BECI starts with calculating average mining revenues based on a 439-day (variable) moving average of the Bitcoin and Bitcoin Cash prices. Then BECI assumes a fixed 60% share of these revenues are spent on electricity costing $0.05/kWh. That is it. There is nothing sophisticated about his model. His first error is that 60% is not representative of current hardware; real-world data shows the lifetime average percentage is between 6.3% and 38.6%. His second error is that the averaging period is excessively variable and poorly justified: it has increased from 60 and 439 days and changed by twofold his electricity consumption figures.
For more info, see: http://blog.zorinaq.com/serious-faults-in-beci/
In reality, Bitcoin's consumption is only a third of the power generated by the Three Gorges Dam as of 4 months ago: http://blog.zorinaq.com/bitcoin-electricity-consumption/
Anyway, the discussion is a good one and I'm glad at least 2 people are concerned about this. This is exactly the kind of thing we should be truth-seeking on, as well as finding use cases... not if Wall Street is finally allowing people who don't know any better to keep fueling the beast.
[0] https://digiconomist.net/bitcoin-energy-consumption#validati... [1] https://digiconomist.net/re-serious-faults-in-beci
Alex of Digiconomist is very good at writing long verbose paragraphs that "seem" to support his model, but when we look at his actual data and logic, it's full of holes, flaws, and poor assumptions.
One of the things we both agree on is that in the future (1 year? 10 years?) the ratio will tend toward 40 to 60%. But that doesn't say anything about the ratio as of today.
Ethereum is a vibrant, growing ecosystem with many many ambitious projects that have or are nearing release (maker, golem, swarm, bat, you need to do some research because the list is extensive). the above statement is really ridiculous but unfortunately the kind of throw away comment that gets voted to the top of an hn thread. what a sad place this is!
Same for the others. Golem is a nice idea with a lot of potential, but as of right now, Azure, AWS, and GCE are where programmers go for rentable computer power. Maybe one day that place will be golem. Maybe not. Right now it is all just potential.
I think part of the parent's point is the potential of crypt-currencies is often touted, but so far little has been shown in actual results. How long until there are actual results?
That's beside the point, though. Knowing that Ethereum and Bitcoin consume more energy than Iceland and Switzerland respectively, why don't we turn them off until they're providing actual use? I get that markets need time to figure out if a product is useful and useless things get built all the time, but has that discovery process ever been done at this scale before? For instance, if proof-of-work ends up spending as much electricity as the United States and we still can't use it for anything meaningful, shouldn't we be concerned? I know "turn them off" sounds silly, but you can tax the miners or impose energy consumption caps until there is a smart contract out there that is arguably benefiting society.
If concern for humanity's wellbeing on this earth is a "throw away" concern to you, then maybe you can provide an argument for why everything's going to be OK? Have we fixed global warming and I just didn't get the memo?
Credit suisse/UBS main operations are in London.
There is no way the numbers come out in bitcoins favor here, the scale just isn't there.
If we can’t find any benefit that Bitcoin brings, then there should be a cost threshold where we start applying the brakes. In other words, how much energy are we willing to spend without a beneficial use case? With Visa and banks, we already know they serve hundreds of millions of people on earth—if not billions.
That's quite a statement. Did you mean without beneficial use cases that you personally know of? Example I posted somewhere in this thread: https://www.forbes.com/sites/ktorpey/2017/12/31/bitpesa-ceo-...
The hint at her anecdote not holding true in the general case comes right after: "While BitPesa is still happy with Bitcoin for now, Rossiello did note that it’s annoying that on-chain transactions have become much more expensive.
Rossiello went as far as to say there will be a mass exodus away from Bitcoin if a countermeasure for high fees is not deployed within the next twelve months; however, she pointed to the Lightning Network as a possible solution to this issue."
OK so the CEO of a Bitcoin payment company is annoyed that Bitcoin is increasingly worse than alternatives and she is hoping the Lightning Network will redeem it, otherwise there will be a "mass exodus".
That's hardly a convincing case for Bitcoin being the better alternative for payments and reads more like all other blockchain things: an experiment. Experiments are good as long as we're keeping track of their costs and being honest about the results.
It obviously is for the CEO who continues to use Bitcoin despite the annoyances, so I'm not really following your logical leaps, nor do I have the energy to chase you through your goalpost moving. You said there are no beneficial uses, and I provided you with one example.
You can say that I said "beneficial use cases" and show me that you can actually send money using Bitcoin. But that's like me showing you a 5-wheeled car and saying "see, you can get from A to B in a 5-wheeled car therefore a fifth wheel is beneficial". That's not sufficient and the CEO of BitPesa seems painfully aware of this.
Furthermore, if we can't find a stable use case for the blockchain and there are no more greater fools investing--the bubble collapses and all the ASICs end up in landfills--does society suddenly stop innovating in energy production?
Can you point me to any good literature on this? I don't understand the point.
The point is that the free market and algorithm finds an efficient equilibrium between what the users value it at (which is a reflection of its utility to them), what it costs to secure the network in relation to the reward paid for it. In these costs are electricity costs, for which miners will find the most efficient means such that they are profitable. The energy prices are determined by the economic actors producing it. The will charge what the market will bear such that they are profitable and competitive. All this is a wonderfully efficient system except for the fact that the externality you are worried about (pollution from energy production) is not accounted for anywhere in the system. In fact, it is almost incentivised, since it is likely that I can produce dirty power cheaper than I can clean power (although this is changing). If you did account for this, dirty power would become more expensive, miner using it would disappear or choose clean energy.
Hence, the problem is not "It uses too much energy! So wasteful!!!!" The problem is "We don't account for externalities!!!! So bad!!!"
"Another issue relates to the negative externalities arising in proof-of-work blockchains. First, as shown above, when choosing individually optimal computing capacity, miners fail to internalise the negative externality their investment generates for other miners by increasing difficulty. This implies that equilibrium capacity acquisition in proof-of-work mining is excessive. Second, proof-of-work mining generates greenhouse-effect negative externalities, whose order of magnitude is significant. As of January 2018, the electricity consumed for Bitcoin mining was equal to the electricity consumption of over 3,400,000 US households, with an average consumption per transaction of around 300 KWh. Pigovian taxation could curb overinvestment in mining, but it might also be difficult to put in place, given the international decentralisation of mining."
source: https://www.tse-fr.eu/sites/default/files/TSE/documents/doc/...
The “let the market find equilibrium” philosophy is reckless when we’re talking about global environmental concerns IMO.
Crypto mining is part of the externalities accounting problem, and for that matter one of the chief motivating factors behind crypto currency adoption has the idea of being able to avoid things like carbon taxes and energy quotas...
The chief motivating factor is a fair economic system.
By your logic anything energy intensive industry should be stopped because of the pollution it causes. Say goodbye to the Steel industry, Aluminium industry, Chemical industry, Textile Industry, Copper industry. https://www.ecofys.com/files/files/ecofys-fraunhoferisi-2015...
I guess this comes down to the fact you think it is more pragmatic, sensible, likely and beneficial to have Bitcoin miners stop mining. I think it is more pragmatic, sensible, likely and beneficial to have a price on carbon.
I'm going to go out on a limb and say that the steel industry, aluminium industry, chemical industry, textile industry and copper industry have done more for humanity than Bitcoin. Use less electricity though...
YMMV, but I think it is more pragmatic, sensible, likely and beneficial to stop shitting on people's lawns than to evangelise wider adoption of a system for producing more shit, funnelling it more efficiently to lawns less likely to attract the attention of shit regulators, whilst insisting the real problem is that the lack of a global shit tax. (I'd have used a less fecal analogy like hazardous waste and clandestine disposal systems but that's a little too close to stuff I'm actually working on making more difficult in the day job!)
!=
"citations for the idea that the theoretical point of Bitcoin is reducing government control of money and a large practical part of the BTC economy is regulatory evasion?!"
It's not entirely clear how to put a price on carbon, though. And, bitcoin's censorship-resistance (which of course I laud as an amazing and important innovation) might make it harder to force everyone to account for externalities.
Is someone doing serious research, writing, and thinking on the relationship between crypto-blockchain tech and the economics of environmental externalities?
An obvious problem, sure, but no easy solution.
You worked at a Bitcoin company and don't understand the value of proof of work? Being able to trustlessly settle shared stated across the entire globe is unproductive output? The globe spend billions of dollars each year blowing up mountains to sift for gold that then sit in central bank vaults to collect dust. At least bitcoin mining can absorb some of the curtailed renewable energy and make the payback on renewables faster.
If there's a total internet && computer && electricity apocalypse, a physical barter exchange with gold is more trustless than bitcoin. You can still transact bitcoin but there's trust involved.
Advantages of Bitcoin over gold:
No trusted third party (If you're moving sizable sums of gold, you're doing it with a trusted third party[TTP))
99.99226%[1] transaction uptime (Your TTP keep business hours)
No countries win the gold-geography lottery (Your country that has no gold reserves can still accumulate)
Instantly assay the value (gold takes time)
You can use latent renewable energy to mine bitcoin (the way that miners remain profitable is by using cheap hydro power that would be wasted if not consumed) No need to blow up mountains to sift for gold.
Bitcoin is programmable value (People in places without functioning judicial/commercial systems can create contracts with each other)
You cannot practically bring more bitcoin to market if the price increases (The price of gold dictates which gold fields are profitable to produce)
Most importantly, it separates money that citizens use and the onus of financial obligations that the oligopoly (Banker's bailout 2008) and political class (Endless war) place on citizens.
Moreover, you fail to account for human nature. I'd be shocked if bribes have not already passed hands, to allow for use of electricity for bitcoin mining near hydro power in remote areas.
Bitcoin as of now, is by volume a mechanism for letting privileged people in China shift their wealth (including Face) from resources in China to somewhere outside of China, with Bitcoin as a proxy. All the while ignoring externalities.
All the other rosy bitcoin has potential and so on, may be true, but please, look at what it is right now and say with a straight face that it's great. What is it, 90% of hash rate in China?
Edit: the bitcoin block reward is irrelevant. When more value is being parked in bitcoin, the transaction fees are going to go up to match the lost rewards. And people will pay.
1) Has been used for millennia. Bitcoin was invented ten years ago.
2) Accepted as valuable by most of human population. Bitcoin mostly as valuable within internet echo bubbles.
3) Although price fluctuates, it's more stable than that of bitcoin, thus better suited as store of value.
4) Has applications for industrial uses or for jewelry, guaranteeing that your gold retains at least a base value. Same cannot be said of prime number data structures.
5) Doesn't corrode and doesn't depend upon hardware or storage media to be working.
I'm not saying that people should hoard gold, just that bitcoin is a pretty poor substitute.
And, BTW:
> No countries win the gold-geography lottery (Your country that has no gold reserves can still accumulate)
Pop quiz: the population of Germany is roughly on par with the population of The Democratic Republic of Congo. Do the citizens of the two countries currently hold the same amount of bitcoin? If not, how come?
Price fluctuations are not what you care about with a store of value. What you care about is the elasticity of supply with respect to price. If the market price of gold increases, the production of it will increase (though to a lesser degree of other commodities). Bitcoin's supply curve is programmed with time as the only input. This means that no matter the change in price, supply cannot increase. This is a coup de grâce against every other existing store of value.
>4) Has applications for industrial uses or for jewelry, guaranteeing that your gold retains at least a base value. Same cannot be said of prime number data structures.
You don't want a store of value to have a myriad of other uses because those uses influence the price of the store of value. Say that 'Beautanium' becomes a more popular substance use for jewelry. Now Gold loses value. The only metric you care about in a store of value is the change in supply vs the available supply. This is Bitcoin's killer app, This is why it has value.
Yet the total amount of gold on this planet is finite, and as the millennia have passed, it has become increasingly expensive to dig it out of the ground. These are the same properties that people laud in bitcoin.
> Bitcoin's supply curve is programmed with time as the only input
As far as I am informed, bitcoin mining farms also tend to be connected to the power grid. May be just to keep the soft drink vending machine running. Dunno.
> This is a coup de grâce against every other existing store of value.
Limited supply is by no means unique to bitcoin. Gold and real estate share the same properties.
>> Bitcoin's supply curve is programmed with time as the only input
>As far as I am informed, bitcoin mining farms also tend to be connected to the power grid.
The electric grid has no influence on the supply curve of Bitcoin. Do you accept the fact that time is the only input for the supply function? It does not matter if there is one miner producing bitcoin or if bitcoin's price is $1 or $1MM dollars the same amount of bitcoins are produced.
>Limited supply is by no means unique to bitcoin.
Again seems like you completely misunderstand the difference between scarcity (limited supply) and controlled supply. There's a finite amount of water on the planet (because there's a finite amount of hydrogen and oxygen on Earth) too, does that mean water would make for good money? No because water is easily synthesized from other elements. It has a very multivariable supply function.
Difficulty readjustment mitigates long term influence of electriciry market on Bitcoin supply curve, but doesn't prevent the current prices from short-term influences
> Do you accept the fact that time is the only input for the supply function?
No.
> It does not matter if there is one miner producing bitcoin or if bitcoin's price is $1 or $1MM dollars the same amount of bitcoins are produced.
First, this isn't true over short terms, though difficulty adjustment means it should be approximately true with a steady state allocation of mining resources over the long term.
Second, though, supply isn't production caapcity, it's the mapping of price to number of units sold.
>No.
Do you accept that the following line is nominally linear? https://blockchain.info/charts/total-bitcoins?timespan=1year
If you do, then you must accept that time is the only input to the supply function.
If you don't, supply a source to show what variable you think is correlated to the change in bitcoin. The data indicates, time alone.
> If you do, then you must accept that time is the only input to the supply function.
Er, no, even assuming the omitted x-axis is linear, I don't. The other inputs being relatively constant or having variation which mostly cancelled out each other's effects over a one year period is indistinguishable, on such a chart, from time being the only input. (Plus, the implicit function to which time would hypothetically be the only input isn't a supply function, which is a mapping from price to quantity people are willing to sell.)
This is really interesting. Gold has physical uses for constructs that were developed during the industrial revolution. And as you say are well tested.
Ethereum or BCH, have information uses, for constructs that either have been developed in the last 20 years, or are being developed in this "information revolution" era.
I think we are about 20 years early to really see if a thing with physical properties will actually be more valuable than something with "informatic" properties. My bet is that the latter will win, given that materials sit on a lower level chain of complexity than information management.
Bitcoin corrodes?
You can encode your private keys in Gold if you want to.
And yeah, you can encode your private keys in gold, and I kind of think that everyone should do that, to keep future archaeologists puzzled :-P
It is difficult to compare the durability of gold and bitcoin. It does seem plausible, given the length of time we've been using it, that gold's value is completely related to its intrinsic properties and difficulty of mining.
Bitcoin can't possibly be valued on its intrinsic properties, because I can fork the entire technical edifice, start a new blockchain called it Bitcoin+1, and the fork will have no value vs the original.
Combined with a passphrase, it also means your money can't be stolen if your house is burgled. Gold would be a sitting duck in this case.
If you're using a computer, its hardware and software stack consist of one or more trusted third parties.
Although it's an interesting thought-experiment to compare the risks!
It means when governments want to borrow more money, their lenders actually need to have the liquidity available and can't just magic it out of thin air.
> You can use latent renewable energy to mine bitcoin (the way that miners remain profitable is by using cheap hydro power that would be wasted if not consumed) No need to blow up mountains to sift for gold.
... you do realize that this is a contradiction, right?
i mean yeah, its technically no longer a literal gold mine, but renewable energy sources where you can place power plants such as your mentioned hydro power aren't that abundant.
For what that's worth.
Some time ago I wrote an essay on proof of work and gold: https://medium.com/@zby/proof-of-work-8d8265def194
If something is worse than the problem currently being discussed, then The problem currently being discussed isn't that important at all. In order for the statement "A is not as bad as B," to suggest a fallacy there must be a fallacious conclusion such as: ignore A.[0]
I keep seeing these types of posts every few months and they always seem very poorly thought out and presented. Almost like people did the actual analysis but didnt like the results...
Don't you need to compare this with the energy usage of the industries bitcoin is supposed to replace (banks, money transmitters, credit cards)? Also take into account the environmental benefits of a deflationary currency (people saving instead of buying crap they dont need)?
What about running web pages that are >99% ads, tracking, and other unwanted content by kb? Also blocking all that unwanted content. Did anyone set out before hand to calculate the energy usage of that?
It just seems so disingenuous to be concerned about the energy usage of this one thing (and its only possible because it happens to be particularly easy to estimate, an advantage!) every time I see it.
The link I included does actually compare Bitcoin energy consumption to Visa by the way, and the difference is staggering.
I never assumed that, nor that it won't (like you just did). I just said "supposed to replace".
Here is an example. Someone came up with the number 1700 TWh per year for data centers, wifi, computers, phones, tvs, etc. I have no idea, so just use that for now: https://www.researchgate.net/publication/320225452_Total_Con...
Anyone familiar with whats on tv or the internet knows that 99% of all that is unnecessary, even unwanted, crap. That gives 1683 TWh being wasted each year on crap, yet people are only concerned about this 5%. I just want consistency so I don't feel like I'm being concern trolled.
EDIT:
Looking at the visa estimate I see right away it is underestimated by 40%:
>"In calendar year 2016, we used an estimated 674,922 gigajoules (GJ) of energy from Scope 1 and 2 sources [...] Of our total emissions, only 7% were from Scope 1 sources, with 54% from Scope 2 (primarily purchased electricity) and 39% from Scope 3 (mainly employee commercial air travel)."
https://usa.visa.com/dam/VCOM/download/corporate-responsibil...
So Visa alone is about 0.25 TWh per year.
Deflation seems bad from our current mindset of growth at all costs, and especially bad if you're one of the upper class who's investments return more than inflation, ensuring you'll keep your status.
But this could possibly bring a brighter future for humanity, where we tame the viral nature of capitalism and the rich getting richer at a far faster rate than the rest of society.
no one judge can take it away from me, and i value that as others do. i fail to see how a decentralized store of value and exchange is nil value
[0] https://www.coindesk.com/why-bitcoins-remittance-disruption-... [1] https://medium.com/cryptonight/bitcoin-doesn-t-make-remittan... [2] https://www.saveonsend.com/blog/bitcoin-blockchain-money-tra...
i don't really see its electricity use as a big issue since market forces will inevitably reduce cost or increase supply.
part of the reason though people will put a lot into an experiment like this is that the network effect is important to its success. needs to be very liquid.
for me personally having a fungible electronic currency that cannot be taken away by law or inflation is itself valuable
What do you mean? Bitcoin already solves problems; e.g., https://www.forbes.com/sites/ktorpey/2017/12/31/bitpesa-ceo-...
A discussion on this cannot take place without a mention of the lightning network.
The vast majority of transaction have a low stake. Most amounts are fairly low, people respect the fact they have parted from some value. Using such a highly price consensus seems like overkill. What you want is to concentrate on the litigious ones, on the higher amounts; while still securing the rest.
This is exactly what is happening in the lightning network smart contract. Bitcoins are taken apart in a channel, and people can do their business as usual. The protocol concentrates the stakes in a single on-chain transaction; with a tremendous pressure in not settling on the last channel state. It sorts of changes the transaction representation space to its dual. From a white-list of all approved transactions -and nothing else- to a black-list of misbehaving business relations -all other consented transactions being allowed-.
Bitcoin's 7 transactions per second is demultiplied by some order of magnitude here. That represents about 3 channels per second. 3 licenses to indefinitely transact up to a certain balance.
The lightning network is about rationalizing the use of this high-priced consensus.
----
As for the electricity consumption, I'm of the opinion that the halvenings will take care of that; In 10 years, the reward will have diminished by about an order of magnitude.
Bitcoin gained too much popularity too quickly if you ask me. I think a super-exponential difficulty adjustment would have been better suited (make it require an exp(x\^2) hash rate instead of exp(x)).
The growth of Bitcoin is capped exponentially, as it is based on human adoption. Only something super-exponential can regulate this properly.
But IMHO, this is only a slight inconvenience. We'll waste electricity but only in the next few years.
How are you measuring "transaction throughput"?
Maybe social dynamics work out to avoid this, but humans have a long history of trending towards various forms of plutocracies.
One of the reasons Ethereum forked was to avoid a single party (the DAO thief) controlling far too large a portion of the total ether which has big consequences in a future with PoS.
Are we going to continue fork networks every time one party gets too rich? We could. Unless of course that party is pretending to be many individuals and we can't tell. Hard to stop this without deanonymization and a loss of censorship resistance. To be fair PoW suffers similarly from mining power all clustering in cheap electricity areas and thus players like Bitmain have a scary amount of centralized control.
I am skeptical of PoW long term without much better globally distributed use of renewalable energy. PoS by dropping this anchor to the physical world seems even crazier.
I am however somewhat optimistic about Bram Cohen's "Proof of space and time" in his "Chia" project which attempts to move the PoW problem from proof of expended electricity to proof of burned disk space, which feels inherently greener and easier to geographically distribute.
Or neither of these PoW alternatives work and the lessons learned from them empower as of yet unknown innovations. Exciting times to be sure.
It's not free to manufacture hard drives, in terms of materials, energy, CO2 etc. Proof of space and time merely shifts the pressure away from compute power to storage.
It has the potential to do for the storage market what PoW as done for the graphics card market - pump up prices and move useful tech out of the reach of most consumers.
Proof of storage is just another ecological disaster in the making and I sincerely hope it never takes off.
Having said that, "Wall Street" should be wary of enabling the use of a system that is using more electricity than the country of Switzerland [0] without any productive output. The energy consumption is rapidly rising too. Furthermore, Bitcoin's transaction throughput is dismally low compared to existing payment systems (Visa, PayPal, etc.) and "smart contract" systems like Ethereum seems to still not show any productive output aside from scams and severe bugs in wallets and contracts that lose millions of dollars.
I'm all for "wait and see", but have we ever done that with a technology that is literally using more electricity than a medium sized nation? When are we going to ask crypto-fans to be accountable for the burned energy? We're asking Google and Facebook to treat our data fairly, can we ask proof-of-work miners to treat our planet fairly? At least wait until the blockchain can solve a real world problem before burning through the mines?
[0] https://digiconomist.net/bitcoin-energy-consumption