Hacker News Reader

> I'm currently reading Digital Gold by Nathaniel. As I progress, its clearer that bitcoin might or might not be a great economic tool but its an awesome case study in game theory.

Yes, and you can abstract that observation from blockchains to cryptography in general, which is the underlying foundation of blockchain technology’s game theoretic fault tolerance. This is particularly clear once you move closer to cryptography’s theoretical and mathematical foundations and away from specific applications (such as encryption or authentication).

Many cryptography textbooks begin with an elementary definition and discussion of the field using notions such as confidentiality, integrity, authentication, non-repudiation, etc. These are all correct, but I prefer Oded Goldreich’s definition, from his Foundations of Cryptography:

Cryptography is concerned with any problem in which one wishes to limit the effects of dishonest users.

In an abstract sense, cryptography has a direct mapping from game theory, and can be considered the use of computational complexity to resolve scenarios in adversarial relationships.

If you’re enjoying looking at blockchains through this lens, you might want to read a few papers on cryptogenography, or zero knowledge proofs (the latter has far more literature). There is also a lot of fun research that uses game theory to model fault tolerance and economics in blockchains.

I haven't read the book, but this 2013 paper from Princeton economists Kroll, Davey, and Felten [1] describes the game theory that keeps strong players from destroying the network (aka a 51% or "goldfinger" attack):

"It is often asserted (for example, in the Bitcoin white paper [22]) that a cartel can double-spend Bitcoins. In a strict sense, this is true: a cartel can spend a Bitcoin by paying it to a player Alice, receiving goods or services, and then shifting the consensus choice of history to a branch where that coin is instead paid to a different player Bob...

Bitcoins have value because people are willing to trade them for goods and services. If players were unwilling to accept Bitcoins for trade or unwilling to spend Bitcoins for fear of having their payments nullified, the value of Bitcoins would diminish significantly as players lost confidence in the system...

Worse, because players are encouraged to generate a new identity for each transaction and because identities are not linked to any side information, players cannot easily determine whether a proffered payment is coming from the double-spending cartel or an honest user.

A rational player should refuse to accept any payments when there is a significant threat of double-spending. As a cartel must outmine the entire Bitcoin network and thus outspend the entire Bitcoin network for as long as it would remain a cartel, we believe it is very unlikely that a cartel could double-spend enough to recover the cost of the attack."

The question at this stage, with the price of Bitcoin so high, is if this rationale still holds. What if a cartel were able to double spend one (or multiple) of the big 40 "whale" Bitcoin accounts before destroying the entire network? The one-time billion dollar theft could offset the cost of putting together the computing power to pull off a 51% attach, especially if the cartel assembled that power maliciously through a botnet.

[1] http://www.econinfosec.org/archive/weis2013/papers/KrollDave...

> What if a cartel were able to double spend one (or multiple) of the big 40 "whale" Bitcoin accounts before destroying the entire network?

I don't think 51% attacks work that way. You can't spend at all without the private keys of the whales because the network nodes won't trust the chain. If they have enough power, they can double spend their own coins by rewinding the history till the point they are spent. This is why around 6 blocks time is considered as a must to prevent double spends. In fact as the price increases the lost opportunity cost also increases (they could have just mined and gotten the reward instead of attacking). How ever, there are still players out of the net currently who don't care about the rewards and have enough hash power (governments). It's just that at this point of time they are ambivalent to the little monster that is growing up before their eyes (like a cute little dinosaur).

> You can't spend at all without the private keys of the whales because the network nodes won't trust the chain.

Here is a simple library to recover the private key of ECDSA and DSA signatures sharing the same nonce k and therefore having identical signature parameter r

[1] https://github.com/tintinweb/ecdsa-private-key-recovery

That only gets you access to the private keys of specific accounts that had previously made transactions in a specific insecure way in old uncommon and no-longer-used clients. That's not at all a generic attack.

You cannot build a botnet large enough to out-hash the army of dedicated ASICs that's responsible for todays' Bitcoin hash rate. Not even a botnet that merges all personal computers in the world would be able to do that - ASICs are just that much faster.

But - some miners and mining pool operators are rumored to be among the bigger Bitcoin whales themselves. They already own (or at least control) a large part of the ASIC hashing power. None of them is even close to 51% today, but then you also don't really need 51% to pull off a double-spend attack, 30-40% should do it, as long as you pull that hashing power from the legit network at the same time. A cartel made up of the top miners and mining pool owners should be able to comfortably reach this critical mass of hash power. So effectively it comes down to whether they assume that it's better for them financially to continue supporting the networks' health or to pull a one-time heist, destroying the networks' credibility in the process. For the moment, the former seems to be still true. But nobody knows for sure whether that may change some day...

Especially as financial motivation is not the only reason for pulling that off. Currently majority of the miners are based in China. What happens if state orders them to approve a transaction of their choosing?

EDIT: I am not saying this is a probable threat, but there are some risks associated with Bitcoin.

Theoretically (I'm not expecting this to be a remotely practical attack), could you hijack ASICs in a similar fashion to reach the required hashrate?

The miners are incentivized not to undertake the attack you envisage because the outcome of such an attack is a fairly routine modification of the consensus protocol to an alternate Proof-of-work algorithm, which instantly renders all of the attacking miners hardware obsolete. They quite rightly want to avoid any strategy that might lead to that outcome.

Something similar happened in Ethereum where miners manipulated an ICO. During Status ICO sale, a particular pool mined only those blocks which had their transactions while rejecting others:

https://steemit.com/ethereum/@dhumphrey/update-f2pool-manipu...

These kinds of things look theoretical but might be happening in real life without any one knowing about it.

woah... that's a huge deal

> Bitcoins have value because people are willing to trade them for goods and services. If players were unwilling to accept Bitcoins for trade or unwilling to spend Bitcoins for fear of having their payments nullified, the value of Bitcoins would diminish significantly as players lost confidence in the system...

My impression is that right now, and for the last couple of years, this is untrue, with speculative demand vastly exceeding demand from people actually wanting to spend bitcoins on goods and services.

Currently buying Bitcoin is a very similar risk profile to buying stock of Facebook, where the value is highly correlated to the network effect, so even as a store of value Bitcoin doesn't work very well because of volatility, we will have to wait what current or future cryptocurrency develops a similar behavior to fiat currencies.

Who do you think you're kidding? It works exceptionally well as a store of value, regardless of volatility. It is approaching an increase of value of 100x in three years.

It worked wonderfully as an speculative investment and that correlates to working as a store of value there's a subtle diference.

Only if you don't like recognising a store of value when you see one.

I think bitcoin price will stabalize once most of the bitcoin are mined. Right now its a race to simply collect as much satoshi as possible if you believe in a future where cryto becomes the dominant form of exchange. 1 satoshi is still far less valuable than a single USD. 1 satoshi is 0.0001 USD right now, still a ways to go before it becomes equivalent.

What does equivalence have to do with it, though?

1 USD is worth .87 Euro or 1.3 Euro (at any given time), where 1 feels like a magic number; but 1 USD is worth (right now) 6.6 Yuan or 113 Yen, where it's more obvious 1 is not a magic number.

So I'm confused by the belief that .001 or .00001 is closer or further than where something 'should' be. How do we reach the conclusion that the value should be this thing or not this thing?

Simply by collectively believing it to be so.

This does not apply to bitcoin, however many alt coins have hash rates low enough that 51% attack is possible at any time of the day. Nobody cares as long as the price keep going up and the true degree of security is determined by the ~~casinos~~ exchanges they play at.

The problem is everything hinges on "rationality" of a player. And there might be a time when the rational decisions of a game theory goes out of the window and things don't work any more.

One big problem of the incentive system of Bitcoin is that it doesn't reward "Full-node operation" only "Mining operation", the result would be a more distributed and democratic system.

The reason is because it isn't expensive to run a full node while running the mining operation is expensive.

"No one anticipated pool mining, so we considered all miners to be full nodes and almost all full nodes to be miners.

https://cointelegraph.com/news/satoshis-best-kept-secret-why...

Now apply this reasoning to our current financial system and how it works. Think of the system in terms of people making decisions and not in terms of perfectly continuous mathematical models. Debt is not a physical object. Inflation and spending are not physical forces of nature.

I was about to start this book but noticed it was from 2015. Is it still relevant? :)

Actually I think bitcoin's history is more relevant than its present :). Definitely worth a read.

The psychological process you've described happens with any investment.

>Its a constantly evolving monster that will gobble up everything eventually (Isn't it beautiful :))

but the problem is that it isn't really. Bitcoin coin cash implemented big block size to expand the transaction capacity. Meanwhile bitcoin stuck with the small block size.

... And Bitcoin got the segwit upgrades that fixes the transaction malleability issue, enabling users to make new types of contract transactions like payment channel and lightning network transactions, which Bitcoin Cash specifically refused. One thinks it's evolving by tweaking constants, and one is working on enabling new things.

> Its a constantly evolving monster that will gobble up everything eventually (Isn't it beautiful :))

As I recall, that has always been part of the dream. Plus the potential for early adopters to become unimaginably wealthy ;)