Hacker News Reader

Comment by Lramseyer | Hacker News Reader

I'm surprised that nobody's mentioned it here, but this crypto boom has sent demand for mining hardware through the roof, which has contributed in no small part to the chip shortage. Miners are buying up ASICs, GPUs and FPGAs at a premium.

In the short term, crashing the crypto market can ease demand for semiconductors, but in the long term, getting cryptos to move to Proof of Steak will hopefully ease it up for good (But don't quote me on that!)

It's no secret that one of their biggest supply chain constraints is chips right now (aside from batteries.) So it's not all that ridiculous to imagine them pulling this kind of publicity stunt off so that they can actually produce and sell more cars. And if they can profit off of it at the end of the day, even better!

View on HN

> Proof of Steak

There’s so much scope for word play humour here that I’m paralysed by choice. Vegetarianism, steak and chips…

Such humor is rare, usually put out to pasture. But this joke is well done. Hopefully it ages well.

Yes, it’s a rare medium for conveying information.

> Decentralized networks are a rare medium well-done.

https://steak.network/

It's obvious that steak needs chips, hence proof of steak will inevitably lead to a chip shortage.

and we all thought the proof was in the pudding

Proof of work keeps the network fair. Non mining nodes have the final say. Proof of stake lets players with the most money have the most say.

>Proof of stake lets players with the most money have the most say.

Okay, and what do you think the mining equipment is acquired with exactly?

It's the non-mining nodes that verify transactions and have the final say, and those can be run on home devices by regular people. That's the difference.

Non mining nodes do diddly squat.

Nonsense.

If miners's new block isnt accepted by the network - they've just lost millions.

Rules are enforced by block verification of the nodes that accept - or refuse to accept mined blocks.

It's amazing that in 2021 people still do not understand the basics of it.

Non mining nodes are programmed to follow the longest chain. That's so they are: readers. They are not writers. They act like slave nodes in a database. They have insignificant power in the grand scheme of things.

Longest _valid_ chain. Invalid blocks are just ignored.

That’s what miners do as well.

Yes. I just wanted to emphasize that in PoW coins, a rogue miner cannot force the whole network to accept blocks because all network participants (even passive non-miners) will just ignore the "invalid" block.

No, but if all the miners fork from all the non mining nodes, the chain with all the miners will be more secure. If a minority of miners side with the non mining nodes, that chain will be more vulnerable to disruption.

Would you hire a security company again, if they refused to protect you when the push comes to shove?

That's all miners are. Mercenaries hired for 10 minutes at a time.

Yeah exactly.

Not sure if you realize the implication of what happens if you arm a bunch of mercenaries and then fire them and turn your backs on them.

This has already played out.

The breakaway hostile fork is now worth 2 percent of the original work.

So..nothing?

I’m pretty sure that shows that miners follow the money. They will mine whatever chain where they can sell the block reward to the highest bidder.

Anyway I don’t think reality is as simple as your last comment suggests. There was a lot going on at the time including a lot of misinformation.

Miners don't really have much say, they cannot impose rules on the network, unlike PoS stakers.

PoS malicious stakers lose nothing in case of a split. Miners should they attempt a 51% attack will lose everything.

Not even in the same category.

>PoS malicious stakers lose nothing in case of a split

You can only support one side of a fork in PoS and if you try to play both sides anyone can form a proof showing you mined both forks and your stake gets slashed. This is the Nothing At Stake problem and well understood which makes your comment pretty confusing.

how do you objectively find out who is the attacker, so that you do not slash the victim on accident?

Can you give an example of what you mean?

There are no attackers or victims in the scenario of the chain forking and a block producer gooing out of their way to mine both chains with the same stake.

the chain splits in two.

how do you know which one to continue staking on to avoid getting slashed?

By default your validator software picks the unmodified chain unless you install a special update that lets it use the newly forked chain. If you do nothing you continue mining just one chain (the old one) and nothing is lost. If you validate with both then you are penalized. That can't really be done on accident.

Proof of work also boils the planet. It needs to be banned, yesterday.

Just put a carbon tax on everything with environmental side effects and let the market sort it out, otherwise who decides what get's banned?

For example, I think cars contribute very significantly to global warming and use lots of fossil fuels, do I get to say, let's ban all cars and force everyone to use public transport? No!

Bans of future sales of ICE vehicles are a thing, a lot of countries either have them or are proposing them.

Gonna have lot of exemptions. Long haul and industrial/farm/mining equipment cannot always run on batteries.

We are going to have ICE's for a long time.

Mindless bans like that, if overzealous, will just push even more of industry to Asia/China, where most of pollution happens today anyway.

I wouldn't be surprised if China funds quite a few green causes, big boost for their industries, and they get to just ignore all emissions rules.

I agree that there will likely be exceptions, but long-haul isn't one of them. The trucking industry is already subject to a lot of regulations around mandatory rest stops, so there are well-established break points along every major long-haul route.

This has lead to things like [1] where it is significantly more economical for long-haulers to electrify their existing fleet than continue using diesel fuel. I suspect long-haul will reach 90% electrification long before last-mile trucking and private-use automotives.

That said, I imagine you are spot on regarding farming - and pushing industry to less restrictive jurisdictions.

[1]: https://www.theguardian.com/australia-news/2021/apr/29/swap-...

I'm willing to bet that Bitcoin mining will be 100% renewable (compared to 40-70% depending on whose numbers you look at today) before ICE vehicles are all gone. I'm basing this on the fact that miners are incentivized to always seek the cheapest source of energy and that renewable/stranded energy is only going to become cheaper over time. At one point, hydro/wind/nuclear will become much more cheaper than coal/fossil fuels, it's inevitable.

As long as there are fossil fuels being used for the production of electricity, either non-clean energy or a mix will be cheaper as a purchaser (regardless of production costs).

All the electricity gets mixed together in the power network anyway, so if you buy 'renewable' electricity you are effectively just paying a premium to say yours came from the pool of renewable electricity. If you increase the power load on the network though, and in-aggregate the power requirement increases, chances are that increase still comes from fossil fuels.

I think we will be waiting a long time until global electricity is 100% renewable, although I hope it happens quick and it's good that you are more optimistic than me :)

If bitcoin mining is 100% renewable, that is 100% of renewable energy being destroyed instead of being put to use displacing other, more polluting energy.

It doesn't matter if bitcoin mining specifically uses renewable energy. What matters is used in total by everyone. And bitcoin mining will always make things worse, no matter what kind of energy it uses.

if renewables can't power a computer network - sounds like it'll never power the world.

What.

So do dishwashers and hot tubs, hypocrite.

The provide actual value and usefulness.

Bitcoin does three transactions per second while using the energy of an entire country. The bitcoin network could be run on a single PC from 2004, but it uses, again, an entire country's worth of energy.

There is nothing in the entire history of humanity that has been as obscenely inefficient and wasteful as bitcoin.

Banning it is a no-brainer. The lowest hanging of low-hanging fruit for environmental improvement.

Ask Argentines, Zimbabweans, Cypriots, Greeks, Venezuelans, Turks, Lebanese and a few other billions around the world who are not privileged enough to have a stable monetary environment why they choose Bitcoin.

It has little to do with tps.

They do not have any real property rights, unable to access world's financial market, unable to save, unable to better their lives by accumulating wealth, and whatever little wealth they are able to accumulate is confiscated on the regular by the oppressive regimes.

Of course, your dishwasher is much more important than their human rights.

They don't choose bitcoin, dude. That is pure propaganda, spinning a handful of people doing it into some big movement. It's not real, it's not happening.

"not real"? Do you have any data to back this up?

Penetration of crypto in Turkey was 18% (page 9) in 2018, likely a lot higher now. Expected to own in the future 45% (page 10): https://think.ing.com/uploads/reports/ING_International_Surv...

One in five own it, half expect to own in the future.

Can you elaborate what you mean by “the most say”? The nodes aren’t voting on political proposals. So long as one entity doesn’t control >50% of the validating nodes, it doesn’t really matter, does it?

The nodes aren’t voting on political proposals.

In the Etherium world, which has more changes, they do. Bitcoin, not so much.

I suppose they are voting on fork chains in that sense, true.

From https://vitalik.ca/general/2020/09/11/coordination.html "A large mining pool publicly showing how they have internally distributed their nodes and network dependencies doesn't do much to calm community members scared of mining centralization. And pictures like these, showing 90% of Bitcoin hashpower at the time being capable of showing up to the same conference panel, do quite a bit to scare people".

If you follow the link, you'll see a picture of 7 people that controlled 90% of Bitcoin hashpower at the time. Remind me how that's more fair to an arbitrary holder of Bitcoin, than say, letting the holders have the power with staking?

Or, for that matter, leaving fiscal matters in the hands of a central bank with a board of governors appointed by democratically elected officials?

Honestly, I don't see how the status quo in crypto actually improves on that at all (unless you are in the ransomware business, of course. Then it makes sense.)

You could try auditing central banks, see how long you can remain alive. Bitcoin is an open ledger that can be audited by anyone. That alone is a significant improvement.

It is also the closest thing we have in practice to the "Ideal money" proposal by Nash, that Nash spent 20 years giving lectures about around the world.

Nash's notes and slides on that: http://web.math.princeton.edu/jfnj/texts_and_graphics/Main.C...

The improvement of status quo is quite obvious. Bitcoin is a potential solution to the Triffin's dilemma.

This is something of the scale that can alter the course of history and impact every country on Earth.

> Or, for that matter, leaving fiscal matters in the hands of a central bank with a board of governors appointed by democratically elected officials?

Money shouldn't be democratic. I'm not OK with a majority deciding that they're entitled to my money.

More generally, private property is a good thing. I don't want majority vote to decide whether I can keep the house or car that I own.

How would you pay for infrastructure?

I see nothing wrong with taxes. I just don’t think a monetary unit managed by politicians works in the long run.

But the power to set monetary policy is just a restricted version of the power to tax. Creating money is effectively a tax on savings. So why are you OK with taxes in general, but not this very restricted tax?

> Creating money is effectively a tax on savings.

No, it’s much more than that. For example, by creating money, and using it to buy bonds, you can push down the rate of interest. This has nothing to do with taxes. It’s an interference with market coordination, because interest rates play a central part in this.

For example, let’s say you have a business worth one billion dollars, and you earn 50 million dollars in profit per year (a 5% yield per annum). By setting the Federal Funds rate at 6%, the government can make it profitable to sell this business and deposit the proceeds in a bank account, hereby earning one percentage point more in yearly returns.

The government shouldn’t have the power to halt economic activity in this manner, because it amounts to central planning, ie. communism. Which has been proven to be a disaster.

> by creating money, and using it to buy bonds, you can push down the rate of interest. This has nothing to do with taxes

Of course it does. It's all just wealth redistribution. Lowering interest rates redistributes wealth from lenders to borrowers, i.e. from savers to debtors.

> By setting the Federal Funds rate at 6%, the government can make it profitable to sell this business

The government can do exactly the same thing with tax policy. The only difference is that tax policy is more flexible (because it has many degrees of freedom rather than just one) and hence more powerful. The power to control the money supply is a proper subset of the power to tax. You can't support the latter but not the former and remain logically consistent.

In addition to miners, running actual Bitcoin nodes keeps the network safe, that's why it's very important for it to be very cheap to run so that many people/volunteers can do it.

They have very little power.

All they can do is refuse to work. Anyone can do that. They cannot force any new changes on you. Proven in forkwars with billions at stake.

PoS have not yet had such a test.

This difference is that in PoS, if anyone ever holds more than ⅓ of the total supply (as is the case with Ethereum's founders currently holding 65% of the overall issuance, for example), they can indefinitely control the chain through their re-org capability.

I don't have a reference but rumours have it Joe Lubin alone controls close to 10%, which would be close to 50 billion today.

With the usually apathetic voters/stakers, one such loud voice can sway a PoS network easily.

"With the usually apathetic voters/stakers"

What makes you think eth stakers are apathetic?

all experiments with onchain voting seem to have that problem, just like with voting in real life.

read up on the DAO exploit and carbon vote. Even in a drastic situation turnout was low.

It's a permissionless industry and I suppose people can try different things, but one thing that sets apart Bitcoin from everything else - it is not a democracry.

Democracy was designed for cities of 50-60k in size, where only rich landowners got to vote and everyone else was basically peons. It doesn't scale and is not particularly great.

PoS is just another string of boneheaded attempts to make democracy great again...by marrying it with plutocracy? idk.

Would you care to vote with your measly stake if you know the insiders have over 50% anyway?

"the insiders have over 50% anyway?" This is false. Hard to take the rest of your premise seriously when they are not based on facts and sound more like maximalism.

That's a hypothetical conjecture.

In any distribution, the small number of players usually controls most of the wealth, as we've seen throughout the history.

What are the odds it will not be a handful people, coordinating their interest vs everyone else? That's inevitable.

There's ~100m eth so your numbers are way off. I have a sneaky suspicion you don't really know what you are talking about.

Like i said, it's just a rumour, but from a fairly informed source.

Do you have a better estimate?

Total amount of fundraising ETH did initially was about 20 mil? 1-2 mil sounds like a feasible early VC check to get 10%, not including games you can play later to increase your stack further.

10% with the current market cap would put him among the top 10 richest people in the world... I highly doubt these numbers are accurate.

Could you: 1. Point me to the source that says, " if anyone ever holds more than ⅓ of the total supply (as is the case with Ethereum's founders currently holding 65% of the overall issuance, for example), they can indefinitely control the chain through their re-org capability." 2. Point me to the addresses of Ethereum's founders? Presumably it's then 65% of 115 million eth.

Would honestly love to see the details.

Apparently Vitalik has 333,000 ETH https://www.forbes.com/sites/michaeldelcastillo/2021/01/11/b.... Which would be 0.29% of the 115 million current supply.

I guess Joe has the other 64.71%, or perhaps there are a whole bunch of other founders that I'm unaware of...

It’s probably for ethereum 2, which is PoS.

The total amount of ETH held by addresses doesn't change when going from ETH to ETH 2.0.

That’s cute.

You would have to crash the crypto market by 95% for years to have any affect on miners and ongoing investment into chips, which is a small small portion of chip demand.

Good luck imagining this is related.

Do you think this relates to low-power MCUs like ARM Cortex-M? These are getting hit hard as well.

No, high performance gpus/fpgas/asics are made at bleeding edge fabs, whereas stm32 and the like are made at slightly older fabs.

Currently for the highest end cortex-m micros you're looking at 28 or 32nm ICs (10 year old nodes) up to 180nm for the older, cheaper micros (20 year old nodes).

There has been a you shortage for a quite while in large part due to crypto. The microcontroller shortage only started happening after car companies shot themselves in the foot and decided to screw everyone else over.

Reply on news.ycombinator.com