Ethereum Is a Dark Forest • Hacker News Board

Re: Ethereum Is a Dark Forest

#241

Post by Taek » Fri, Aug 28, 2020, 11:32 PM UTC

TeMPOraL wrote: post #233

Taek wrote: post #210

Earlier quoted context omitted.

> 1) Proof-of-work systems are pure, unadulterated energy waste (and an ecological disaster as long as we depend on fossil fuels). They cannot, ever, be allowed to become significant in the economy, lest our future will be building a Dyson sphere around the Sun just to power everyone's ability to pay for a hot dog on their way to work. Proof of Work gets a bad reputation because people have a hard time wrapping their…

I haven't done the exact math (I'm sure somebody did, though), but I based on some rough estimates I do believe that for widespread use as currency, trust-based systems beat trustless systems significantly in terms of energy efficiency. Partly because scaling factors - trust scales sublinearly with the number of participants (like average path between two arbitrary vertices in an acyclic graph). PoW energy use scales…

> PoW energy use scales superlinearly.

Citation needed? I don't believe this is true. And techniques like payment channels allow you to stretch a significant amount of mileage out of a single transaction. The Sia network for example has payment channels that have over a million payments made per on-chain transaction.

Re: Ethereum Is a Dark Forest

#242

Post by lmeyerov » Fri, Aug 28, 2020, 11:33 PM UTC

literallycancer wrote: post #234

lmeyerov wrote: post #216

Earlier quoted context omitted.

For background, I've built verifiers for harder languages, reviewed papers for crypto systems, seen the inside of crypto operations & crypto security startups, and help build software for adversarial, investigative, & high performance scenarios. I'm not an expert in blockchain stuff, but I'm also not unfamiliar with the software challenges. * I would agree that paying consultants to audit contracts is prohibitively e…

How do you check for front running? What if your checker tells you that you can't fix it just by changing your own code?

I'm unclear what level you're asking this at.

-- Modeling: You start with basics like using the small universe assumption to bound checking to X transactions. I'd expect most front running to show up as small cycles here, so the typical case is a super small X. Later, you might get into a time cost semantics to better tune what you consider interesting, but almost no one in the crypto space is at the level of modeling maturity. I'd predict a team's time is way better spent building up a stdlib of contract checks, verified contract helpers, & whitebox attack heuristics/guides.

-- Modeling II: Also, in verification, it's way better (ex: realizes more of the ROI) to verify the program has the properties you want ("money goes from a->b without getting stuck"). You can dream up individual attacks and model those one by one ("front-running where ..."), but then you potentially miss some, or some aspect of one. That's basically the difference between verification and testing. You still do stuff like check sample scenarios & individual attacks, but that's more about testing the verification conditions & model fidelity.

-- Fixes: A good (while still cheap & easy) checker gives you summarized examples of attacks. Likewise, it makes it interactive, so you can tune what you consider in/out of scope. More R&D-level verifiers suggest patches (verification and synthesis are two sides of the same coin), but that's not necessary. If your idea sucks or the attack is unavoidable, the verifier isn't the problem, and if you decide to still proceed with the now proven-bad idea, you can at least now price the risk in.

Re: Ethereum Is a Dark Forest

#243

Post by seibelj » Fri, Aug 28, 2020, 11:34 PM UTC

xg15 wrote: post #214

ak_111 wrote: post #156

Earlier quoted context omitted.

This is a well-articulated list, and captures a lot of my negativity towards crypto. But as I like to keep an open mind about new tech, I always wonder if something like this could have been said about the early days of Linux: 1) Open-source is pure unadulterated theft of other peoples' work. This model cannot ever be allowed to become significant in the software economy. 2) That people who are into open source have…

Open Source didn't have as its only unique feature the ability to evade regulation.

I work in the crypto space and the ability to side step regulations that I find ridiculous is a main selling factor. Red tape has stifled not just the financial space but the entire economy. If you want to cut hair you need 2000 hours of training and tens of thousands of dollars paid to scam schools, when YouTube and brochure on how to sweep hair from the floor is perfectly adequate. How is regulation fair again?

Re: Ethereum Is a Dark Forest

#244

Post by literallycancer » Fri, Aug 28, 2020, 11:35 PM UTC

_7fvc wrote: post #224

I'm not sure when Ethereum supporters would wake up. Ethereum's design creates this kind of Dark Forest problem. When it is programmable, you are at risk of being overpowered by bots. Computers are far more efficient than humans. Smart contracts are not smart. If they're too complicated (smart), humans can't comprehend them. So they're mostly dumb. Bots will overrun humans. With Ethereum, you know you're entering a d…

Most trading is automated. This is not something that's unique to smart contract platforms. How come the world didn't end already?

Re: Ethereum Is a Dark Forest

#245

Post by shajznnckfke » Fri, Aug 28, 2020, 11:35 PM UTC

Taek wrote: post #187

josh2600 wrote: post #10

I really think that all of this DeFi stuff is playing with fire. If these tools scale large enough, it's easy to imagine breaking the right link in the system at the right time to cause catastrophic failures. Remember that all complex systems operate in a degraded state. If there's ever a way that only part of a complicated swap executes correctly the trade can get really far out of position. People in Ethereum land…

This DeFi stuff is playing with fire because the products being released have significantly outpaced the state of the art in building safe smart contracts. To make an analogy, imagine that instead of DeFi, we were talking about skyscrapers. Imagine that thousands of engineers funded by millions of people who believed in them were building 25 kilometer tall towers using technology that they discovered in Isaac Arthur…

I don’t disagree with your claim here, but aren’t existing systems even worse? The conventional electronic payments system is in many ways permissionless. Even if crypto doesn’t live up to all the promises it makes, it may still add value.

If crypto is building poorly-engineered space elevators to get out payments to and from the sky, maybe the current system is throwing them in artillery and parachutes and hoping they land where you aim.

Re: Ethereum Is a Dark Forest

#246

Post by schoen » Fri, Aug 28, 2020, 11:36 PM UTC

iSnow wrote: post #228

schoen wrote: post #198

Earlier quoted context omitted.

> They _cant'_ be corrupted, because a combination of incentives and cryptography keep everyone safe. I would like this to be right but then I ran into https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3197300 which discusses economic limitations to the incentives for blockchain mining. (One part is that if a cryptocurrency gets too valuable, the value of a double-spend attack could exceed some models of the reward…

The paper is 2y old - which is a bit of time in blockchain space - and yet, no one has performed this attack successfully in the wild. So, either everyone is just loving Bitcoin and wouldn't harm it for profit, or it's not so easy (hint: you cannot short Bitcoin for hundreds of millions, it's not liquid enough). Also, this is strictly about PoW, the ETH roadmap (and this is where we are coming from in this submission…

> you cannot short Bitcoin for hundreds of millions, it's not liquid enough

That seems like a great explanation, but the paper's argument is definitely not that cryptocurrencies can't exist or can't work, just that they have a limited range of levels of adoption where the incentives will continue to point in the right direction. In the paper's model it seems that Bitcoin has just not reached that level, right? If you could double-spend or short enough value in it, the incentives would reverse.

> PoS where attacks are potentially way more expensive

I don't think Eric Budish agrees that PoS verification is categorically immune to this. The very last sentence of the paper mentions that it "will be interesting to watch [PoS] research develop, and see whether or not it constitutes a valid response to the critique in this paper".

Re: Ethereum Is a Dark Forest

#247

Post by seibelj » Fri, Aug 28, 2020, 11:36 PM UTC

orthecreedence wrote: post #96

dawkins wrote: post #80

Earlier quoted context omitted.

> Proof-of-work systems are pure, unadulterated energy waste The more you pollute, the richer you get. I can't understand why nobody talks about this.

They do, they blabber about how it's not actually waste because people value the waste. It's effectively a logical extension of the marginal value theory applied to ignorant self-destruction. While they will talk about marginal value, they don't talk about entire systems incentivized to operate by creating value destructively , aka externalization at its finest.

If there is a geothermal source of energy in a location that is impossible to transmit electricity from, and you use it to mine bitcoin, is this immoral and a “waste” of energy? This literally happens in Iceland.

Re: Ethereum Is a Dark Forest

#248

Post by seibelj » Fri, Aug 28, 2020, 11:39 PM UTC

TeMPOraL wrote: post #157

young_unixer wrote: post #151

Earlier quoted context omitted.

> It's a wild west. Trading unregulated money tends to disproportionately attract the worst kind of people. It means that you have to take precautions instead of the government telling you what you can or can't do (using money they took from you anyway). The energy concerns are valid. I hope that the cryptocurrency that ends up winning is energy efficient.

> It means that you have to take precautions instead of the government telling you what you can or can't do (using money they took from you anyway). That's true! And I do prefer the world in which I pay the government to take care of this, because as an individual I have no chance against organized, professional scammers. Wild West is a fun thing to watch in movies (or at least the fictional recreation of it, with mo…

What’s so wonderful about voluntary society is that you can avoid the whole and others can dive right in! Different strokes for different folks.

Re: Ethereum Is a Dark Forest

#249

Post by seibelj » Fri, Aug 28, 2020, 11:41 PM UTC

robbomacrae wrote: post #73

emerged wrote: post #52

Writing this sort of bot seems like a legitimately fun and interesting thing to work on, but somehow I have less than zero interest in actually doing it. There's just something intrinsically repulsive about the entire blockchain world to me where I just don't want to touch it. I don't mean to offend people who do love blockchain tech, in many ways I don't blame you. But is this feeling I have somewhat common? I'm not…

The feeling is common and here is a justification for you. The bitcoin network now takes 7 nuclear power plants to run [1]. Last year it was responsible for 0.21% of the worlds supply of energy [2]. Whatever benefits it brings (are there any?) I don't think it justifies the environmental cost even if some of the energy used is renewable and close to the source. Also think what all those engineers could do with all th…

Uh... who cares? How many power plants keep houses cool when you can just sit in the shade and meditate? If the market pays the fee then that’s the reality whether you like it or not. I find the fact you use a computer at all mildly distasteful given the energy required to power it.

Re: Ethereum Is a Dark Forest

#250

Post by sickygnar » Fri, Aug 28, 2020, 11:41 PM UTC

cesarb wrote: post #176

sickygnar wrote: post #14

The whole thing is a complicated, wacky game. The DeFi stuff is especially fun right now. Opportunities (and danger) abound. There is so much money locked up in DeFi. It's not necessarily always good for the bots either. They can be exploited and tricked as well.

> There is so much money locked up in DeFi. Is the money really "locked up"? No money actually enters these systems; whenever someone buys a token with money, there was someone else selling that token for money, and the money went from the buyer to the seller, who is free to do anything with that money.

hmm, maybe you are referring to money as currency, where i am referring to it as assets (tokens)

anyway, yes there is money locked in the lending platforms. most of it is used as collateral, which is locked when borrowed against. the reserve ratios are high (60-80%), so unlike a regular bank, no money is "printed" when lent.

"real money" does enter these systems. people spend resources to acquire real money which is subsequently traded then locked into these platforms. also, while a lot of ethereum tokens are simply minted with no real backing, some are mined (even erc20), which again people dedicate finite resources to