back
29 comments
Ethereum does to coding what wagering real money does to a friendly game of Poker

Expected value (EV) changes behavior significantly.

I'm skeptical too but didn't let that keep me from learning Solidity and the Ethereum Virtual Machine (EVM). This dark forest is like having a red team looking at your code 24/7. I feel like I learned some novel things from it.

Then why are we building it? To fleece people who are not as smart? All this just seems insane to me.
It's a super massive stretch to imagine this invalidates Ethereum. Only some specific types of contracts are vulnerable to front-running issues and there are workarounds. The article is about the difficulty of being the first to take from a contract offering free money. This isn't an issue about regular contracts being unsafe to use.

I think the title misportrays the issue as being much more general than it is. For some reason I see the article brought up often in Twitter conversations by people who don't understand Ethereum and are trying to argue as if regular transactions are vulnerable to theft.

It's not just contracts that are the issue. You need to be an expert in cyber security just to not to get robbed.

The general recommendations I've heard go like this:

- Have a dedicated machine just for transferring money

- Only boot from a read only flash drive to interact with the network

- Store credentials in a separate place, maybe on paper in a very hidden spot

- Transfer a small amount of money before large transfers and wait 24 hours to make sure it works ok

- Spread your money between many accounts so it can't be stolen all at once

Does that not seem at least a little insane to anyone? No one is going to be able to practically use the network without some expert user as a proxy. And I don't see that changing any time soon with how terribly designed the protocol is at controlling these issues.

> - Have a dedicated machine just for transferring money - Only boot from a read only flash drive to interact with the network - Store credentials in a separate place, maybe on paper in a very hidden spot

You're overstating the problem. Companies like Ledger make "hardware" wallets as you're describing and they're not complicated to use. If you're using crypto in small amounts and not as long-term storage, I think a phone wallet or desktop wallet is fine.

I think you’re understating the problem. Imagine having your entire life savings on a flash drive. It’s a disaster. Crypto has no chance of replacing fiat until that and many other issues are resolved.
The only reason it's popular is the low bar of entry, stealing large sums of money, and (to an extent) the promise of some utopia of truly free wealth transfer and algorithmic mechanisms around it.

But IMO etherium is not that by a long shot. From the protocol, to all existing contracts, and the user experience resulting from those, etherium is fucking amateur hour.

Why did we build nuclear weapons? Even knowing that they could potentially annihilate humanity - indeed, that was the point. Because if we didn't, Nazi Germany would've, and then we would've been the ones annihilated.

History has a shape, and that shape is largely dictated by science, technology, and game theory. Whenever a technology gets discovered that gives its controllers a significant advantage over others, and its discovery is obvious enough that it could likely have been discovered elsewhere, you have to adopt it. Otherwise, others will, and gain enough of an advantage over you to threaten your existence. There's a reason that the permanent members of the UN Security Council are the U.S, Russia, China, Britain, and France, and that Germany and Japan are not on it.

Similar dynamics are why we still have arms races; climate change; lying & exaggeration; corruption; profiteering; and privatization. For more thoughts on this, you might want to check out this SlateStarCodex blogpost:

https://slatestarcodex.com/2014/07/30/meditations-on-moloch/

It's tempting to look at a situation as a whole and say "This is bad for everyone. Why are we doing it?" without looking at the incentives and levers of control for each of the individual actors. Sometimes, these people even resolve to not engage in the bad behavior themselves, not realizing that not engaging in the bad behavior is marginalizing themselves out of the discourse, and that's why bad behavior persists.

So by the analogy, the reason to develop Ethereum is so that we can fleece people before the nazis can fleece them?
Pretty much. I was gonna say that Nazis might be a strong term for people who just wanna get rich in today's day and age, but a lot of the people who have gotten rich off crypto are literally being called Nazis by those who oppose them.
> but a lot of the people who have gotten rich off crypto are literally being called Nazis by those who oppose them.

Bullshit, show me where this is happening...

They are not merely being called nazis, they (=some of them) are nazis.

https://www.economist.com/united-states/2022/02/05/the-charm...

Note (for dang): it's "Ethereum", not "Etherium"
I'm curious whether you could eliminate or at least mitigate this "dark forest" environment with privacy preserving transactions?

There are privacy preserving smart contract schemes (see https://eprint.iacr.org/2020/543 for one example) however to my knowledge few if any have ever been implemented/utilised at scale.

At least in theory this should eliminate or at least greatly mitigate the opportunity for MEV since the actions taken and the results are opaque. In reality however I'm not sure and I am absolutely not qualified to answer.

Discussed at the time:

Ethereum Is a Dark Forest - https://news.ycombinator.com/item?id=24308588 - Aug 2020 (447 comments)

I hope your link is not correct!

The Aug 2020 article is about real estate transactions at the close of escrow. Does that mean people are accidentally using Ethereum? Yikes!

The discussion is about a common scam in real estate, no Ethereum involved. The point the poster was making is that scams occur whenever there's potentially large profit to be made, and they're as much a feature of the "real" financial system as the cryptocurrency world.
Yeah this has been posted before, and the message is always the same: If you want to trade in ethereum, just remember that you're dealing with a technology created by psychopaths. The Dark Forest wasn't somewhere you went for a walk.
Is the benefit of Ethereum really larger than the value these intelligent people could be creating in non-crypto industries? I see the allure. They're super hard, super high-stakes problems. Is there really no other industry that provides this level of flow and excitement for these researchers?
It's kind of the opposite.

Highly intelligent people are most productive when they are tackling problems they enjoy. You cannot just order them around and tell them to go tackle climate change or whatever.

As long as they enjoy tackling crypto shaped problems, they will keep jumping in until enough crypto shaped problems are solved that only really frustrating ones are left.

If you want to change how things work, you have to change the environment for problem solving. Make it more or less enjoyable.

You don't need to change the environment, you need to change the compensation. Do you think that for the last 15 years highly intelligent people have found the most enjoyment in developing advertising software? Or found the most enjoyment in investment banking in the years before that?
The compensation is not just financial is how I would put it.

The compensation for working in those fields is often that you're working alongside other smart people, in a situation where your work is highly valued.

The money is just a factor in the equation, money equals respect in a lot of different ways.

People want to work on projects where meaningfulness and respect flows to them. Money is one of the things you want to see flowing.

Crypto offers that money, and offers a lot of technical challenges because programming distributed systems that must be error free is tricky.

In terms of the ad tech, a lot of the work on ad tech happens because Facebook/Google/et all have cultivated environments that offer that kind of respect, and offer serious technical problems.

Advertisements per se are not so much a part of it.

I’d probably rather they keep flocking to the crypto honeypot instead of doing weird pseudo intellectual crap to my teams code base.
Excuse my useless comment but lol
Ethereum Layer1 should be only for storing meta data and provide trust and arbitration when necessary.
I don't really understand this take, and I hear it a lot. What's the point of using a blockchain if you're going to abstract everything away to a centralized ledger? The whole "gas is expensive" shtick doesn't warrant centralizing your ledger and doing all of the important stuff off-chain. Am I missing something here, or are layer 2 protocols really that wishy-washy?
Blockchains are just tools. We should focus use cases to choose different types of blockchains. Layer1, Layer 2, sidechains, private chains can be combined together to solve the problem.
> What's the point of using a blockchain if you're going to abstract everything away to a centralized ledger?

I guess I'll echo GP and ask what is the problem that requires features specific to blockchain to solve some of these problems? I've followed eth since 2012 or so and I've only seen increased centralization in it and similar projects.

Telling investors you need to invent something new called an "L2 protocol" lets you ask for more money.

It might cause you to fix the insurmountable problems with a blockchain, insofar as it's just moving back to regular databases. But also it's now just something else you have to trust.