Hacker News Reader

Comment by Uptrenda | Hacker News Reader

You can say this about anything that uses lots of electricity. Sure, removing it solves the immediate 'problem' but its also not any more practical than having everyone walk everywhere solves the immediate carbon impact of driving. The solution is to invest in more solar, nuclear, and energy storage solutions. Our Earth is already plenty rich in resources to scale up power requirements, and our sun provides near unlimited energy. The energy problem isn't unique to Bitcoin. Using a lot of resources is also the main benefit to the security behind proof-of-work because it makes building a duplicate ledger immensely impractical. The security of the chain actually increases the longer blocks are joined after an input which is pretty cool.

View on HN

> The solution is to invest in more solar, nuclear, and energy storage solutions.

> The energy problem isn't unique to Bitcoin.

The unique problem with Bitcoin and PoW in general is that its energy usage is inversely proportional to the cost of energy. If the total value of Bitcoins mined every day is $X, collectively Bitcoin miners will use just a little less than $X of power every day.

That means that any time we do improve our energy infrastructure, by increasing supply with more renewable sources or reducing demand via efficiency improvements elsewhere (all of which reduce energy prices) Bitcoin mining becomes much more profitable, miners expand and new miners appear, and PoW's energy usage goes up accordingly to use as much power as they can possibly afford (up to $X).

This isn't true for anything else - there's no other industry that so directly turns energy into money without limits. While there is a relationship to other energy usage, if energy drops to 1/100th of its current price, people don't start heating their houses 100x more, they don't start gaming on 100x more power-hungry computers, and they don't start taking 100x more flights than they do today, because in each case, there are significant other limits and marginal costs involved.

PoW meanwhile will absolutely increase its power usage by 100x, as fast as the network possibly can. This is why PoW's energy usage is a unique problem.

> This isn't true for anything else - there's no other industry that so directly turns energy into money without limits.

Aluminum refining maybe, but even then you are going to run out of ore to turn into aluminum. This really highlights how novel the mining economics are.

Refining more aluminum at a lower energy cost is a societal gain.

Using more cheaper energy to hit a higher difficulty to protect the same network isn't a societal gain in the same way.

I.e. aluminum still follows the normal supply-demand curve even if it is highly sensitive to energy prices

They both suffer from the negative energy externalities, but you get something more with higher aluminum production. You get more aluminum.

You make good points on how bitcoin could grow to use a larger share of generated energy even without increases in its exchange price, but don't forget Andy and Bill's law: "what Andy giveth, Bill taketh away".

More generally, the Rebound effect suggests other expansions will consume a sizeable portion of efficiency gains (or possibly exceed it). Video gamers will want higher res VR devices, they'll want richer worlds with better physics, animation, AI units and high fps high resolution graphics. AI software is absolutely guaranteed to use up any 100x gain in efficiency and more will be demanded of computers.

A backfire effect is predicted as possible if new capabilities once discarded as infeasible open up (ala aluminum smelting, constant use of AC) and when new economic activity generated by gains in efficiency results in a larger number of wealthier participants. Such effects from efficiency gains are fine so long as they're focused on energy and not say, polluting fuel use efficiency.

Buckminster Fuller proposed ephemeralization as the manner in which prosperity could increase given growing population and finite resources. This takes place because as material use decreases; information, computation and energy use intensity increase as more activity is displaced into them.

TL;DR; the comment above yours is absolutely wrong about having 100x more energy won't bring 100x more flights. If my car could expend 100x more energy, it would be easily able to fly, so everyone would probably start flying as their regular commute.

https://www.lynalden.com/bitcoin-energy/

The author's opinion on this topic is that energy arguments are basically bunk, but she doesn't mention the scaling you do. I'm unequipped to judge who's right.

Most energy arguments are little more than hot takes, speculating on what might happen in hypothetical worlds where the cost of energy behaves in a way that it literally has never behaved in human history. The efficiency of energy production, storage, and transmission has always been and will always be the fundamental anchor of worldwide economic growth.

PoW also requires ASICs, and as the recent semiconductor shortage illustrates there are definitely limits to how much supply that industry can crank out. Mining rigs depreciate and must be replaced eventually, so it’s not like the supply can just grow without bound either.

>You can say this about anything that uses lots of electricity.

Yes. You can.

Let's actually generalize it to anything with an outsized carbon footprint (e.g. commuter flights, car dependent housing development, &c).

We will almost certainly continue doing that stuff but it's incompatible with any vaguely desirable ecological outcome. PoW crypto is special in that it is both highly optional (it's not essential to the lifestyles of 99.999% of people) and has easy substitutes.

Why don't we just relay on the price energy to allocate its use kind of like we do with almost everything else? That's the point of prices. If energy is not priced correctly then the government should adjust the price through a tax. But let's drop the morality shtick every time cryptocurrencies come up.

We do tax lots of other kinds of energy, but you have the dilemma where crypto will use the least taxed, least cost energy anywhere in the world, dump the externalities into the atmosphere, and no one can tax energy everywhere to make them pay to clean it up. That's why you have things like coal fired electrical plants that used to no longer be competitive firing up at full capacity again to mine cryptocurrencies.

Crypto mining can't be curbed by carbon taxes in any energy market like normal electrical consumption.

You address this by taxing at the fiat on/off ramps. The ones with access to the deepest liquidity also happen to operate in jurisdictions that have functioning governments that are amenable to passing taxes on PoW. And if that doesn't work, you impose sanctions and tariffs on the countries that permit limitless PoW mining, up to and including black-holing their IP blocks.

It can be done. We banned CFCs after all.

You can ban CFCs in refrigerators/products that come through customs and monitor their emissions from space. The first is just impossible and the second probably impractical to pin on the carbon output of specific mining operations.

The taxes on PoW specifically interfacing with the financial system is a weak form of banning it and is definitely something that would work though, but what the gp seems to be arguing strongly against. They are asking why the energy price can't be updated to encompass the negative externalities instead of indirectly taxing it through the PoW's blockchain when it interfaces with the financial system.

I don't think this is right.

Carbon taxes on energy used to mine bitcoin will cut into profitability, leading to some proportion of the least-efficient miners turning off because they are no longer profitable.

How do you enforce the carbon taxes? If I mine somewhere off the books in a random part of the world, my block is not distinguishable from one that paid a carbon tax on its emissions.

Carbon taxes can very effectively reduce the amount of carbon-emitting energy used by bitcoin in jurisdictions it is enacted.

How to coordinate a global carbon tax seems like a much bigger problem than cryptocurrency. At 0.5% of global energy use, even if crypto mining was magically eliminated in an instant, 99.5% of the problem would still remain.

Also, feels weird to try to address this problem on the demand side instead of the supply side (which has much more direct influence). Machine Learning training is power intensive and location-insensitive, but it feels odd to worry about 'how to enforce carbon taxes on ML training'. "Which uses of energy are economically profitable" seems tangential to "how much carbon is produced by energy generation".

“Random parts of the world” are not cranking out the electricity required to secure the Bitcoin network. Your hypotheticals need better anchoring in reality.

I think what happened when china banned Bitcoin is instructive to what would happen if a nation attempted to increase the energy prices for mining in it's borders - much of it moved to the next cheapest option, like coal plants in Mongolia, however being banned in China did decrease the price and use overall. Which is basically my point, you would be playing wack-a-mole with policy across the whole world trying to work on it from the supply side and energy prices, it will just keep moving to the most enviromental-legislation-lagging internet connected place. However, legislating the interfaces between your country and the network do drop the overall use and demand for mining effectively.

That would just lead to all the mining happening in North Korea and Iran since they can't be sanctioned for defecting from climate accords.

> PoW crypto is special in that it is both highly optional (it's not essential to the lifestyles of 99.999% of people) and has easy substitutes.

I’d like to hear what other economic activities you consider “highly optional” and whether they are more or less deserving of economic resources than Bitcoin.

We're seeing the fruits of years of relentless climate narratives connecting all energy usage to pollution, and therefore all energy consumption (for any purpose) is considered inherently evil.

Does that mean everyone who plays video games after work is literally satan for consuming energy for mere entertainment?

Life itself consumes energy; does this mean the universe would be better off without us in it?

This is obviously a nihilistic viewpoint once you follow it to it's natural conclusion.

I expect the future will have a combination of both PoW and PoS due to fundamental trade-offs. So I got news for everyone. Proof-of-work exists and isn't going anywhere. You can't just yell at the clouds wishing we could make it illegal for "environmental reasons". PoW is not "incompatible with climate change mitigation", this is obviously a straw-man. There's plenty of energy to go around, and we have to solve the climate+energy crises either way. Not to mention Co2 emissions alone is just one small part of the overall climate concern.

I'm more interested in individuals and organizations willing to actually solve technical problems related to energy and climate, and choose to ignore those just yelling at politicians to solve it for them.

> Does that mean everyone who plays video games after work is literally satan for consuming energy for mere entertainment?

If that replaces them and the three friends they are playing with from each driving in their own car to the bowling alley 25 minutes away then it's a net carbon savings. It's all about the alternatives, and in the case of cryptocurrencies, a lot of people think there are comparable, much more energy efficient alternatives for everything you use crypto for legally. There really is no ridiculous conclusion here, it's about the alternatives.

Taking your analogy: that doesn't make going to bowling alley as an evil alternative to video games? We should all acknowledge that's a straw-man argument. But this is the kind of argument you hear against PoW. For all we know they are driving electric cars to the bowling alley. Does that make the scenario different? How's that different from PoW powered by solar?

It's a difference of degree. Going to the bowling alley in a gas car is certainly worse for the environment and if they were perfectly substitutable you probably shouldn't choose the bowling alley. They certainly aren't perfect substitutes.

Neither is PoW and PoS, but it's still a question whether those differences are meaningful and worth the tradeoff. The argument is not absurd, there are real costs to the PoW approach worth considering.

The PoW powered by solar is different but it's also a mythical beast. By design there would be no way to legislate the kind of energy the mining part uses because it's opaque to the users of the network. The only thing reasonably open to legislation by a country are the interfaces of the network with the countries financial system and by then it is impossible to determine the providence of the energy it used. It's also impractical to make enough energy where coal or other fossil fuels aren't economical to mine blocks in a popular PoW scheme.

> there are comparable, much more energy efficient alternatives for everything you use crypto for legally

No. The only alternative that is even remotely similar to Bitcoin is gold, and gold mining is horrendous to the environment. Proof of stake coins are not Bitcoin, they don’t provide the same value proposition, and if you don’t understand why then I’d encourage you to do some more reading (including the entirety of the article linked in this post).

Imo this is like arguing we should all have private helicopters instead of bicycles- PoW currencies should be criticized on environmental grounds if the thing they replace (a bunch of bank servers doing ach and whatnot) is an order of magnitude more electrically efficient.

Bitcoin isn't just a replacement for the centralized database: it also replaces the government which administrates the financial system, the courts which enforce rules, and the military which enforces dominance on the rest of the globe.

Strong disagree, (some of) those things are all still necessary to prevent me from coming to your house and hitting you with a wrench until you give me your private key; to prevent me from taking payment and then ghosting you; and innumerable other financial system problems we solve with our current system.

> You can say this about anything that uses lots of electricity

Sure, but it's especially important to say here because proof-of-work cryptocurrencies are novel, popular, and consume orders of magnitude more electricity per transaction than more common transaction processing systems.

Quantifying the energy per transaction of fiat is hard. Yes, changing a number in a centralized database is cheap, but the cost of maintaining trust in that centralized database is immense and has far-reaching tentacles.

Ex. How much of the energy budget of the US military should be included in the cost of a dollar denominated transaction?

Answer: it doesn't matter. If a relatively small number of people (say, 10m) stop using the US dollar, its military isn't going away. But if the same number of people stop using POW "currencies", the energy impact could drop drastically.

And if we were charging for underlying infrastructure costs, then Bitcoin would have to put in its share. Unlike paper/specie currencies, Bitcoin is entirely dependent on a functioning global internet. There are an awful lot of long-tail scenarios under which Bitcoin becomes effectively useless and therefore valueless.

> If a relatively small number of people (say, 10m) stop using the US dollar,

This is a very weak straw man. The majority of people on the planet interact with the dollar economy, either directly or indirectly, because of its extreme dominance. Of course a ~0.3% reduction in users isn't going to change much.

Dollar hegemony is fairly binary: either it's the dominant world currency or it isn't. If hegemony falls, the US military will have to reduce in size, because it can no longer be funded by being the gatekeeper of the global economy.

Interesting speculation, but all wildly off the point. Which is that ending POW "currencies" would have a significant environmental benefit with an amount of inconvenience somewhere between marginal and net negative.

It would not be a "significant environmental benefit". POW mining is ~0.5% of global energy usage. If it was magically eliminated, that 'progress' would be eliminated in a year as growth in the more urgently-relevant sectors continues marching on.

It seems you don't see the value of POW cryptocurrencies at all. The attitude that "bitcoin is worthless, so any energy spent on it is wasteful" has been parroted a lot on HN recently, but... scoreboard: the market disagrees. I wish more bitcoin skeptics kept some intellectual curiosity around it rather than just writing it off as "I don't get it: its worthless" - there are reasons people are paying tens of thousands of dollars for a bitcoin, even if you don't understand them.

I think these attitudes are indicative of cognitive dissonance in individuals who dismissed Bitcoin a long time ago and don't want to recognize their own regret for discounting it: if they didn't dismiss it their net worth would probably be higher today.

Redirecting 0.5% of global energy to productive use is very clearly significant. That's the whole-country level of energy usage.

Markets also value Ponzi schemes highly. So what? Markets can be a useful guide to value in certain circumstances, but it's correlative, not probative.

I've had plenty of intellectual curiosity around Bitcoin, blockchains, etc. For a while I was hoping something good would come out of it. But it's been 10 years and there's been almost no economically positive use demonstrated. It's great for certain kinds of light financial crime (ransomware, money laundering, etc) and it's great for creating the sort of unregulated markets in which scammers, fraudsters, and other economic parasites thrive.

I'm personally very happy never to have put a nickel into cryptocurrencies. Could I have made a bunch of money? Maybe! Would have it been through creating value? No, not at all. It's a "greater fool" game, and I don't play those. I haven't since I quit working for financial traders in the 1990s, as I don't think it's moral to get paid when no value is created.

Bitcoin is a hard currency first; it includes a mechanism for transactions because that is a hard requirement for anything that functions as a currency but the network is optimized for security not throughput. If my only problem is paying people and getting paid, then Visa/ACH works just fine. If my problem is lack of confidence in my nation’s money supply (and this is a very real problem in quite a few countries) then Bitcoin is unquestionably the most convenient alternative.

I don't buy it. Bitcoin has much higher volatility than other currencies and much higher costs to use it. If your problem is lack of confidence in your nation's money supply, you're better off getting some more stable currency, like any one of these: https://fxssi.com/top-10-world-most-stable-currencies

But maybe the real answer is to just have trusted authorities? So much of civilization is possible because of trust, I don’t think we should throw that away.

You need trusted authorities anyway. Anyone may show at your home and steal your crypto currency at gun point, you may pay for items that are not delivered, your employer may not pay you. Authorities and the rule of law are fully needed for any functional society. Crypto currencies as an anarchist dream are impossible as a very complex society is needed to produce them, and that society requires to be functional.

If someone wants anarchism they should simplify and use gold or some other easier to carry and exchange.

> Anyone may show at your home and steal your crypto currency at gun point, you may pay for items that are not delivered, your employer may not pay you. Authorities and the rule of law are fully needed for any functional society.

In a crypto anarchist’s dream world, the employee’s income claimable token could be resold, purchased by a loan shark, and the loan shark could provide another permission-less contract where anybody can upload a picture of the employer’s knees being broken, likely verified on chain by some other pain oracle.

In my experience, your described state of affairs is common in poor economies. Robberies and kidnapping ransoms are relatively frequent. Shootings are rare but still way too frequent. Workers often go unpaid and it's not unusual for wages to total less than transportation costs. Landlords are known to have poor memories about paid rent and it's not uncommon for deliveries to be stolen and containers disappear. There's a real risk of valuable goods ordered internationally "disappearing" in customs.

The countervailing factors are, the poor are mostly ignored (except for kidnappings of large groups of children) while the rich can afford private security at an equilibrium price point where betrayal is not rational. Although crime rates are high, most do not participate because people are good for the most part. The major drags are scams, unpaid workers, phone theft (a phone is more likely to be stolen than a laptop), equipment sabotage (from wiring to rail tracks), political campaign related violence and most especially, theft, corruption and embezzlement by government officials.

In an environment of rampant corruption, minimizing the surface area exposed to trust is a rational preference. It might even be helpful in exiting a high corruption steady state.

You obviously live in a society where central authorities are, in fact, trustworthy enough to handle these things. A cursory glance at history—or at other societies even in the present day—shows that this is a shockingly rare state of affairs.

These trusted authorities of which you speak are no longer trusted.

More than I trust whatever the latest crypto get-rich-quick scheme is.

Neither are great, but various crypto scams have been more common than even bank fraud, plus they use more energy.

Why not choose one that's been around for a few years, then?

Five? Ten?

No one is suggesting that everyone must use Bitcoin, but please understand that over the past 100 years the majority of the world's currencies have collapsed one way or another. Not every country is the U.S. which has been able to use its military might to maintain the value of its currency. Russia, Mexico, India, many European nations prior to the Euro, most of South America, have all seen their currencies collapse at some point in the past 50 years or so, and in many cases within the past 30 years.

It's possible that the US dollar doesn't collapse at all in the next 50 years... and that would be great... but it's also possible that at some point in the next 50 years the US dollar ends up being devalued significantly in which case it isn't such a horrible idea to take some of your cash and store it in a globally accessible cryptocurrency in addition to other assets such as gold.

Okay so I kind of see where you're coming from, but I still strongly disagree with you.

You're saying that because world currencies collapse (your claim is the majority of them have within the past 100 years, which I don't have the data on, but let's just assume for the sake of argument that you're right), bitcoin is needed.

But bitcoin collapses not just every 30-100 years, it collapses every 0.5-5 years. [1]

How is that in any way better? Why are we spending all this energy on this?

[1] https://www.yahoo.com/now/7-biggest-bitcoin-crashes-history-...

> But bitcoin collapses not just every 30-100 years, it collapses every 0.5-5 years.

When someone says that a fiat currency collapsed, it means the value went to zero forever. Bitcoin hit a new all-time high price in USD just recently. Not the same usage of the word collapse.

This paper by the European Central Bank has a very in-depth discussion of the definition of currency collapse in section three [1], including three proposed definitions, two of which are given as mathematical formulae.

It specifically does not require permanence for any of the definitions, there's even an entire subsection on the varying degrees of persistence of various currency collapses.

Interestingly, from a first glance, it would also appear that most of the bitcoin collapses linked above fit the definitions in the paper, or iow, are just that, collapses.

[1] https://www.ecb.europa.eu/pub/pdf/scpwps/ecbwp1226.pdf

Thanks for the arcane technical note, I’m sure the OP who referred to currency collapse was using that term in the precise technical sense that the ECB uses it /s

You tried to derail my argument by casting doubt on my use of the term, I demonstrated at great length my use was in fact correct, you responded with snark.

Sometimes it's okay to admit when you're wrong.

Alden, the author, has another article on Bitcoin and energy use.

“Bitcoin's Energy Usage Isn't a Problem. Here's Why.”

https://www.lynalden.com/bitcoin-energy/

Also, fraud is absolutely everywhere. Everywhere.

Crypto is an attempt to solve a political problem, untrusted authorities, with a technology. This can never work, and without a political fight any crypto/web3 infrastructure will not catch on in a meaningful way unless it is bent to serve the interests of those authorities.

But it does and is working. Did you not even read the article this discussion is about?

The immediate carbon impact of driving should be solved by improving public transportation and weening people off individual transportation.

The immediate carbon impact of proof-of-work should be solved by banning proof-of-work. What is the problem any of this is solving?

It seems to me like your defense of proof-of-work is "it needs to be this way to work properly and it's not that big a deal", but why do we even need it to work? If it has a clear carbon cost and no perceivable benefit, then it seems obvious to me we should get rid of it.

> no perceivable benefit

I see no “perceivable” benefit in lots of things that other people choose to spend their time and money on. But at least I have the decency not to say it out loud.

Reply on news.ycombinator.com