We can confirm that there was a successful 51% attack on Ethereum Classic • Hacker News Board

Re: We can confirm that there was a successful 51% attack on Ethereum Classic

#161

Post by schoen » Tue, Jan 08, 2019, 3:15 AM UTC

philwelch wrote: post #121

johndunne wrote: post #7

This is inevitable. Ethereum Classic (ETC) isn't the only currency such attacks have been successful on. The site https://www.crypto51.app/ puts the cost of running a 51% attack on ETC at ~$5k per hour. The incentive for running these attacks for profit becomes higher as the market cap of these coins increases, making long-term 'investment' in these coins nonsensical.

I wonder if there's a formula for the maximum wallet or transaction size you can safely have on a particular cryptocurrency given the cost of a 51% attack. If a 51% attack costs $5,000 per hour, you're only going to do it if you can extract more than $5,000 of value (or if your motivation is to see the world burn). In theory you might be able to set things up such that the cost of a 51% attack is necessarily greater…

Budish (2018) gives some attempts at formulas to determine attackers' incentives under various hypotheses: http://faculty.chicagobooth.edu/eric.budish/research/Economi...

It's tricky to directly apply these to cryptocurrency adopters' risk, although they apply more directly to "transaction size" than to "wallet size". You can probably apply some of Budish's calculations to transaction size unless you anticipate there's a way that your transaction counterparty could be defrauding you and other people in roughly the same way at roughly the same time. (For transaction size, the main risk is that you give someone something else of value in exchange for a transaction that's subsequently removed from the consensus history.) But for wallet size, the risk is not that someone steals your cryptocurrency from your wallet, but that your cryptocurrency becomes less valuable because other people recognize risks more immediately as a result of discovering and publicizing a successful attack. But that has most to do with how other people respond to the news of the successful attack, which is harder to predict.

Re: We can confirm that there was a successful 51% attack on Ethereum Classic

#162

Post by sgeisler » Tue, Jan 08, 2019, 3:18 AM UTC

lokedhs wrote: post #156

Klathmon wrote: post #32

Earlier quoted context omitted.

You won't find any disagreement from me that there is a LOT of garbage out there using blockchains and cryptocurrency tech inappropriately. The extreme vast majority is trash, manipulative, and awful in most every way. Not to mention the charlatans and fanboys that hype it as able to do everything and anything perfectly. But the few that use it well are doing fairly well, and innovation is happening. You shouldn't lo…

Are they though? I'm honestly curious as to what those things could be. A lot of people talk about good uses of blockchains, but all examples I've seen are either not actually good uses, or they have some flaw that makes practical implementation impossible (or at least highly unlikely). What examples of good use of blockchains are you thinking of?

* Trustless time stamping to proof that you had certain information at a certain point in time without revealing it https://opentimestamps.org/

* (depending on your stance on personal liberty) censorship resistant, pseudonymous e-cash/e-gold (Bitcoin)

Re: We can confirm that there was a successful 51% attack on Ethereum Classic

#163

Post by s3xham » Tue, Jan 08, 2019, 3:26 AM UTC

jancsika wrote: post #108

jonahx wrote: post #81

Earlier quoted context omitted.

Would you mind elaborating on this for someone unfamiliar?

Imagine all the Bitcoin miners out there right now using their ASICs to do extremely efficient hashing in the hopes of generating a block reward. Let's didactically suppose there are 100 such miners total. Now imagine Dinkycoin comes along and releases their cryptocurrency that uses the same hashing mechanism for the block reward. Initially they have the block difficulty level pretty low as there aren't that many peo…

This phenomenon of "mysterious Bitcoin miner smashing your shitcoin" was pretty common in the early days, especially around the time namecoin merged mining came into play. Any theoretical attack that might be possible on Bitcoin would be attempted elsewhere if you didn't have the hashpower to carry it out on mainnet.

Origin of the term shitcoin would most likely be unable to defend the 51% attack

Good write-up op

Re: We can confirm that there was a successful 51% attack on Ethereum Classic

#164

Post by anonymous_ch » Tue, Jan 08, 2019, 3:28 AM UTC

jancsika wrote: post #108

jonahx wrote: post #81

Earlier quoted context omitted.

Would you mind elaborating on this for someone unfamiliar?

Imagine all the Bitcoin miners out there right now using their ASICs to do extremely efficient hashing in the hopes of generating a block reward. Let's didactically suppose there are 100 such miners total. Now imagine Dinkycoin comes along and releases their cryptocurrency that uses the same hashing mechanism for the block reward. Initially they have the block difficulty level pretty low as there aren't that many peo…

If this is true, then does the pre-existing size of BTC and other widely mined crypto-currencies create a moat? How can a new up-start coin ever hope to gain significant traction with the possibility of a 51% attack?

Re: We can confirm that there was a successful 51% attack on Ethereum Classic

#165

Post by SilasX » Tue, Jan 08, 2019, 3:37 AM UTC

anonymous_ch wrote: post #164

jancsika wrote: post #108

Earlier quoted context omitted.

Imagine all the Bitcoin miners out there right now using their ASICs to do extremely efficient hashing in the hopes of generating a block reward. Let's didactically suppose there are 100 such miners total. Now imagine Dinkycoin comes along and releases their cryptocurrency that uses the same hashing mechanism for the block reward. Initially they have the block difficulty level pretty low as there aren't that many peo…

If this is true, then does the pre-existing size of BTC and other widely mined crypto-currencies create a moat? How can a new up-start coin ever hope to gain significant traction with the possibility of a 51% attack?

The parent’s scenario was predicated on using the same hashing algorithm for the mining/proof of work as the dominant coin. So you can avoid that by using a different hashing algorithm.

Monero plans to do that indefinitely, by constantly changing to a new hashing algorithm that doesn’t have dedicated ASIC chips for it.

Re: We can confirm that there was a successful 51% attack on Ethereum Classic

#166

Post by krrrh » Tue, Jan 08, 2019, 3:53 AM UTC

skwirl wrote: post #99

CydeWeys wrote: post #43

Earlier quoted context omitted.

Because the market cap isn't truly $5M. That figure uses the naive calculation of "whatever coins sold for most recently times the total number of outstanding coins". But if you tried to sell some significant fraction of the outstanding coins then the sale price would plummet, and you'd never actually reach $5M total. The actual depth of the order book at any given moment isn't close to $5M, plus a lot of outstanding…

>But if you tried to sell some significant fraction of the outstanding coins then the sale price would plummet, and you'd never actually reach $5M total. You can't focus on sellers and ignore buyers when discussing the valuation of something. If many people want to sell and few people want to buy the price will of course go down. But if many people want to buy and few people want to sell then the price will go up. Th…

Another salient point is that when an entity buys all the shares of a company they pay a premium over the current share price indicating that the company is perceived by the market to be more valuable with a single shareholder than with widely dispersed shares.

A cryptocurrency with a single sharehodler is a non sequitor and essentially worthless.

Re: We can confirm that there was a successful 51% attack on Ethereum Classic

#167

Post by samatman » Tue, Jan 08, 2019, 3:53 AM UTC

pryce wrote: post #144

xur17 wrote: post #139

Earlier quoted context omitted.

> You can increase your chances by increasing your mining percentage. Make it a 75% attack and you have an 18% chance of success. To have a 50/50 chance of success you really need to mount about a 90% attack which is pretty ambitious. I'm not sure this is accurate. You don't need to mine 6 blocks in a row on the existing chain. Clients are programmed to recognize the longest chain, so you just need to silently mine n…

The six-blocks-in-a-row problem seems less of an impedence to me - because the "legitimate" blocks are still available to the malicious actor. Therefore, if our malicious miner identifies that they have had poor luck and begun to fall behind the "legitimate" chain by a block or two they can start the make-a-longer-competing-chain process over again from the legitimate head with -as far as I can see- no downsides exce…

They still have to get out ahead by six blocks, however.

This drops to zero every time they start over, so it doesn't change the basic calculus.

Re: We can confirm that there was a successful 51% attack on Ethereum Classic

#168

Post by madmax96 » Tue, Jan 08, 2019, 3:56 AM UTC

anonymous_ch wrote: post #164

jancsika wrote: post #108

Earlier quoted context omitted.

Imagine all the Bitcoin miners out there right now using their ASICs to do extremely efficient hashing in the hopes of generating a block reward. Let's didactically suppose there are 100 such miners total. Now imagine Dinkycoin comes along and releases their cryptocurrency that uses the same hashing mechanism for the block reward. Initially they have the block difficulty level pretty low as there aren't that many peo…

If this is true, then does the pre-existing size of BTC and other widely mined crypto-currencies create a moat? How can a new up-start coin ever hope to gain significant traction with the possibility of a 51% attack?

You have to use a PoW that's different enough from popular blockchains that they are unable to use speciality hardware against you.

Re: We can confirm that there was a successful 51% attack on Ethereum Classic

#169

Post by philwelch » Tue, Jan 08, 2019, 3:59 AM UTC

schoen wrote: post #161

philwelch wrote: post #121

Earlier quoted context omitted.

I wonder if there's a formula for the maximum wallet or transaction size you can safely have on a particular cryptocurrency given the cost of a 51% attack. If a 51% attack costs $5,000 per hour, you're only going to do it if you can extract more than $5,000 of value (or if your motivation is to see the world burn). In theory you might be able to set things up such that the cost of a 51% attack is necessarily greater…

Budish (2018) gives some attempts at formulas to determine attackers' incentives under various hypotheses: http://faculty.chicagobooth.edu/eric.budish/research/Economi... It's tricky to directly apply these to cryptocurrency adopters' risk, although they apply more directly to "transaction size" than to "wallet size". You can probably apply some of Budish's calculations to transaction size unless you anticipate there…

> But for wallet size, the risk is not that someone steals your cryptocurrency from your wallet, but that your cryptocurrency becomes less valuable because other people recognize risks more immediately as a result of discovering and publicizing a successful attack. But that has most to do with how other people respond to the news of the successful attack, which is harder to predict.

And that's going to be governed by irrationality, especially given the irrationally exuberant bubble effects of the cryptocurrency market in the first place. But once that wears off...I don't know. If you somehow managed to make a cryptocurrency that couldn't be profitably attacked, and made the base minimum cost of such an attack high enough, you'd still be vulnerable to a state-level adversary, but every system is vulnerable to a state-level adversary.

Re: We can confirm that there was a successful 51% attack on Ethereum Classic

#170

Post by bouncycastle » Tue, Jan 08, 2019, 4:17 AM UTC

ThrustVectoring wrote: post #142

bouncycastle wrote: post #127

Earlier quoted context omitted.

But if they throw hashpower on Dinkycoin, wouldn't Dinkycoin's difficulty increase? If the difficulty for Dinkycoin re-targets more aggressively, then Dinkycoin would quickly gain more security in the process!

The attacks tend to do something like "build a chain based off Dinkycoin's previous blocks, but don't publish it until a later date". The difficulty changes can't help because the chain doesn't know about the attack until the attacker publishes a longer chain.

Yes, of course! And once the damage has been done, it's even harder to reverse. Perhaps if Dinkycoin's devs are fast enough, they could release a fork to undo it?