Elon Musk wants clean power. But Tesla's carrying Bitcoin's dirty baggage
reuters.com
reuters.com
First, the only thing that's energy-intensive about Bitcoin is block production. Making transactions, relaying them, validating them and storing them have negligible energy cost.
Second, nothing about the Bitcoin protocol requires block production to use fossil fuels. Just like with every other energy-intensive industry, the problem isn't the industry. The problem is the fossil fuel use.
Therefore, if you want to get mad about the high energy use of block production leading to environmental pollution, you should direct that anger at the appropriate target: miners who use fossil fuels to mine.
How do we fix this? The same way we fixed it in every other energy-intensive industry: through taxes and regulation. Let's get some laws passed to require miners in your area to use only renewable energy, and to require exchanges to impose a carbon tax on coins whose miners rely on fossil fuels to mine. Properly applied, these laws would make fossil-fuel mining unprofitable, which is exactly what we want.
Bitcoin and PoW aren't going anywhere at this point. So let's make sure it's continued existence doesn't make the world worse. Sound good?
I checked the bitcoin docs - but couldn't find much about energy use (or I'm just not very good at searching ;) )
Of course, the estimates of how much energy is used are always a bit suspect. I've seen articles like this that just don't quote any sources at all, and I've seen a number of them that all ultimately refer the same reddit comment from a decade ago where someone made an estimate of the per-hash electricity cost. They then scale that up to the current hash rate and use that as an authoritative figure.
Of course, nobody adds up all the electricity used heating and cooling all the bank branches in the world, or the cost of the looping animations on the screens of all the ATMs in the world, or the food eaten by all of the bank tellers in the world. Single-book accounting can make anything look bad.
But aside from that, the real story here is that the proof of work implementation allows the bitcoin system to spend money on something that makes certain kinds of attacks difficult or impossible. Other banking systems have to protect against those same attacks, but they do it by using trust between humans, exhaustive reporting and accounting requirements, and so on. As a result, those attacks are easier than in the bitcoin system, especially when international banks are involved. That big scandal in Germany where Wirecard told it's auditors that it had €1.9B in a bank in the Philippines comes to mind. It turns out that they forged the documents in an attempt to cover up their failing business. Forging a bitcoin transaction like that is impossible.
The energy spent on block production is something that, in theory, _could_ be regulated so as to reduce or cap the pollution it may create. We already do this with other industries, like banning CFCs from refrigerants. Why not do the same with mining?
But before we can have a serious conversation about this, the folks on this site need to understand that the problem is _using fossil fuels to mine blocks_. It makes no sense to talk about how much electricity a Bitcoin transaction consumes, as if that meant anything. Loading and rendering HN takes more electricity, and no one complains about that.
Even "clean" energy isn't free energy.
We need to consider both the costs on the supply side as well as what's driving the demand side for our energy consumption if we want sustainable systems.
The costs of running both our Internet as well as the Bitcoin blockchain are components of that, and if we aren't considering them, it's a mistake.
The energy costs of the Bitcoin blockchain will become highly significant very quickly if the network continues to grow at the pace that it has to date, even if the work is done by machines relying on energy sources multiple times cleaner than those that currently exist.
If we want to alleviate this, I suspect that we'll have to consider what the underlying social factors are that are driving the demand for such a network.
No one said it was. But some energy sources produce far less pollution than others, and are far safer. Also, everything I'm saying about regulating and taxing pollution from PoW applies to every single energy-intensive industry; PoW just happens to be the particular industry at hand.
> If we want to alleviate this, I suspect that we'll have to consider what the underlying social factors are that are driving the demand for such a network.
This is literally what regulation and taxation could address. If you tax cryptocurrency transactions based on how much pollution their miners create, the problem could fix itself -- miners would then be incentivized to use only clean energy to mine (and prove that they are doing so), since that's the way to stay competitive. Also, because literally every transaction is a public record, a government could (a) require KYC for you to trade crypto -- up to requiring you to register your address(es) -- and (b) send you a tax bill each year for the pollution caused by each block that included one of your transactions. Exchanges operating under its jurisdiction would only let you cash out once you paid that tax.
In addition, suitable regulation could require that no more than X% of a utility's electricity can go towards mining (e.g. only during off-peak hours; only if its excess; etc.).
You said "It makes no sense to talk about how much electricity a Bitcoin transaction consumes, as if that meant anything".
The only way for the amount of consumption to not matter is if there is a free or nearly free energy source.
For what it's worth, I agree with you somewhat that regulation and taxation could help, at least to the extent that such actions are enforceable - my issue is simply with your proposition that it is only the source of the energy that matters and not the amount, when in fact it is both.
Yes, but that's not saying that energy is free. I was saying that the energy used to create a Bitcoin transaction is negligible. It only takes a handful of ECDSA signing and cryptographic hash operations, as well as some serialization operations to produce a transaction. The act of validating a Bitcoin transaction only takes a handful of ECDSA verifying and cryptographic hash operations, as well as some deserialization operations (such as evaluating Bitcoin scripts). Loading HN takes more CPU cycles than this, but as I was saying here and elsewhere, it's quite often the case that people mistakenly believe that the high energy cost of producing a block must also mean that the act of producing and validating a Bitcoin transaction takes a lot of energy. This simply is not the case, per the above.
> my issue is simply with your proposition that it is only the source of the energy that matters and not the amount, when in fact it is both.
We're trying to control pollution, not energy output. They are linked, but we shouldn't penalize low-pollution high-yield power plants (like hydro-electric dams or wind farms).
My proposition considers how to effectively regulate miners (really, any high-energy industry) who can easily pack up operations and move between countries. To me, this means that it makes more sense to apply a universal carbon tax/tariff than it is to try and target individual industries. This is something we can actually get decent data for -- all we need to do is use satellites to monitor how much pollution gets emitted from each country or other political division, and apply a tax/tariff on its goods/services proportional to the pollution emitted per capita. Pollution data is something that already gets tracked; we just need to translate it into tax policy.
Life as we know it doesn't care about which industry creates which pollutant, nor does it care about how many gigawatt-hours we can create. Life as we know it only cares that the environment is not so polluted that it can't keep on living.
I agree with you.
> Some bitcoin proponents note, meanwhile, that the existing financial system with its millions of employees and computers in air-conditioned offices uses large amounts of energy too.
There's 0 details or comparisons in there.
On first order of approximation you can take how much a miner gains on average per transaction (mining+fees), which is probably much easier to calculate than the real cost of the other side (I guess one could check how much Western Union takes ?)
Sometimes people compare it with Visa/Mastercard costs, but it's unfair when you're talking about unallied countries, as those benefit from the discount coming from operating at scale in allied countries, and the bitcoin network cannot be consistently shut down without shutting down the Internet itself !
If for every 35 people who pay for their tesla in bitcoin, 1 is someone who wouldn't have bought an electric vehicle otherwise, there is still a net overall decrease in CO2 emissions.
Source?
The entire bitcoin mining network which processes all transactions does 120 Exahashes (120,000,000 terahashes) per second. Were all mining done on comparable asics this would be 4.56 GW to run the whole network, which in turn means electricity consumption would be 4.56 million kwh per hour. At the time of this writing bitcoin is averaging 14,814 transactions per hour, meaning the average electricity consumption per transaction is 308 kwh.
You can see the hash rates and power consumption for bitcoin miners at [0]. Hash rates for the network at [1]. Transaction rate at [2].
[0] https://www.asicminervalue.com/efficiency/sha-256
[1] https://www.blockchain.com/charts/hash-rate
[2] https://www.blockchain.com/charts/transactions-per-second
Wouldn't those windy German days when electric energy prices turn negative be very attractive to them ?
Or is bitcoin mining hardware still quickly getting obsolete ?
Anecdotes about a brother of their friend's cousin mining using GPUs will be invoked to argue people are mining in their basements using grid electricity and thus the Bitcoin network uses fossil fuels, even though mining with GPUs is impossible for many years now.
The benefits of Bitcoin will be handwaved away even though they criticize the "high cost" of transactions - who the hell uses these costly transactions if Bitcoin is so useless and can be replaced by cheaper and better methods?