The way I like to explain it is that Bitcoin is collectively giving away about $1.5 billion a month in prize money to miners. That's the root of the problem. Miners will spend up to $1.5 billion a month on electricity (mainly) and their other expenses. Currently it's about 10x Google's electricity usage.
Total revenue is proportional to the block reward (an algorithmic parameter) and Bitcoin price. So to fix this, either they could reduce the block reward ahead of schedule (won't happen) or something could happen to crash Bitcoin's price. More transaction volume would actually be better if it were due to a fire sale.
Taxing holdings of Bitcoin would be a way to convince people and businesses to sell and discourage buying, at least for those that pay their taxes.
Or who knows, maybe threatening to do this would result in enough consensus to accelerate the block reward halving schedule?
If you implement proof of stake, for example (where the scarce resource becomes the currency of the network itself, which is a lot trickier to implement and especially quite hard to bootstrap from nothing because in the early days of a currency there's going to be only a few people owning it), then there's no need for the reward process to be a lottery (though it still can be).
The proof of stake proposals don’t seem to work that way. They seem to be more a matter of stakeholders having machines that vote about who is trustworthy.
You need something to be the limiting factor on the amount that you can try to win such a lottery. With proof of work, it's electricity. (assuming everyone is about as efficient as each other) Early on, they talked about it being proportional to cpu-time. That has become more abstract since.
I haven't heard of any cryptos that require you to risk your coins to mine. Sounds like a neat gimmick, but hard to implement.
There is already massive interest in finding more efficient ways to solve the problem. It's a difficult problem to solve (as Elon Musk will likely learn in the coming year).
The cost of proof of work is paid in two forms. The first is in the form of inflation. Merely holding onto Bitcoin means eating dilution as new Bitcoins are printed to fund the mining ecosystem. And the second cost is in terms of transaction fees. At times, it can cost $20+ to send a transaction over the Bitcoin network, and a more efficient / higher throughput system could significantly reduce these fees.
The users of Bitcoin are already themselves naturally incentivized to look for better solutions, even completely absent any care about the environment.
More enlightened self-interest would be to want to see the price go up without using more electricity, but this is secondary to the primary incentive.
To get action from enlightened self-interest, there needs to be leadership for making changes. (Ethereum seems to have this?)
Knowledge preceedes enlightenment. For example, Git is also green-friendly, and shares a virtually identical consensus model to pure PoS consensus designs. It’s just that “green-friendliness” wasn’t a design goal of cryptocurrency — creating a lasting store of value sans institutions, exchangeable pseudonymously over the internet, was.
As it were, pure PoS consensus suffers from the misfeature of not even having a quantitative fork ranking protocol — i.e. it lacks a way to objectively compare the truthfulness of divergent blockchains [1]. In PoW networks, cumulative hashing power can be simply and easily mathematically compared to produce cold, quantitative consensus. In PoS networks, there is no hashing power, and human intervention is required — just like Git.
Which begs the question, why do PoS networks even need a blockchain? I think we all know the reason for that, though — Git doesn’t have a hype train. Git just isn’t good enough at separating the general public from their hard-earned money.
Git was designed for cooperative work, and critically depends on trust (commit access).
Merkle trees are easy, scalable schemes of trustless consensus, less so :(
This would imply the users of Bitcoin use it for anything more than speculative investment.
Something that also seems to get lost in the mix frequently is the fact that bitcoin owners themselves are the ones paying for the electricity. The security benefit comes at a cost - manifested as inflation (not just transaction fees) - and that cost is borne by all of the owners of Bitcoin. To that extent, the participants in Bitcoin are actually incentivized to use more efficient systems, because it means the inflation cost of being a participant is lower.
That they don't use cheaper systems indicates (though there are plenty of confounding factors) that they see value in paying the larger cost of holding Bitcoin.
And Bitcoin users are only paying the fraction of the cost paid for energy. But that ignores negative externalities like the pollution that's the subject of the article.
Then you have to define value. I don't think it's transactions. Any random bank can make transactions. People can transact with cash.
Cryptocurrencies create value by providing other properties such as decentralization, trustlessness, privacy. For example, Monero enables private and untraceable transactions and for privacy enthusiasts it has essentially infinite value.
With this definition of value, we can see how bitcoin isn't providing much. By now it's centralized, has no privacy, has high fees... It sucks basically. The only reason it's still the number one currency seems to be inertia. It's the most frustrating thing about this market honestly. Not only is bitcoin still king but it drags down better coins when its value plummets.
If you wanted to know how much people are willing to pay for decentralisation, trustlessness and privacy (most certainly not an infinite amount of money) you could also do that, but that's answering a different question.
Why transact via cryptocurrencies instead of other systems? Because of these other desirable properties. They're the value cryptocurrencies add to the world and the reason why people choose to them despite the inefficiency. People transact in Ethereum because they want to run smart contracts. People transact in Monero because they want privacy.
> So it absolutely makes perfect sense to measure how efficient these systems are processing transactions, and the way to measure that is by looking at how much it costs to process a transaction with each of these systems.
I don't think so. People measure this in order to compare with traditional banking systems which are centralized, require trust in institutions and affords no privacy. This is not a fair comparison.
This is a big assumption, and almost certainly untrue.
That's an ok general rule, but it breaks down in a lot of circumstances. Look at scams, for example. People pay quite a lot for them, but at a societal level, they have strong negative value.
Interesting.
> For example, Monero enables private and untraceable transactions [...]
So so...
> Cryptocurrencies create value by providing other properties such as decentralization, trustlessness, privacy.
Let's add "a largely unregulated gambling area where people with too much money can bet their excess money on being among those who pull out before the music eventually stops" to that list, shall we?
That's also where your frustration comes from: Bitcoin is perfectly fine for this kind of "value proposition", hence it stays King of the Hill.
What I mean is people can transact via traditional systems just fine and they choose cryptocurrency because of other qualities.
> largely unregulated gambling area
The speculation is a separate thing from the coins themselves. People also speculate with fiat currencies, stocks, real estate.
> That's also where your frustration comes from: Bitcoin is perfectly fine for this kind of "value proposition", hence it stays King of the Hill.
Yes. What I don't understand is how it drags the whole market along with it no matter where it goes.
No, speculation is the primary use case for all of these coins right now.
And crypto currency has some advantages over the other speculation options you mentioned:
- more people can participate because of them being unregulated (compare: forex trading, which has much higher barriers of entry)
- it is rather easy to not pay taxes on crypto gains in many countries, especially if you're staying below certain limits
And the most important of them all: the crypto cake is still expanding and has been expanding for over ten years now. That is the crucial difference between classic forex trading and crypto - in forex, the comparative value of different currencies usually stays within a certain range. It goes up and it goes down. In crypto, the crypto just has been getting more and more valuable over the long term. It goes up, and up, and down, and further up than before again, and then some more up.
As long as this continues, any gambler betting on crypto has an edge and will win more than he loses. Of course it can't continue like this forever, but that's easy to ignore when looking at a ten year history of unbelievable gains.
That would be like saying "you need light to your math homework so we will measure how many math problems you can do per lumen of light".
The throughput of transactions is not coupled to the amount of electricity used to mine. They are two elements in the system, that doesn't mean they move proportionally in any way.
So fine, "kWh per available transaction-second" may be a more apt moniker however that's strictly what I've been talking about. Since 100% of electricity usage is attributable to mutations, this is a completely fair quantization. As blocks have been overwhelmingly full for years, the point on discrepancy is pretty much moot.
And further, the point is a dump truck is totally unsuitable for use as a school bus.
You are more than welcome to measure Wh of electricity used per math question in regards to the lighting setup. You'll find that number small and irrelevant, however. On the other hand if it requires a few thousand kWh worth of light to do a math question, you're well within your rights to ask (a) why you're doing math questions in what must be a stadium setting and (b) whether there's a more efficient lighting solution.
Which are exactly the questions we're asking here.
So, respectfully, that's exactly how logic works.
Bitcoin's electricity use it because of its sustained price. If the miners could not consistently sell for high prices the overall electricity use would go down. Price/electricity usage going up or down does not affect throughput over the long term.
Bitcoin's throughput is from being artificially crippled to sell a second layer. If the maximum block size was raised, it would have more throughput. The max block size does not change the amount of electricity used from the block reward.
The same way that you would not measure your internet connection on mb per kw/hr since the energy used by 10mbs service is going to be the same as 1000gbs service. The two things are barely correlated.
Energy per transaction is barely correlated. I know it is simple and easy to look at two numbers and pretend they depend on each other, but that isn't the dynamic here, which is why everything you are saying is nonsense.
> ... which is why everything you are saying is nonsense.
You're being rude again. Once again this is against the site terms. I'm happy to engage in any civil conversation with you that doesn't involve name-calling, as I'm sure you've seen a number of times. This thread, however is closed.
You may wanna chat with your mother for a second opinion ;)
> No one is complaining and no one said "haters", these are things you hallucinated.
I didn't say you said 'haters' that's a stylized/flourished way of saying an opposing viewpoint. That's why it's in quotes. If anything it was self-deprecating/deferential with respect to my biases. Jumping in with "that's not how logic works" is certainly a valid thing to call out as whining or complaining because again it's not particularly civil. Off the basis of your misinterpretation you once again accuse your counter-party of 'hallucinating.'
As I said this thread is over and should in the future you decide to have a civil conversation I'll happily engage.
You seem happy to engage on how something is being said, but not so much on what is being said.
Of course, because I'm here to have an interesting conversation about an idea on its own merits, not be accused of hallucinating when you misinterpret something I've said. If you stop attacking the speaker, assumed some good faith, and have a civil conversation I suspect you'd have a better time.
Normally I have much more tolerance for shenanigans but you've demonstrated a pattern in our interactions.
We know that. This is not an argument against cost per transaction being an adequate measure of efficiency for a transaction processing system.
Not only that, but not all cryptocurrencies are even trying to optimize for this. I would say that bitcoin is actually in the opposite direction and trying to be as expensive per transaction for both the miners and users as possible.
Not for a general audience. The purpose of a currency is to enable transactions. Internally, I get why people involved might fuss about the metric. But from the societal perspective it's reasonable to ask, "What does it cost per unit of value created?"