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

Mainnet Merge Announcement

blog.ethereum.org

Post reply on HN
Print view View original

581–590 of 609 posts

Re: Mainnet Merge Announcement

#581

Post by immibis »

AlexandrB wrote: post #170
TakeBlaster16 wrote: post #100

Earlier quoted context omitted.

I suppose that is possible, but we're thinking of two different incentive structures. If 66% of miners censor their own blocks to comply with OFAC, then eventually the transaction will get in. If 66% of miners not only censor their own blocks, but also collude to continually reorg the chain, then obviously all bets are off, the project has failed, and the exchange rate is headed to 0. I think the former is infinitely…

> If 66% of miners not only censor their own blocks, but also collude to continually reorg the chain, then obviously all bets are off, the project has failed, and the exchange rate is headed to 0. Is it though? If crypto ever becomes a common currency people will still want to transact in it regardless of corruption/collusion among miners. Most are not going to stop using it for ideological reasons like "miners are a…

Many people do want to transact in rubles, but rubles are still a failed currency and will eventually be replaced with something else.

Re: Mainnet Merge Announcement

#582

Post by immibis »

spywaregorilla wrote: post #103
immibis wrote: post #65

Earlier quoted context omitted.

I don't know the specific rules for Ethereum, but in general PoS systems, anyone can slash. All you have to do is submit two different headers for the same block signed by the same validator. This proves the validator is cheating. The person who submits the proof may get to keep some of the validator's slashed money, which is a good incentive to run slashing nodes as this can amount to thousands of dollars per slash,…

> This proves the validator is cheating. To whom? Who gets to decide to slash the funds? I caught you cheating and can prove it. You don't give a shit and control 51% of the network. How do you get punished?

If you control 51% of the network, the network is fucked.

If you control 20% of the network, the other 80% is following the rules and colluding against you. (Even another 20% that are double-validating themselves are incentivized to slash you)

Re: Mainnet Merge Announcement

#583

Post by immibis »

Balladeer wrote: post #472
leppr wrote: post #159

Earlier quoted context omitted.

> In principle, inflationary currencies have a force pushing away from wealth centralization That depends entirely where the new money gets injected. If it gets injected at the top, like with our current fiat currencies, then there's nothing wealth-decentralizing about it, all the contrary. [1] Mineable cryptocurrencies inject the new money to whoever mines them, which could be anyone (less true with the current stat…

> Mineable cryptocurrencies inject the new money to whoever mines them, which could be anyone While true in theory, is it true in practice? Those with the most resources to mine will mine the most and thus receive the most, allowing them to mine more, and the cycle of centralization continues. That's true regardless of PoW or PoS, isn't it?

It is true. It's more true on PoS than PoW, since the mechanism in PoW is indirect. We can also consider how it works with traditional currencies, such as dollars, which has a hybrid PoS/??? system: you can receive resources by having the most resources (federal reserve deposits) and you can also receive them by performing services which the democracy deems useful (government spending), and to balance the numbers, resources are passively taken from people who spend their money but not from those who hoard it... maybe not the best design :)

Re: Mainnet Merge Announcement

#584

Post by RjQoLCOSwiIKfpm »

stablefusion wrote: post #564
RjQoLCOSwiIKfpm wrote: post #527

Earlier quoted context omitted.

Could you perhaps please answer this Ask HN thread I have posted: "Ask HN: Does the Ethereum foundation really not develop a post-Merge client?" https://news.ycombinator.com/item?id=32586172 TL;DR: To use the new Proof of Stake network, you'll need 2 pieces of software in parallel ("execution client" and "consensus client"). It seems only one of those is developed by the Ethereum foundation - implementations of the o…

Thr udeal scenario would be for ethereum foundation to keep becoming more and more irrelevant. Why would they take on 'more' responsibility if the goal is decentralisation? afaik, all clients are open source projects. And many are being funded by gitcoin grants now(crowdsourcing)

Because they earned $ 1 billion with the Ether presale [0], and if the community gives a billion dollars to a non-profit, the non-profit ought to deliver a usable product from that, not hide behind some handwaving of why that's the job of other people.

If they deliver a usable product that doesn't prevent anyone else from also doing so and thus achieving decentralization.

But it would mean that they actually do their job for the billion they were given.

[0] Calculated with today's Ether price, based on: https://www.coindesk.com/business/2022/04/19/ethereum-founda...

Re: Mainnet Merge Announcement

#585

Post by simiones »

tboyd47 wrote: post #261
remram wrote: post #180

Earlier quoted context omitted.

51% can do double-spend in Bitcoin, e.g. they can undo entire blocks of transactions and spend that money again. They can also prevent you from getting your transaction in ever . The 49% cannot prevent this because they cannot make a longer chain, that is the entire point of proof-of-work.

Adding the double-spend concept into it makes it even more silly. If you are changing the validation rules of Bitcoin to allow double-spend, then your block is rejected by the entire network and you must fork, even if you had 99% of the hashpower.

This thing actually happens ALL THE TIME in Bitcoin. That's why no one accepts a transaction (at least not one for any significant amount of BTC) as soon as it is in the latest block - they wait until the transaction is in the 3rd, 4th or older block.

The Bitcoin blockchain constantly "forks" for the latest 1-2 blocks. If you sell your car for BTC and hand over the car the moment you see the latest block contains the transaction where the BTC enters your wallet, you may see 1h later that the longest chain does NOT contain your transaction, and in fact your transaction is now invalid, because the buyer wallet has sent all the money somewhere else.

Re: Mainnet Merge Announcement

#586

Post by jbaczuk »

stale2002 wrote: post #546
jbaczuk wrote: post #516

Earlier quoted context omitted.

I understand. My point is there is no reason to reduce energy usage. There is reason to reduce "dirty" energy usage. There is an abundant supply of "clean" energy.

> My point is there is no reason to reduce energy usage. There is reason to reduce "dirty" energy usage Reducing energy usage is almost exactly the same thing as reducing dirty energy usage. This is because if there is X less total energy required, that means we can shutdown the X worst energy producing sources.

If that's how it actually worked, that'd be great. But I don't think you can count on energy demand to decrease or even stay the same over the long term.

Re: Mainnet Merge Announcement

#587

Post by leppr »

Balladeer wrote: post #472
leppr wrote: post #159

Earlier quoted context omitted.

> In principle, inflationary currencies have a force pushing away from wealth centralization That depends entirely where the new money gets injected. If it gets injected at the top, like with our current fiat currencies, then there's nothing wealth-decentralizing about it, all the contrary. [1] Mineable cryptocurrencies inject the new money to whoever mines them, which could be anyone (less true with the current stat…

> Mineable cryptocurrencies inject the new money to whoever mines them, which could be anyone While true in theory, is it true in practice? Those with the most resources to mine will mine the most and thus receive the most, allowing them to mine more, and the cycle of centralization continues. That's true regardless of PoW or PoS, isn't it?

I absolutely hate cryptocurrency shilling on HN so forgive me if the following appears like so, but this answers your question.

> Those with the most resources to mine will mine the most and thus receive the most, allowing them to mine more, and the cycle of centralization continues.

This is not a problem if the reward distribution is proportional to the investment from miners, which happens when the mining competition has low or inverse economies of scale, and a low barrier to entry.

In PoW and Bitcoin especially, the economies of scale are huge. "Industrial" miners are greatly more profitable than home miners, and ASICs require an upfront investment which comes with risk that smaller operations might not want to take. The result of this high barrier to entry and economies of scale are empirically verifiable by looking at the hash rate from block wins.

A correctly designed PoS system almost completely removes economies of scale (notice I said correctly designed, because many other PoS cryptocurrencies actually explicitly favor larger stakes), but might add some artificial barrier to entry, like Ethereum's 32 ETH minimum. The wealth centralization effect is also amplified in PoS by how directly acquiring the scarce resource required to participate in the competition influences the value of existing stakers' investment.

In Proof of Space consensus protocols (PoST) [1], the scarce resource is storage space. This can have minimal or even inverse economies of scales relative to money invested, as the running costs become almost as minimal as a PoS operation. And since the quality of the storage material is also irrelevant, what matters most is the cost per Tb of acquisition of storage space. The best savings are done on second hand drives, which makes investment in free time rather than money the profit differentiator. The free time to wealth ratio scales inversely, so smaller PoST miners can have a proportional advantage over larger ones. This decentralization effect is empirically verifiable with how much more individual mining nodes the biggest PoST cryptocurrency Chia currently has than Bitcoin and Ethereum, despite the comparatively very low absolute amount of rewards its consensus delivers to its miners.

[1]: full name: Proof of Space and Time, with Time referring to a competition required to get reliable timestamps on the blockchain happening in parallel to the competition over Space.

Re: Mainnet Merge Announcement

#588

Post by spywaregorilla »

zodiac wrote: post #561
spywaregorilla wrote: post #473

Earlier quoted context omitted.

> Yes, this was ad-hominem because it seems obvious the OP is not asking this in good faith. There's nothing here that was done in bad faith and I should not need to defend myself for asking a question in a tech forum. Assuming I'm not asking in good faith is bad faith. The reason I asked was a description pasted a short while ago: https://news.ycombinator.com/item?id=32535409 https://news.ycombinator.com/item?id=325…

> Which sure sounds like a decentralized process that is ultimately just centralized around the ETH foundation at the end of the day. The validators need to be decentralized (i.e. prevent "harmful collusion"), but the slashers don't need to be in the same way (as long as the validators are).

Nope. That's very inaccurate. If only the ETH foundation is allowed to slash, then it's just a centralized system with extra steps. Because only the ETH foundation is allowed to decide what is bad behavior.

It is not the case that only the ETH foundation can slash, which was the crux of my original question.

The fundamental misunderstanding people seem to be having is that slashers show objective proof of bad events. That's accurate. But validators are still free to agree or disagree with that objective proof of bad events. If you can create a means to prevent consensus on slashing for your overtly malicious behavior, your stake is only at risk if the entire community agrees to fork, which is probable for huge malicious behaviors but I'm not convinced is true for smaller things or circumstances where large actors have a vested interest in preferring the original chain.

Re: Mainnet Merge Announcement

#589

Post by zodiac »

spywaregorilla wrote: post #588
zodiac wrote: post #561

Earlier quoted context omitted.

> Which sure sounds like a decentralized process that is ultimately just centralized around the ETH foundation at the end of the day. The validators need to be decentralized (i.e. prevent "harmful collusion"), but the slashers don't need to be in the same way (as long as the validators are).

Nope. That's very inaccurate. If only the ETH foundation is allowed to slash, then it's just a centralized system with extra steps. Because only the ETH foundation is allowed to decide what is bad behavior. It is not the case that only the ETH foundation can slash, which was the crux of my original question. The fundamental misunderstanding people seem to be having is that slashers show objective proof of bad events.…

I agree with your 3rd, 4th, and 5th sentences, but not with your 1st and 2nd. Assuming a large number of independent validators, the Nash equilibrium is close to "there is no equivocation, hence no slashable evidence, hence no incentive to run a slasher node". (Remember that Nash equilibrium implicitly assumes that agents are not allowed to cooperate with each other). Suppose some fraction of the validators are deviating from this equilibrium by doing lots of equivocation, now there is an incentive to run a slasher node. So what this modelling suggests is that in the real world, we can have a small number of slashers and a large number of validators, and the security comes from the fact that anyone could be a slasher (it's OK that the number of people who actually are is small). But we cannot conclude that "it's OK to have a small number of validators, as long as anyone can be a validator".

Re: Mainnet Merge Announcement

#590

Post by marshray »

rglover wrote: post #553
marshray wrote: post #547

Earlier quoted context omitted.

> What are your standards for a good study, then? Prospective. For example, they decide in advance to test Ts for a randomly-selected group of men. This study reads as basically "We queried our healthcare system database and made up the rules as we went along." For example: * How did they choose the date ranges 1/2006 through 3/2019? * How did they decide to divide the dates into four ranges of 4, 3, 3, and 4 years e…

> Try to describe how you have considered and rejected other possible explanations for the observed phenomena. Sure. I use abductive logic/reasoning to connect different data points (this could be a study, social trend, major event, etc) and reason out how/why that outcome might be possible (and whether or not there are any gaps/inconsistencies in that reasoning). For example, our current economic situation: massive,…

> massive, unpayable federal debt, rising inflation, and a serious constriction of the middle class

I've been hearing people say all this since the 1970's. What's different now?

> If you factor in rising crime rates, well, what do people do when they're desperate for basic needs? Yep, crime.

Crime rates are not rising, they are low compared to the peak in the early 1990's. Whomever is telling you that is lying to you. Stop listening to them.

> So, in this one example, we've got the government's mismanagement of the economy leading to mental health problems, physical health problems, and relationship problems.

This is classic begging the question. By the second sentence you have concluded that the government spending is the source of all woes and use the rest of your post explaining again why you believe everything is the government's fault.

You are unable to consider other possible explanations.

Your reasoning begins and ends with the premise that the problem can only be government.

Post reply on HN