > > > > > > > > your burn investment is always "at stake", any redaction ca= n result in a loss-of-burn, because burns can be tied, precisely, to block-= heights > > > > > > > > > > > > > > So you're saying that if say someone tries to mine a block on= a shorter chain, that requires them to send a transaction burning their co= ins, and that transaction could also be spent on the longest chain, which m= eans their coins are burned even if the chain they tried to mine on doesn't= win? I'm fuzzy on how proof of burn works. > > > > > > > > > > > > > > > proof of burn can be more secure than proof-of-stake > > > > > > > > > > > > > > FYI, proof of stake can be done without the "nothing at stake= " problem. You can simply punish people who mint on shorter chains (by rewa= rding people who publish proofs of this happening on the main chain). In qu= orum-based PoS, you can punish people in the quorum that propose or sign mu= ltiple blocks for the same height. The "nothing at stake" problem is a solv= ed problem at this point for PoS. > > > > > > > On Mon, May 24, 2021 at 3:47 AM Erik Aronesty erik@q32.com wr= ote: > > > > > > > > > > > > > > > > I don't see a way to get around the conflicting requireme= nt that the keys for large amounts of coins should be kept offline but thos= e are exactly the coins we need online to make the scheme secure. > > > > > > > > > > > > > > > > proof of burn clearly solves this, since nothing is held on= line > > > > > > > > > > > > > > > > > how does proof of burn solve the "nothing at stake" probl= em in your view? > > > > > > > > > > > > > > > > definition of nothing at stake: in the event of a fork, whe= ther the > > > > > > > > fork is accidental or a malicious, the optimal strategy for= any miner > > > > > > > > is to mine on every chain, so that the miner gets their rew= ard no > > > > > > > > matter which fork wins. indeed in proof-of-stake, the proof= s are > > > > > > > > published on the very chains mines, so the incentive is mag= nified. > > > > > > > > in proof-of-burn, your burn investment is always "at stake"= , any > > > > > > > > redaction can result in a loss-of-burn, because burns can b= e tied, > > > > > > > > precisely, to block-heights > > > > > > > > as a result, miners no longer have an incentive to mine all= chains > > > > > > > > in this way proof of burn can be more secure than proof-of-= stake, and > > > > > > > > even more secure than proof of work > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > On Sun, May 23, 2021 at 3:52 AM Lloyd Fournier via bitcoin-= dev > > > > > > > > bitcoin-dev@lists.linuxfoundation.org wrote: > > > > > > > > > > > > > > > > > Hi Billy, > > > > > > > > > I was going to write a post which started by dismissing m= any of the weak arguments that are made against PoS made in this thread and= elsewhere. > > > > > > > > > Although I don't agree with all your points you have done= a decent job here so I'll focus on the second part: why I think Proof-of-S= take is inappropriate for a Bitcoin-like system. > > > > > > > > > Proof of stake is not fit for purpose for a global settle= ment layer in a pure digital asset (i.e. "digital gold") which is what Bitc= oin is trying to be. > > > > > > > > > PoS necessarily gives responsibilities to the holders of = coins that they do not want and cannot handle. > > > > > > > > > In Bitcoin, large unsophisticated coin holders can put th= eir coins in cold storage without a second thought given to the health of t= he underlying ledger. > > > > > > > > > As much as hardcore Bitcoiners try to convince them to ru= n their own node, most don't, and that's perfectly acceptable. > > > > > > > > > At no point do their personal decisions affect the underl= ying consensus -- it only affects their personal security assurance (not th= at of the system itself). > > > > > > > > > In PoS systems this clean separation of responsibilities = does not exist. > > > > > > > > > I think that the more rigorously studied PoS protocols wi= ll work fine within the security claims made in their papers. > > > > > > > > > People who believe that these protocols are destined for = catastrophic consensus failure are certainly in for a surprise. > > > > > > > > > But the devil is in the detail. > > > > > > > > > Let's look at what the implications of using the leading = proof of stake protocols would have on Bitcoin: > > > > > > > > > > > > > > > > > > ### Proof of SquareSpace (Cardano, Polkdadot) > > > > > > > > > > > > > > > > > > Cardano is a UTXO based PoS coin based on Ouroboros Praos= 3 with an inbuilt on-chain delegation system5. > > > > > > > > > In these protocols, coin holders who do not want to run t= heir node with their hot keys in it delegate it to a "Stake Pool". > > > > > > > > > I call the resulting system Proof-of-SquareSpace since mo= st will choose a pool by looking around for one with a nice website and off= ering the largest share of the block reward. > > > > > > > > > On the surface this might sound no different than someone= with an mining rig shopping around for a good mining pool but there are cr= ucial differences: > > > > > > > > > > > > > > > > > > 1. The person making the decision is forced into it just= because they own the currency -- someone with a mining rig has purchased i= t with the intent to make profit by participating in consensus. > > > > > > > > > > > > > > > > > > 2. When you join a mining pool your systems are very muc= h still online. You are just partaking in a pool to reduce your profit vari= ance. You still see every block that you help create and you never help cre= ate a block without seeing it first. > > > > > > > > > > > > > > > > > > 3. If by SquareSpace sybil attack you gain a dishonest m= ajority and start censoring transactions how are the users meant to redeleg= ate their stake to honest pools? > > > > > > > > > I guess they can just send a transaction delegating t= o another pool...oh wait I guess that might be censored too! This seems rea= lly really bad. > > > > > > > > > In Bitcoin, miners can just join a different pool at = a whim. There is nothing the attacker can do to stop them. A temporary dish= onest majority heals relatively well. > > > > > > > > > > > > > > > > > > > > > > > > > > > There is another severe disadvantage to this on-chain del= egation system: every UTXO must indicate which staking account this UTXO be= longs to so the appropriate share of block rewards can be transferred there= . > > > > > > > > > Being able to associate every UTXO to an account ruins on= e of the main privacy advantages of the UTXO model. > > > > > > > > > It also grows the size of the blockchain significantly. > > > > > > > > > > > > > > > > > > ### "Pure" proof of stake (Algorand) > > > > > > > > > > > > > > > > > > Algorand's4 approach is to only allow online stake to par= ticipate in the protocol. > > > > > > > > > Theoretically, This means that keys holding funds have to= be online in order for them to author blocks when they are chosen. > > > > > > > > > Of course in reality no one wants to keep their coin hold= ing keys online so in Alogorand you can authorize a set of "participation k= eys"1 that will be used to create blocks on your coin holding key's behalf. > > > > > > > > > Hopefully you've spotted the problem. > > > > > > > > > You can send your participation keys to any malicious par= ty with a nice website (see random example 2) offering you a good return. > > > > > > > > > Damn it's still Proof-of-SquareSpace! > > > > > > > > > The minor advantage is that at least the participation ke= ys expire after a certain amount of time so eventually the SquareSpace atta= cker will lose their hold on consensus. > > > > > > > > > Importantly there is also less junk on the blockchain bec= ause the participation keys are delegated off-chain and so are not making a= s much of a mess. > > > > > > > > > > > > > > > > > > ### Conclusion > > > > > > > > > > > > > > > > > > I don't see a way to get around the conflicting requireme= nt that the keys for large amounts of coins should be kept offline but thos= e are exactly the coins we need online to make the scheme secure. > > > > > > > > > If we allow delegation then we open up a new social attac= k surface and it degenerates to Proof-of-SquareSpace. > > > > > > > > > For a "digital gold" like system like Bitcoin we optimize= for simplicity and desperately want to avoid extraneous responsibilities f= or the holder of the coin. > > > > > > > > > After all, gold is an inert element on the periodic table= that doesn't confer responsibilities on the holder to maintain the quality= of all the other bars of gold out there. > > > > > > > > > Bitcoin feels like this too and in many ways is more iner= t and beautifully boring than gold. > > > > > > > > > For Bitcoin to succeed I think we need to keep it that wa= y and Proof-of-Stake makes everything a bit too exciting. > > > > > > > > > I suppose in the end the market will decide what is real = digital gold and whether these bad technical trade offs are worth being abl= e to say it uses less electricity. It goes without saying that making bad t= echnical decisions to appease the current political climate is an anathema = to Bitcoin. > > > > > > > > > Would be interested to know if you or others think differ= ently on these points. > > > > > > > > > Cheers, > > > > > > > > > LL > > > > > > > > > On Fri, 21 May 2021 at 19:21, Billy Tetrud via bitcoin-de= v bitcoin-dev@lists.linuxfoundation.org wrote: > > > > > > > > > > > > > > > > > > > I think there is a lot of misinformation and bias again= st Proof of Stake. Yes there have been lots of shady coins that use insecur= e PoS mechanisms. Yes there have been massive issues with distribution of P= oS coins (of course there have also been massive issues with PoW coins as w= ell). However, I want to remind everyone that there is a difference between= "proved to be impossible" and "have not achieved recognized success yet". = Most of the arguments levied against PoS are out of date or rely on unprove= n assumptions or extrapolation from the analysis of a particular PoS system= . I certainly don't think we should experiment with bitcoin by switching to= PoS, but from my research, it seems very likely that there is a proof of s= take consensus protocol we could build that has substantially higher securi= ty (cost / capital required to execute an attack) while at the same time co= sting far less resources (which do translate to fees on the network) withou= t compromising any of the critical security properties bitcoin relies on. I= think the critical piece of this is the disagreements around hardcoded che= ckpoints, which is a critical piece solving attacks that could be levied on= a PoS chain, and how that does (or doesn't) affect the security model. > > > > > > > > > > @Eric Your proof of stake fallacy seems to be saying th= at PoS is worse when a 51% attack happens. While I agree, I think that line= of thinking omits important facts: > > > > > > > > > > > > > > > > > > > > - The capital required to 51% attack a PoS chain can = be made substantially greater than on a PoS chain. > > > > > > > > > > - The capital the attacker stands to lose can be subs= tantially greater as well if the attack is successful. > > > > > > > > > > - The effectiveness of paying miners to raise the hon= est fraction of miners above 50% may be quite bad. > > > > > > > > > > - Allowing a 51% attack is already unacceptable. It s= hould be considered whether what happens in the case of a 51% may not be si= gnificantly different. The currency would likely be critically damaged in a= 51% attack regardless of consensus mechanism. > > > > > > > > > > > > > > > > > > > > > Proof-of-stake tends towards oligopolistic control > > > > > > > > > > > > > > > > > > > > People repeat this often, but the facts support this. T= here is no centralization pressure in any proof of stake mechanism that I'm= aware of. IE if you have 10 times as much coin that you use to mint blocks= , you should expect to earn 10x as much minting revenue - not more than 10x= . By contrast, proof of work does in fact have clear centralization pressur= e - this is not disputed. Our goal in relation to that is to ensure that th= e centralization pressure remains insignifiant. Proof of work also clearly = has a lot more barriers to entry than any proof of stake system does. Both = of these mean the tendency towards oligopolistic control is worse for PoW. > > > > > > > > > > > > > > > > > > > > > Energy usage, in-and-of-itself, is nothing to be asha= med of!! > > > > > > > > > > > > > > > > > > > > I certainly agree. Bitcoin's energy usage at the moment= is I think quite warranted. However, the question is: can we do substantia= lly better. I think if we can, we probably should... eventually. > > > > > > > > > > > > > > > > > > > > > Proof of Stake is only resilient to =E2=85=93 of the = network demonstrating a Byzantine Fault, whilst Proof of Work is resilient = up to the =C2=BD threshold > > > > > > > > > > > > > > > > > > > > I see no mention of this in the pos.pdf you linked to. = I'm not aware of any proof that all PoS systems have a failure threshold of= 1/3. I know that staking systems like Casper do in fact have that 1/3 requ= irement. However there are PoS designs that should exceed that up to nearly= 50% as far as I'm aware. Proof of work is not in fact resilient up to the = 1/2 threshold in the way you would think. IE, if 100% of miners are current= ly honest and have a collective 100 exahashes/s hashpower, an attacker does= not need to obtain 100 exahashes/s, but actually only needs to accumulate = 50 exahashes/s. This is because as the attacker accumulates hashpower, it d= rives honest miners out of the market as the difficulty increases to beyond= what is economically sustainable. Also, its been shown that the best proof= of work can do is require an attacker to obtain 33% of the hashpower becau= se of the selfish mining attack discussed in depth in this paper: https://a= rxiv.org/abs/1311.0243. Together, both of these things reduce PoW's securit= y by a factor of about 83% (1 - 50%*33%). > > > > > > > > > > > > > > > > > > > > > Proof of Stake requires other trade-offs which are in= compatible with Bitcoin's objective (to be a trustless digital cash) = =E2=80=94 specifically the famous "security vs. liveness" guarantee > > > > > > > > > > > > > > > > > > > > Do you have a good source that talks about why you thin= k proof of stake cannot be used for a trustless digital cash? > > > > > > > > > > > > > > > > > > > > > You cannot gain tokens without someone choosing to gi= ve up those coins - a form of permission. > > > > > > > > > > > > > > > > > > > > This is not a practical constraint. Just like in mining= , some nodes may reject you, but there will likely be more that will accept= you, some sellers may reject you, but most would accept your money as paym= ent for bitcoins. I don't think requiring the "permission" of one of millio= ns of people in the market can be reasonably considered a "permissioned cur= rency". > > > > > > > > > > > > > > > > > > > > > 2. Proof of stake must have a trusted means of times= tamping to regulate overproduction of blocks > > > > > > > > > > > > > > > > > > > > Both PoW and PoS could mine/mint blocks twice as fast i= f everyone agreed to double their clock speeds. Both systems rely on an hon= est majority sticking to standard time. > > > > > > > > > > On Wed, May 19, 2021 at 5:32 AM Michael Dubrovsky via b= itcoin-dev bitcoin-dev@lists.linuxfoundation.org wrote: > > > > > > > > > > > > > > > > > > > > > Ah sorry, I didn't realize this was, in fact, a diffe= rent thread! :) > > > > > > > > > > > On Wed, May 19, 2021 at 10:07 AM Michael Dubrovsky mi= ke@powx.org wrote: > > > > > > > > > > > > > > > > > > > > > > > Folks, I suggest we keep the discussion to PoW, oPo= W, and the BIP itself. PoS, VDFs, and so on are interesting but I guess the= re are other threads going on these topics already where they would be rele= vant. > > > > > > > > > > > > Also, it's important to distinguish between oPoW an= d these other "alternatives" to Hashcash. oPoW is a true Proof of Work that= doesn't alter the core game theory or security assumptions of Hashcash and= actually contains SHA (can be SHA3, SHA256, etc hash is interchangeable). > > > > > > > > > > > > Cheers, > > > > > > > > > > > > Mike > > > > > > > > > > > > On Tue, May 18, 2021 at 4:55 PM Erik Aronesty via b= itcoin-dev bitcoin-dev@lists.linuxfoundation.org wrote: > > > > > > > > > > > > > > > > > > > > > > > > > 1. i never suggested vdf's to replace pow. > > > > > > > > > > > > > > > > > > > > > > > > > > 2. my suggestion was specifically in the context= of a working > > > > > > > > > > > > > proof-of-burn protocol > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > - vdfs used only for timing (not block height) > > > > > > > > > > > > > - blind-burned coins of a specific age used to = replace proof of work > > > > > > > > > > > > > - the required "work" per block would simply be= a competition to > > > > > > > > > > > > > acquire rewards, and so miners would have to = burn coins, well in > > > > > > > > > > > > > advance, and hope that their burned coins got= rewarded in some far > > > > > > > > > > > > > future > > > > > > > > > > > > > > > > > > > > > > > > > > - the point of burned coins is to mimic, in eve= ry meaningful way, the > > > > > > > > > > > > > value gained from proof of work... without so= me of the security > > > > > > > > > > > > > drawbacks > > > > > > > > > > > > > > > > > > > > > > > > > > - the miner risks losing all of his burned coin= s (like all miners risk > > > > > > > > > > > > > losing their work in each block) > > > > > > > > > > > > > > > > > > > > > > > > > > - new burns can't be used > > > > > > > > > > > > > - old burns age out (like ASICs do) > > > > > > > > > > > > > - other requirements on burns might be needed t= o properly mirror the > > > > > > > > > > > > > properties of PoW and the incentives Bitcoin = uses to mine honestly. > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > 3. i do believe it is possible that a "burned co= in + vdf system" > > > > > > > > > > > > > might be more secure in the long run, and tha= t if the entire space > > > > > > > > > > > > > agreed that such an endeavor was worthwhile, = a test net could be spun > > > > > > > > > > > > > up, and a hard-fork could be initiated. > > > > > > > > > > > > > > > > > > > > > > > > > > 4. i would never suggest such a thing unless i b= elieved it was > > > > > > > > > > > > > possible that consensus was possible. so no, = this is not an "alt > > > > > > > > > > > > > coin" > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > On Tue, May 18, 2021 at 10:02 AM Zac Greenwood za= chgrw@gmail.com wrote: > > > > > > > > > > > > > > > > > > > > > > > > > > > Hi ZmnSCPxj, > > > > > > > > > > > > > > Please note that I am not suggesting VDFs as a = means to save energy, but solely as a means to make the time between blocks= more constant. > > > > > > > > > > > > > > Zac > > > > > > > > > > > > > > On Tue, 18 May 2021 at 12:42, ZmnSCPxj ZmnSCPxj= @protonmail.com wrote: > > > > > > > > > > > > > > > > > > > > > > > > > > > > > Good morning Zac, > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > VDFs might enable more constant block times= , for instance by having a two-step PoW: > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > 1. Use a VDF that takes say 9 minutes to r= esolve (VDF being subject to difficulty adjustments similar to the as-is). = As per the property of VDFs, miners are able show proof of work. > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > 2. Use current PoW mechanism with lower di= fficulty so finding a block takes 1 minute on average, again subject to as-= is difficulty adjustments. > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > As a result, variation in block times will = be greatly reduced. > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > As I understand it, another weakness of VDFs = is that they are not inherently progress-free (their sequential nature prev= ents that; they are inherently progress-requiring). > > > > > > > > > > > > > > > Thus, a miner which focuses on improving the = amount of energy that it can pump into the VDF circuitry (by overclocking a= nd freezing the circuitry), could potentially get into a winner-takes-all s= ituation, possibly leading to even worse competition and even more energy c= onsumption. > > > > > > > > > > > > > > > After all, if you can start mining 0.1s faste= r than the competition, that is a 0.1s advantage where only you can mine in= the entire world. > > > > > > > > > > > > > > > Regards, > > > > > > > > > > > > > > > ZmnSCPxj > > > > > > > > > > > > > > > > > > > > > > > > > > bitcoin-dev mailing list > > > > > > > > > > > > > bitcoin-dev@lists.linuxfoundation.org > > > > > > > > > > > > > https://lists.linuxfoundation.org/mailman/listinf= o/bitcoin-dev > > > > > > > > > > > > > > > > > > > > > > > > -- > > > > > > > > > > > > Michael Dubrovsky > > > > > > > > > > > > Founder; PoWx > > > > > > > > > > > > www.PoWx.org > > > > > > > > > > > > > > > > > > > > > > -- > > > > > > > > > > > Michael Dubrovsky > > > > > > > > > > > Founder; PoWx > > > > > > > > > > > www.PoWx.org > > > > > > > > > > > > > > > > > > > > > > bitcoin-dev mailing list > > > > > > > > > > > bitcoin-dev@lists.linuxfoundation.org > > > > > > > > > > > https://lists.linuxfoundation.org/mailman/listinfo/bi= tcoin-dev > > > > > > > > > > > > > > > > > > > > bitcoin-dev mailing list > > > > > > > > > > bitcoin-dev@lists.linuxfoundation.org > > > > > > > > > > https://lists.linuxfoundation.org/mailman/listinfo/bitc= oin-dev > > > > > > > > > > > > > > > > > > bitcoin-dev mailing list > > > > > > > > > bitcoin-dev@lists.linuxfoundation.org > > > > > > > > > https://lists.linuxfoundation.org/mailman/listinfo/bitcoi= n-dev