Hacker News new | past | comments | ask | show | jobs | submit login
Cooperation among an anonymous group protected Bitcoin during failures (arxiv.org)
139 points by Anon84 on June 10, 2022 | hide | past | favorite | 53 comments



A 51% attack in those days would have made no sense. These were the days before exchanges or even an exchange rate. To trade you had to post on a forum or IRC and wait around for an answer. If the network was attacked, it would have been announced on those same forums. To pull off an attack you'd have to set up a room full of 100 computers CPU mining, hope that nobody noticed the hashrate suddenly doubling, and maybe get a few pennies worth of BTC for your efforts.

There was a much easier way to get bitcoin back then: Go to the bitcoin faucet and type in your address.


The internet button that gives you $150,000 of June 2022 money for free once a day every day.


Gavin funded a lot of these personally in the early days.


Oh, I saw $500/transaction.

Thought that was the limit


It was five bitcoins per day. I actually looked at it back then, and decided it wasn't worth the effort.


This paper acts like it's surprising that miners would cooperate and play by the rules:

> Although bitcoin was designed to rely on a decentralized, trustless network of anonymous agents, its early success rested instead on cooperation among a small group of altruistic founders.

But the Bitcoin whiteppaer says very clearly that it's designed to give miners and incentive to cooperate:

> The incentive may help encourage nodes to stay honest. If a greedy attacker is able to assemble more CPU power than all the honest nodes, he would have to choose between using it to defraud people by stealing back his payments, or using it to generate new coins. He ought to find it more profitable to play by the rules, such rules that favour him with more new coins than everyone else combined, than to undermine the system and the validity of his own wealth.

https://bitcoin.org/bitcoin.pdf


It also says, as you left out, that those early miners had the power to attack the network.


This is exactly how it was predicted to work in the whitepaper. I've always been skeptical about it holding longterm, but this paper seems to be confirming that's held for now.

> If a greedy attacker is able to assemble more CPU power than all the honest nodes, he would have to choose between using it to defraud people by stealing back his payments, or using it to generate new coins. He ought to find it more profitable to play by the rules, such rules that favour him with more new coins than everyone else combined, than to undermine the system and the validity of his own wealth


What if he does it without being detected?

There are probably a bunch of wallets from back in the day whose owner lost their key. Take some coin out of those. Then no one knows. Sell it. Move on.


A 51% attack cannot take other people's coins. It can only double-spend coins owned by the attacker and there's no way to do a 51% attack without people noticing.


A double spend would sort of be taking other people's coins though. Once you revert the first spend, the recipient loses their coins.


yes, but you cant take random coins from random walets, you can only take what you transacted recently.


It's interesting that the centipede game experiment uses a high rate of appreciation - doubling the payoff every round. On the one hand, this offers strong incentive to defect if you strongly believe the next player will defect - you double your reward (this is the strategy game theory predicts). But if you are less confident, it doesn't take much cooperation before you profit significantly. If only the next two players also cooperate, you will have done better than if you had defected.

It seems like rapid expected appreciation, combined with only a small amount of confidence that the other players will not defect (e.g. because you know they share your ideological preference for Bitcoin to succeed) is enough to stave off defection.

I'd be curious to see what the human-player dynamics look like if the expected future price does not increase so sharply, or is even flat or expected to decline.


The last halvening, the price increased less than what a lot of people expected (didn't reach $100k or $250k as per the Stock-To-Flow model). And the peak was much smoother.


Did anyone really believe that the Stock-to-Flow model would be a correct predictor of price? That seems awfully naive, it's just some pseudoscientific nonsense.


If you used the raw model, without any smoothing, it still predicted the $50k price correctly.

I think the smoothing gave too much weight to S2F and not enough to plain time.

But I am really curious what will happen after 2024. That is when S2F really diverges from a plain time power-law.

https://danuker.go.ro/bitcoin-price-predictions.html


This is an interesting paper in a lot of ways. I think the least interesting part is what most media is reporting, that shockingly, Bitcoin didn't suddenly become a massively decentralized network in its first two years of operation.

The most intriguing part to me is this: "game in which anonymous players can choose to exploit, and thereby undermine, an appreciating good...we show that, even when individual payoffs are unchanged, cooperation is more frequent when the game is played by an anonymous group."


> we reveal that between launch (January 3rd, 2009), and when the price reached $1 (February 9th, 2011), most bitcoin was mined by only sixty-four agents.

So who can work out which addresses were mine?


There were pools even then though


My addresses were never part of a mining pool.


Interesting paper, but whether the early cooperation was necessarily altruistic in nature remains a mystery to me. The highly centralized structure which transpired to make Bitcoin popular and regarded as secure raises suspicion about its legitimacy and true intentions, in my mind. Over a decade later, digital currencies have become much more palatable and even attractive to people, that there's even talk of government-backed DC's, and this resulting Overton window shift has significant implications for individual freedom and society.


No central bank digital currency is planning to use a blockchain. A blockchain would add fundamental insurmountable drawbacks. Digital money using databases has been around since before the internet. Central bank digital currency is just a minor addition to our current digital financial infrastructure.


A central bank and blockchain technology are antithetical concepts, and if a central bank is ever offering you something blockchain it's probably a scam by means of sham decentralization.


>>A central bank and blockchain technology are antithetical concepts,

I feel I need to screenshot this. It demonstrates more succinctly than I could write an essay on, how much we are muddying up technology, and purpose / politics / ideals when we start discussing this area (Crypto currency etc)

How is a human institution in your mind "antithetical" to a technology? Is a government "antithetical" to 3rd normal form or airline antithetical to a b-tree? You are inserting a large number of undisclosed assumptions to even begin to make such a statement (whether aware of it or not).

>>if a central bank is ever offering you something blockchain it's probably a scam by means of sham decentralization.

I don't disagree with this... I just think it neatly extends fully to "if anybody is ever offering you something block chain" :-)


I feel like you're intentionally playing semantic games, i.e. trolling, so let me tell you straight that I don't appreciate that.

The fact a central bank is a "human institution" is irrelevant. The fact blockchain is a "technology" is also in itself irrelevant. The point is that each concept necessitates characteristics that are irreconcilable. Most importantly a central bank can not function without central authority. The defining property of the blockchain is the decentralization of authority. Authority can not be simultaneously centralized and decentralized, and so the blockchain and the central bank are antithetical concepts.


>> The defining property of the blockchain is the decentralization of authority.

So "unaware assumptions" it is then.

I'm sorry if you perceive that as snarky (and not everyone that disagrees is a troll btw), but to me, decentralized authority is what some people perceive an important (if not zealous) use case. It is not the defining property of blockchain. There are many companies that are building distributed ledger with centralized authority. You may disagree with that use case. I feel it's lack of self awareness to think a valid and clearly implemented use case is somehow antithetical to a simple neutral well defined technology.


>There are many companies that are building distributed ledger with centralized authority.

The fundamental innovation of the bitcoin protocol is how it made the blockchain, a traditionally centralized data structure invented in the early 1990s, a distributed append-only database using PoW consensus. It's incredibly inefficient, its unimaginable inefficiency is there to buy trust in a trustless peer-to-peer network. You're basically forcing every writer to cut down a pound of their own computational flesh to write to the database. How can this be necessary or useful in a trusted setting? why can you not rely on trust or traditional register-and-authenticate to write to the database?

You're either using PoW consensus or you're not. If you're, you're not a central authority (unless the blockchain is somehow not an authoritative state and you're simply using it as a fancy bulletin board), if you're not, then whatever you're doing is different from what everbody else have been calling 'blockchain' since 13 years and you're just inviting confusion by calling it that.


I think there's a difference between "blockchain + PoW" and "blockchain". Blockchain taken literally is just verification of history via hashing. It's a form of provenance that is hard to fake. Even if you are a central authority as to what transactions are being made, you can't rewrite history beyond any point that you've made publicly visible, without losing the trust of people who rely on you as a central authority, as to do so you would need to break the hashing algorithm.


But I'm still rejecting the premise. There's TONS of small startups and large companies that are building blockchain for verification purposes and centralized or known limited authorities. Many of them are showing pragmatic promise in supply chain verification etc. Certainly they're calling it blockchain. What they're not calling it is Crypto currency. Are we saying they're synonymous and there's no other use case?

I'm an outsider laughing at the whole Blockchain brouhaha, have zero horse in any race. I'm just amused endlessly how any given individual strongly believes their perspective is the only valid one and all others are simply "wrong", and heretics to the one true religion I mean bitcoin I mean blockchain I mean Crypto currency (Note how people are switching which one we are talking about in the thread. Due to various assumptions of equivalence ). Everything in this thread has firmed up my perspective.


>Are we saying they're synonymous and there's no other use case?

No, we're saying that modern usage of "blockchain", dating back 13 years, is specifically about distributed trustless append-only ledgers, which is only useful for distributed authoritative state. Crypto currencies just happen to be a really good application for distributed authoritative state. You can do trusty applications with trustless ledgers, but that would be pointless and needlessly wasteful. You can do a centralized 1990s blockchain, but calling it 'blockchain', though technically true, just smells of marketing hype.

>amused endlessly how any given individual strongly believes their perspective is the only valid one and all others are simply "wrong", and heretics to the one true religion

Snark.

>Note how people are switching which one we are talking about in the thread. Due to various assumptions of equivalence

People on HN probably don't need to be reminded by the entry level 101 stuff about the difference between bitcoin and the blockchain, so your point is just trivial hair splitting. Bitcoin is a special case of crypto currencies, which is enabled by PoW driven blockchains, which without PoW are just boring old data structures. People switch between them because they are a natural fit that exist for and because of each other, and anyone of them alone either don't make any sense or make little sense.

>Everything in this thread has firmed up my perspective.

I don't see why you're so proud that you have learned nothing, but anything that makes you happy I guess.


Fair enough to the last point; but I did enter this thread open to be proven wrong, and then keep seeing sentences in the form of "which is only useful for distributed authoritative state", which is specifically my point. I have made reference several times to successful businesses, from startup to multinational, which are implementing blockchain without distributed authoritative state. Rather than specific discussion on validity of those use cases, all I am repeatedly seeing is casual dismissal without discussion. It is that repeated response that firms up my perspective.

I also do not believe blockchain, bitcoin, and cryptocurrency are actually synonyms and easily interchangeable, in knowledgeable or not knowledgeable audiences. I do think they're deemed interchangeable for some subset of people, per my initial point, who believe them to be trivially synonymous; and who cannot themselves be educated or be opened minded enough or escape internal context to see how anybody else can possibly have a different perspective. Here, too, we have failed to reach each other, and I suppose to neither of our surprise either.


> The defining property of the blockchain is the decentralization of authority

No it’s not. A blockchain can be used in a way that only specific central authorities can add to it. https://en.wikipedia.org/wiki/Blockchain#Private_blockchains


I was kind of expecting you to link to an article about Git repository permissions.


You can use a rock as a spoon but that just makes you retarded.


Are you familiar with git repositories?


>How is a human institution in your mind "antithetical" to a technology?

It's not human institution vs technology, it's the very structure of central banking vs decentral banking. It's the concepts themselves that are antithetical.

-Edit I couldn't remember the word, but I believe the term that applies here is 'diametrically opposed'


But that's the unspoken assumption I'm talking about - that blockchain and decentralized banking are one and the same / that this is the only design use case for the technology. There are many people actively working in block chain space who do not believe these are identical concepts. They may yet win out long term.


I'm not claiming that blockchain is a one use concept. Yes it can be used for other things then banking. I personally feel that NFTs are a great example of that, and that they hold many possibilities for the future.

But the main benefit that blockchain provides is a way to decentralize a network, it's quite inefficient from a performance standpoint. So to use it for anything other than that is a waste (at least currently). Centralized currencies are better than decentralized ones performance wise, and I don't predict that will change anytime in the near future.

To quote Eldenwrong's perfect comment here "You can use a rock as a spoon but that just makes you retarded"

Hense blockchain currencies (the ones that aren't scams ofc) are decentralized by their nature, since that's the only draw to using that method.

Because of the Pareto distribution this should fail at some point, but with the innumerable amount of new currencies coming into existence, that shouldn't be a problem.


Blockchain has a purpose: consensus without trust.

Central banks are institutions that are trusted and cooperate with trusted institutions.

In general as long as you have a mechanism to establish trust-based consensus by relying on people or a government there is really no advantage by using a blockhain and a huge number of disadvantages.


I agree.frequently it seems I'm in twilight zone minority :-)


Both are, always have been


> No central bank digital currency is planning to use a blockchain

That's hilariously incorrect. You should revisit this comment later this year.


> No central bank digital currency is planning to use a blockchain.

   f a l s e
What is this then? [0][1] Where we have those same blockchain technologies from the likes of Stellar, XPRL, Algorand, etc that are compliant with the ISO 20020 standard?

As I said before on what is really going on, both the crypto maximalists and the crypto skeptics are going to be very disappointed in achieving their unobtainable and extreme goals.

[0] https://www.prnewswire.com/news-releases/ukraine-electronic-...

[1] https://www.businesswire.com/news/home/20210922005878/en/Roy...


That is not a central bank...

From Wikipedia A central bank, reserve bank, or monetary authority is an institution that manages the currency and monetary policy of a state or formal monetary union, and oversees their commercial banking system. In contrast to a commercial bank, a central bank possesses a monopoly on increasing the monetary base.


> That is not a central bank...

Assuming you read the entire press releases, so the Royal Monetary Authority of Bhutan as I previously mentioned [0], and the National Bank of Ukraine' NBU' mentioned in the press release and is part of the pilot program [2] are not central banks?

What does this say here? [0]

   The Royal Monetary Authority of Bhutan is the central bank of Bhutan...
And here [1]

   The National Bank of Ukraine is the central bank of Ukraine...
It also says here [2]:

   The National Bank of Ukraine (NBU) selected the distributed ledger technology, in particular the private version of the Stellar protocol, to study opportunities of issuing e-hryvnia and to carry out the Pilot Project.
The Reserve Bank of Australia have also implemented an Ethereum-based CBDC proof-of-concept which was successful. [3]

So the claim 'No central bank digital currency is planning to use a blockchain.' is still false. I think there is a strange sense of denial with the commenter who created that unfounded and baseless absolute claim.

[0] https://en.wikipedia.org/wiki/Royal_Monetary_Authority_of_Bh...

[1] https://en.wikipedia.org/wiki/National_Bank_of_Ukraine

[2] https://bank.gov.ua/admin_uploads/article/Analytical+Report+...

[3] https://www.rba.gov.au/media-releases/2021/mr-21-30.html


Every reputable organization that rolled out a pilot project did nothing afterwards. There is no plan to actually use a blockchain for anything beyond speculation, money laundering, criminality, and scams.


> Every reputable organization that rolled out a pilot project did nothing afterwards.

I see. So after you realised your absolute claim was refuted, you set out moving the goalposts to create another absolute claim, yet that is still wrong again.

The Central Bank of Nigeria (CBN) is using a blockchain for their e-Naira CBDC and that has already launched many months ago. [0] So even when you tried moving the goalposts, your claim has still been refuted once again!

> There is no plan to actually use a blockchain for anything beyond speculation, money laundering, criminality, and scams.

So Moneygram is not a reputable organisation using a blockchain? [1] Perhaps Stripe isn't either [2] who's first partner with crypto payouts is Twitter. Maybe that is why they both waited for regulatory clarity before using these technologies?

This is the problem of having very absolutist statements and creating sweeping claims as you have done. It forces you to move the goal posts on your original claims, create fallacies and do whatever to dodge counter questions.

[0] https://www.prnewswire.com/news-releases/bitt-develops-afric...

[1] https://www.prnewswire.com/news-releases/moneygram-launches-...

[2] https://stripe.com/blog/expanding-global-payouts-with-crypto


So you and I can run a full node for the Central Bank of Nigeria's blockchain? No, it is obviously a private database because not even Nigeria has judgment that poor.


> So you and I can run a full node for the Central Bank of Nigeria's blockchain?

That is not my point and no one is arguing that the use case for blockchains in CBDCs was 'decentralized' or 'decentralization' in the first place and none of the evidences I've shown ever said that. Why do you think that in my first reply to your false claim on blockchains not being used in CBDCs, one of the sources explicitly says that it was '...Using Private Blockchain' [0]?

My whole point was towards your original false claims such as:

> > No central bank digital currency is planning to use a blockchain.

> > Every reputable organization that rolled out a pilot project did nothing afterwards.

So interestingly you have proved my last point after you yourself realised that your claims were refuted with no counter refutation and zero counter evidence towards my replies or even answering any of my other questions which suggests that you are simply trying to escape and avoid the discussion.

What is the point of screaming unfounded and absolute claims and then having to end up running away or dodging other question(s) after someone refutes them in front of you?

[0] https://news.ycombinator.com/item?id=31700999


It's not a huge mystery. "altruism" is author opinion not science. Double spending is a short term benefit, but risks destroying the value of your holdings by discrediting the whole system. Double spending only works in truly arms length transactions, where the victim can't retaliate AND trusts you enough to accept the transaction in the first place.


Also the gains for such action must be larger than resources spend. You are over spending the rest of the network in work, thus the transactions must be larger than possible gains from doing the same work in "main" chain.


> The highly centralized structure which transpired to make Bitcoin popular

What centralized structure?


Read the paper this thread is about.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: