Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

On the growing risk of the 51% attack against BCash

notehub.org

Post reply on HN
Print view View original

41–50 of 63 posts

Re: On the growing risk of the 51% attack against BCash

#41

Post by JepZ »

When the BTC-BCH split happend I didn't really understand what everybody was talking about transaction fees. So after BTC continued to rise the past months and we have transaction fees >20$ I understand what was discussed earlier this year.

I just wonder why the core devs were against increasing the blocksize. Yes, it would not have been a solution for the next 20 years, but at least it would have left the fees at a place where BTC was still suitable for the everyday use until other solutions could be established.

Re: On the growing risk of the 51% attack against BCash

#42

Post by em3rgent0rdr »

tree_of_item wrote: post #38
gizmo686 wrote: post #27

Earlier quoted context omitted.

If you do a double spend attack of any significance, the other party will notice. Unless you are doing something completely anonymous, the other party will also know who you are and be able to pursue other remedies (eg. sue you for fraud).

How could you possibly sue someone for double spending on a blockchain? I don't get it, isn't the point of this blockchain stuff to be the wild west? What is the grounds for a lawsuit? Why would a 51% attack be illegal? "Code is law", right?

"isn't the point of this blockchain stuff to be the wild west?"

Not necessarily. Even in an "Code is law" society, there could be some code/law specifically addressing 51% double spending attacks.

Re: On the growing risk of the 51% attack against BCash

#43

Post by moduspol »

em3rgent0rdr wrote: post #36
oh_sigh wrote: post #28

Earlier quoted context omitted.

It would incur a huge opportunity cost for them to switch over, especially because they would tank bcash by doing it.

Yes it would. This article is surprisingly ignorant of past discussion of the %51 attack.

It also fails to account for the possibility that Bitcoin Cash becomes more valuable, and then Bitcoin is the "vulnerable" chain.

It also doesn't mention the name of the author, and the HN user who posted this was created two hours ago.

HNers should understand there is a huge political divide in the Bitcoin community right now on this very subject. Anonymous write-ups of known attack vectors framed as genuine warnings are nothing more than unsubstantiated fear, uncertainty, and doubt.

Re: On the growing risk of the 51% attack against BCash

#44

Post by jancsika »

sillysaurus3 wrote: post #16
toomim wrote: post #11

A 51% attack only allows you to (1) censor transactions for a short time, and (2) double-spend. The first part is a mild nuisance. The second one can be retaliated against off-chain, which puts the attacker (who must be a large entity) at great risk in the real world.

Yes. A 51% attack is a popular whipping boy. It's not as dangerous as it seems. The price would tank, for sure. But your money won't be gone. And eventually when the attack dies down you'll get your wealth back. However, it remains a very interesting way for governments to completely destroy Bitcoin. It's the only way, and they have to do it now before centralized, dedicated mining facilities come online. It would di…

> And eventually when the attack dies down you'll get your wealth back.

Here's the attack the author is describing:

block 42: exchange sends 1 BCH to foo

block 42: in secret, BTC meanies solve a different block 42

block 62: foo is comfortable believing the transaction is confirmed, so is the exchange

block 62: in secret, BTC meanies are now solving secret branch block 63, or possibly secret branch block 64/65/etc. depending on how much hashing power they have

block 123: foo and many others have made transactions, exchange has made many other transactions. Maybe some users transfered out dollars during this time.

block 123: BTC meanies reveal their branch, which is at block 130 or whatever, propagate it, and all the BCH nodes accept it as having the greatest total difficulty because that is the rule for how the protocol works

block 150: the once secret branch now has twenty confirmations, some of those blocks holding transactions from bona fide BCH users. Maybe the exchange halted trading at block 124 because something looked screwy. Maybe the exchange eats the cost of sending more BCH to the users whose coins "disappeared."

Or perhaps the BTC meanies keep their branch secret for several days.

The point is the appearance of the secret branch undermines faith that 20 confirmations is enough to confirm a transaction, and the users don't know how to judge what constitutes the canonical state of the blockchain.

Avoiding that kind of confusion was a stated goal of the Bitcoin whitepaper.

Hm... it occurs to me that such an attack could be even more insidious. I'll call it the "Robinhood Attack." The BTC meanies could begin hashing their secret branch and include all the transactions they see happening on the main chain. Then for each new secret block, they could send the block reward to a random recipient chosen from the transactions on the previous block.

When the BTC meanies finally reveal their alternate chain, it gets picked up and includes all the transactions, except the block rewards have now been moved from the miners to the speculators. :)

Edit: typo

Re: On the growing risk of the 51% attack against BCash

#45

Post by notsrg »

JepZ wrote: post #41

When the BTC-BCH split happend I didn't really understand what everybody was talking about transaction fees. So after BTC continued to rise the past months and we have transaction fees >20$ I understand what was discussed earlier this year. I just wonder why the core devs were against increasing the blocksize. Yes, it would not have been a solution for the next 20 years, but at least it would have left the fees at a…

Increasing the blocksize increases centralization at a time where miners have unprecedented power.

Re: On the growing risk of the 51% attack against BCash

#46

Post by xiphias »

yourrng wrote: post #24
jVinc wrote: post #20

I like the math and theory on this. But really why would someone who's making millions on the status quo risk tanking the market for a one time gain? Seems unlikely.

You don't have to risk anything. If the trend continues, at some point one would be able to rent the mining power required. On top of it, if you are a small miner, and you can't afford capital investments required to keep up with latest mining hardware, while loosing market share, you can use your soon-osolete hardware to make a one last buck. A big buck this time.

At this point 5-10% of hash rate is a huge cost in mining hardware (maybe hundreds of millions). Also the hardware improvements already slowed down a lot. Antminer S9 is more than a year old. I wouldn't risk decreasing the BTC price by attacking BCH.

Re: On the growing risk of the 51% attack against BCash

#47

Post by JepZ »

notsrg wrote: post #45
JepZ wrote: post #41

When the BTC-BCH split happend I didn't really understand what everybody was talking about transaction fees. So after BTC continued to rise the past months and we have transaction fees >20$ I understand what was discussed earlier this year. I just wonder why the core devs were against increasing the blocksize. Yes, it would not have been a solution for the next 20 years, but at least it would have left the fees at a…

Increasing the blocksize increases centralization at a time where miners have unprecedented power.

How does it increase centralization and why have miners more power than 5 years ago? (Just because the value increased? Do they have more influence to the community?)

Could you pleace elaborate on that?

Re: On the growing risk of the 51% attack against BCash

#48

Post by sillysaurus3 »

jancsika wrote: post #44
sillysaurus3 wrote: post #16

Earlier quoted context omitted.

Yes. A 51% attack is a popular whipping boy. It's not as dangerous as it seems. The price would tank, for sure. But your money won't be gone. And eventually when the attack dies down you'll get your wealth back. However, it remains a very interesting way for governments to completely destroy Bitcoin. It's the only way, and they have to do it now before centralized, dedicated mining facilities come online. It would di…

> And eventually when the attack dies down you'll get your wealth back. Here's the attack the author is describing: block 42: exchange sends 1 BCH to foo block 42: in secret , BTC meanies solve a different block 42 block 62: foo is comfortable believing the transaction is confirmed, so is the exchange block 62: in secret , BTC meanies are now solving secret branch block 63, or possibly secret branch block 64/65/etc.…

Sure, but that is still a costly one-time attack. It only affects those who transfer coins, not those who hold them. So if you're holding wealth, it will be safe.

It will certainly tank te price for a little bit, but the market will readjust based on the probability that it will happen again.

And the theory is that miners would need to choose between spending $180k to execute this attack for a couple hours, causing relatively little damage, or just mining BCH faster than anyone else, which would increase their real-world bottom line.

The only one to do this consistently would be a government, probably. It's hard to imagine someone willingly burning $200k mostly for lulz. But I suppose BTC millionaires have more than enough, if they wanted to.

Also, why was this article flagged? That's frustrating. It was technical and informative. I don't see a vouch button either.

Re: On the growing risk of the 51% attack against BCash

#49

Post by moduspol »

JepZ wrote: post #41

When the BTC-BCH split happend I didn't really understand what everybody was talking about transaction fees. So after BTC continued to rise the past months and we have transaction fees >20$ I understand what was discussed earlier this year. I just wonder why the core devs were against increasing the blocksize. Yes, it would not have been a solution for the next 20 years, but at least it would have left the fees at a…

I think they have valid arguments for keeping smaller blocks, but this FUD is getting ridiculous.

This very post has no new ideas, has no listed author, and is posted by a two hour old user account. It's not even about discussing what's best for Bitcoin any more.

Re: On the growing risk of the 51% attack against BCash

#50

Post by Sir_Substance »

tree_of_item wrote: post #38
gizmo686 wrote: post #27

Earlier quoted context omitted.

If you do a double spend attack of any significance, the other party will notice. Unless you are doing something completely anonymous, the other party will also know who you are and be able to pursue other remedies (eg. sue you for fraud).

How could you possibly sue someone for double spending on a blockchain? I don't get it, isn't the point of this blockchain stuff to be the wild west? What is the grounds for a lawsuit? Why would a 51% attack be illegal? "Code is law", right?

>I don't get it, isn't the point of this blockchain stuff to be the wild west?

I guarantee you no judge in the world is going to accept that interpretation.

>"Code is law", right?

From a legal perspective, obviously not, and given how well accepted the etherium hard fork was in the wake of the DOA smart contract being exploited, you can be pretty sure that a large percentage of the crypto-user population doesn't think so either.

Post reply on HN